685 20 17MB
English Pages 344 [317] Year 2021
Thermodynamic Analysis and Optimization of Geothermal Power Plants
Thermodynamic Analysis and Optimization of Geothermal Power Plants
Edited by
Can Ozgur Colpan Mehmet Akif Ezan Onder Kizilkan
Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States Copyright © 2021 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-12-821037-6 For information on all Elsevier publications visit our website at https://www.elsevier.com/books-and-journals
Publisher: Brian Romer Acquisitions Editor: Graham Nisbet Editorial Project Manager: Leticia M. Lima Production Project Manager: Poulouse Joseph Cover Designer: Miles Hitchen Typeset by SPi Global, India
Contributors Numbers in parenthesis indicate the pages on which the authors’ contributions begin.
Sertaҫ Akar (17) National Renewable Energy Laboratory (NREL), Golden, CO, United States Olusola Charles Akinsipe (3) School of Engineering & Built Environment, Griffith University, Brisbane, QLD, Australia Panagiotis Alexopoulos (131) Laboratory of Soft Energy Applications and Environmental Protection, Mechanical Engineering Department, University of West Attica, Athens, Greece Sharjeel Ashraf Ansari (249) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan Chad Augustine (17) National Renewable Energy Laboratory (NREL), Golden, CO, United States
C. Ozgur Colpan (153) The Graduate School of Natural and Applied Sciences; Faculty of Engineering, Department of Mechanical Engineering, Dokuz Eylul University, Buca, Izmir, Turkey Ibrahim Dincer (207, 225) Clean Energy Research Laboratory, Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON, Canada Anil Erdogan (153) The Graduate School of Natural and Applied Sciences, Dokuz Eylul University, Buca, Izmir, Turkey Mehmet Akif Ezan (153) The Graduate School of Natural and Applied Sciences; Faculty of Engineering, Department of Mechanical Engineering, Dokuz Eylul University, Buca, Izmir, Turkey
Muhammad Aziz (97) Institute of Industrial Science, The University of Tokyo, Tokyo, Japan
Muhammad Farooq (315) Department of Mechanical Engineering, University of Engineering and Technology, KSK Campus, Lahore, Pakistan
Young-Jin Baik (315) Thermal Energy Systems Laboratory, Korea Institute of Energy Research, Daejeon, Republic of Korea
Milad Feili (167) Department of Mechanical Engineering, Faculty of Engineering, University of Mohaghegh Ardabili, Ardabil, Iran
Yusuf Bas¸ o gul (113) Department of Mechanical Engineering, Engineering Faculty, Adıyaman University, Adıyaman, Turkey
Hikari Fujii (83) Graduate School of Engineering and Resource Science, Akita University, Akita, Japan
Riccardo Basosi (53) Center for Colloid and Surface Science, University of Firenze, Sesto Fiorentino; R2ES Lab, Department of Biotechnology, Chemistry and Pharmacy, University of Siena, Siena; National Research Council—Institute for the Chemistry of OrganoMetallic Compounds, Sesto Fiorentino, Italy
Hadi Ghaebi (167) Department of Mechanical Engineering, Faculty of Engineering, University of Mohaghegh Ardabili, Ardabil, Iran Onur Vahip G€ uler (113) Department of Energy Systems Engineering, Technology Faculty, Mugla Sıtkı Koc¸man University, Mugla, Turkey
Arianna Bonzanini (43) Turboden S.p.A., Brescia, Italy
Muhammad Imran (315) School of Engineering and Applied Science, Aston University, Birmingham, West Midlands, United Kingdom
G€ urcan C ¸ etin (263) Department of Information Systems Engineering, Technology Faculty, Mu gla Sıtkı Koc¸man University, Mu gla, Turkey
Mohammad Ashar Jamal (185) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
George Charis (131) Laboratory of Soft Energy Applications and Environmental Protection, Mechanical Engineering Department, University of West Attica, Athens, Greece
Rao Hamza Jamil (185) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Joseph Bonafin (43) Turboden S.p.A., Brescia, Italy
xi
xii Contributors
Firman Bagja Juangsa (97) Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Bandung, Indonesia
Farayi Musharavati (279) Department of Mechanical and Industrial Engineering, Qatar University, Doha, Qatar
Khurram Kamal (249) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Greg F. Naterer (225) Clean Energy Research Laboratory, Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON; Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL, Canada € Osman Ozkaraca (263) Department of Information Systems Engineering, Technology Faculty, Mu gla Sıtkı Koc¸man University, Mugla, Turkey
Prasad Kaparaju (3) Institute for Applied Sustainability Research (iiasur), Quito, Ecuador; School of Engineering & Built Environment, Griffith University, Brisbane, QLD, Australia Spyridon Karytsas (65) Geothermal Energy Department, Division of Renewable Energy Sources, Centre for Renewable Energy Sources and Saving (CRES), Pikermi; Department of Home Economics and Ecology, School of Environment, Geography and Applied Economics, Harokopio University (HUA), Kallithea, Greece Kosmas A. Kavadias (131) Laboratory of Soft Energy Applications and Environmental Protection, Mechanical Engineering Department, University of West Attica, Athens, Greece Ali Kec¸ ebas¸ (113, 263) Department of Energy Systems Engineering, Technology Faculty, Mu gla Sıtkı Koc¸man University, Mu gla, Turkey Shoaib Khanmohammadi (279) Department of Mechanical Engineering, Kermanshah University of Technology, Kermanshah, Iran Onder Kizilkan (153, 279) Department of Mechanical Engineering, Faculty of Technology, Isparta University of Applied Sciences, Isparta, Turkey Parthiv Kurup (17) National Renewable Energy Laboratory (NREL), Golden, CO, United States Saeid Mohammadzadeh Bina (83) Graduate School of Engineering and Resource Science, Akita University, Akita, Japan Diego Moya (3) Department of Chemical Engineering & Grantham Institute—Climate Change and the Environment, Science and Solutions for a Changing Planet DTP, Imperial College London, London, United Kingdom; Institute for Applied Sustainability Research (iiasur), Quito; Carrera de Ingenierı´a Meca´nica, Facultad de Ingenierı´a Civil y Meca´nica, Universidad Tecnica de Ambato, Ambato; Coordinacio´n de Investigacio´n e Innovacio´n, ABREC, Quito, Ecuador Hafiz Ali Muhammad (315) Thermal Energy Systems Laboratory, Korea Institute of Energy Research, Daejeon, Republic of Korea
Murat Ozturk (207) Faculty of Technology, Department of Mechatronics Engineering, Isparta University of Applied Science, Isparta, Turkey Mohammad Mustafa Pardesi (185) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan Maria Laura Parisi (53) Center for Colloid and Surface Science, University of Firenze, Sesto Fiorentino; R2ES Lab, Department of Biotechnology, Chemistry and Pharmacy, University of Siena, Siena; National Research Council—Institute for the Chemistry of OrganoMetallic Compounds, Sesto Fiorentino, Italy Olympia Polyzou (65) Geothermal Energy Department, Division of Renewable Energy Sources, Centre for Renewable Energy Sources and Saving (CRES), Pikermi, Greece Tahir Abdul Hussain Ratlamwala (185, 249) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan Zabdur Rehman (315) Department of Mechanical Engineering, Air University Islamabad, Aerospace and Aviation Campus, Kamra, Pakistan Ron R. Roberts (225) Clean Energy Research Laboratory, Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON, Canada Hadi Rostamzadeh (167) Energy and Environment Research Center, Niroo Research Institute (NRI), Tehran, Iran Lalit Chandra Saikia (293) Department of Electrical Engineering, National Institute of Technology, Silchar, Assam, India Muhammad Nouman Saleem (249) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Contributors
xiii
Muhammad Afzal Sheikh (249) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Washima Tasnin (293) School of Electrical Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu, India
Farooq Sher (315) School of Mechanical, Aerospace and Automotive Engineering, Coventry University, Coventry, United Kingdom
Lorenzo Tosti (53) Center for Colloid and Surface Science, University of Firenze, Sesto Fiorentino; R2ES Lab, Department of Biotechnology, Chemistry and Pharmacy, University of Siena, Siena, Italy
Uzair Aziz Suria (185) Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Shunsuke Tsuya (83) Graduate School of Engineering and Resource Science, Akita University, Akita, Japan
Chapter 1
Various cycle configurations for geothermal power plants Diego Moyaa,b,c,d, Olusola Charles Akinsipee, and Prasad Kaparajub,e a
Department of Chemical Engineering & Grantham Institute—Climate Change and the Environment, Science and Solutions for a Changing Planet DTP,
Imperial College London, London, United Kingdom, b Institute for Applied Sustainability Research (iiasur), Quito, Ecuador, c Carrera de Ingenierı´a Meca´nica, Facultad de Ingenierı´a Civil y Meca´nica, Universidad T ecnica de Ambato, Ambato, Ecuador, d Coordinacio´n de Investigacio´n e Innovacio´n, ABREC, Quito, Ecuador, e School of Engineering & Built Environment, Griffith University, Brisbane, QLD, Australia
1.1
Introduction
Globally, the transition from the present petroleumdependent energy technology to green energy is fundamentally contingent on making a decision on result-driven renewable systems [1]. Mitigating climate issues as well as promoting sustainable development are unattainable without innovative systems and technology transfer. Features such as low greenhouse gas emissions, minimized environmental disruption, and the viability of technology extraction have demonstrated geothermal energy as a sustainable energy resource [2] that could be harnessed and exploited regardless of climatic factors [3]. The Earth’s crust houses the renewable geothermal energy source [4], usually associated with tectonic activity and volcanic activities [5]. Typically, geothermal energy is located as a heat source in hot rocks [6] as well as hydrothermal reservoirs in the Earth’s crust. Geothermal energy is considered a sustainable clean renewable energy resource. The global installed geothermal capacity is estimated to be 12.729 MW and that is projected to grow by 68.46% in 2020. Among the geothermal power plant configurations, single flash (41%) is the predominant configuration followed by dry steam (23%), double flash (19%), and binary (14%). The triple flash (2%) and back pressure (1%) plant configurations are less popular [7]. Factors such as the heat reinjection mechanism, geothermal applications, and geologic time scale are backstopping the accelerated geothermal reservoir heat extraction in comparison with the replacement of heat in the reservoirs. However, methods for heat reinjection have been developed to ensure geothermal energy as a renewable energy resource. Geothermal fluids are generally used to capture heat energy from the Earth’s crust and transport it to the surface through production wells drilled into the geothermal
reservoir [8]. Depending upon the location, temperature, and depth of the geothermal reservoir, geothermal fluids consisting of mineral-coated hot water known as brine and steam (vapor-dominated fluids) are generally used [9]. The heat energy that is transported to the ground surface is further processed for electricity generation and/or direct uses [8]. Depending on the geothermal site features, thermal energy can be harvested from depths of 300 to 3000 m and beyond. At greater depths, thermofluids are naturally occurring as the porous hot rock of the hydrothermal and geopressured geothermal reservoirs [10]. Harvesting the hot fluids is a contingent feature of these reservoirs, and various control measures are factored to maximize the utilization of this heat energy [11]. Finally, the electrical power generation is dependent upon the application of various geothermal heat [12]. In this chapter, current geothermal power plant systems and their significance in applying cuttingedge geothermal configurations as well as undertaking research on hybrid configurations are presented.
1.2
Geothermal power plant system
The abundance of hydrothermal resources has influenced the development of geothermal power plant arrangements and systems [13]. Geothermal power plants can be classified as single-flash steam power plants, double-flash steam power plants, dry steam power plants, binary (Organic Rankine-Kalina Cycle) power plants, and advanced geothermal energy systems. The advanced geothermal energy systems are further classified as hybrid single-double-flash systems, hybrid flash-binary systems, hybrid fossilgeothermal technologies, and hybrid other renewable heat source-geothermal systems [12]. Geothermal power plants are best grouped into steam and binary cycles for cycles for higher well enthalpies and lower enthalpies, respectively
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00005-6 Copyright © 2021 Elsevier Inc. All rights reserved.
3
4 PART
I Basics of geothermal power plants
[14]. This chapter will assess the thermodynamic aspects of the five geothermal power plant configurations.
1.2.1
Single-flash steam power plants
The single-flash steam power plant is the simplest geothermal power transformation arrangement. This configuration facilitates liquid-vapor production from the geothermal production wells. Based on their large density disparity, these are separated into two dissimilar phases: steam and liquid with the support of a cylindrical cyclone pressure vessel [14]. The term “single” depicts a single flashing mechanism of the geofluid obtained by depressuring the geothermal fluid pressure [15]. This can be achieved in a production well, a reservoir, or a cyclone inlet to support the transition of pressurized liquid to liquid-steam mixture production. A 30 MW single-flash geothermal power plant demands between 5 and 6 production wells and 2–3 reinjection wells appropriated along with the geofluid resource [15]. Further, pipes are deployed for mixture accumulation and transportation from the various geo-
thermal production wells. The identified constraint is the drop in vapor pressure due to the pipe frictional force associated with harvesting mechanisms [16]. Empirical correlations are investigated, considering their complication and reliability, by deploying factors such as the vapor mass flow rate, the density, the pipe diameter, the length, and the components of the pipe to forecast pressure loss. The variables are significant compared to the investment cost of the power plant and the energy conversion technology [16]. A turbine generator produces electrical energy from vapor (around 99.95% dry) after separation of the vapor and liquid [15]. The choice of a single-flash process is applicable when the geothermal fluid temperature exceeds 260°C with the attainment of a capacity factor between 95% and 100% [10]. Fig. 1.1 shows the single-flash process of energy conversion technology. Station 1 is where the single-flash steam power process commences while the geothermal fluid accesses the production well through the source inlet temperature. Between stations 1 and 2 (producing pipes), a pressure drop occurs, and this facilitates the boiling of the fluid (a vapor-liquid mixture) before it is transported to
CV
SV
MR T
5
EG
G
BCV
c2 3
6
SE/C C
ST
CT
CS WV
S
WV
ST
CP
WH
c1 CWP 7
2 IW PW
PW S WV CS IW G
4
1
Production well Silencer Well valves Cyclone separator Injection well Generator
FIG. 1.1 Single-flash geothermal power cycle [17].
BCV MR ST CV EG T
Ball check valve Moisture remover Steam tramp Control valve Electric grid Turbine
SV SE C CP CWP WH
Stop valve Steam jet ejectors Condenser Condensate pump Condensed water pump Wellhead
Various cycle configurations for geothermal power plants Chapter 1
station 2. At station 5, it collects steam from the mixture fluid after separation while the brine (mineral-laden hot water) is collected at station 3 before reinjecting at station 4. At station 5, the induced motion of the turbine generator supports the electrical energy at the entrance of the steam, followed by the production of steam expansion along with the turbine to station 6 at condenser pressure. An air cooler condenser may be used at station c1 to allow the cooling of air and exit at station c2 [15]. The significance of certain plant equipment in the operation of a single-flash steam geothermal power plant is greatly acknowledged. In this configuration, large amounts of freshwater are not needed for cooling [18]. Nevertheless, cooling the tower, particularly the dry regions without fresh water, is achieved by deploying cooling water obtained from condensed steam [16]. Fig. 1.2 is a thermodynamic state (T-s) diagram analyzing the single-flash steam conversion process. It should be noted that mass conversion and the energy conversion
FIG. 1.2 Temperature-entropy (T-s) diagram of a single-flash cycle [4].
principle, the two fundamental thermodynamic principles, are considered in the investigation of the process with the aid of the diagram. In a single-flash power plant, flashing occurs when the geothermal fluid under pressure initiates the process close to the saturation curve at state 1 (Fig. 1.2). The change in the potential or kinetic energy is not considered, and the enthalpy (h) is designed as constant, h1 ¼ h2, as can be seen in Eq. (1.1) in Table 1.1. After the flashing process, the separation process occurs at state 2, and is simulated at constant pressure. Further, the vapor and liquid mixture is shown, ascertaining the mixture quality in this state. The dryness fraction (x2) drives this quality, as shown by Eq. (1.2) in Table 1.1. The quantity of vapor entering the turbine is represented by the steam mass fraction. Eq. (1.3) in Table 1.1 shows the work per unit mass (w1) generated by the turbine expansion process between states 4 and 5. The potential and kinetic energy are not generally considered, and heat losses are neglected when the thermal fluid enters and leaves the turbine. Eq. (1.4) in Table 1.1 is the isentropic turbine efficiency, which is denoted by t. This is considered the ratio of the actual work to that of the isentropic work, which is the ideal process from states 4 to 5. Eq. (1.5) indicates the turbine gross mechanical power (W_ t ). The electrical power output of the generator (Eq. 1.6) is given as the turbine’s mechanical power times the efficiency of the generator (g). Lastly, at states 5 and 6, the condensation and cooling processes occur, and Eq. (1.7) gives the cooling water flow rate [4]. For the purpose of analyzing the whole plant efficiency, the second law of thermodynamics is investigated [16]. This allows its examination in contrast to the actual power output toward the finish of the single-flash process and to the utmost theoretical power that is generated by the geothermal fluid [15]. Exergy is the energy feasible to be used and the capacity to produce work from energy [16]. Eq. (1.8) presented in Table 1.2 defines the specific exergy (ex) of the
TABLE 1.1 Equations used for thermodynamic state analysis [15]. State
Main characteristics
Equation
Equation number
Flashing process
Constant enthalpy
h1 ¼ h2
(1.1)
Separation process
Constant pressure Liquid plus vapor mixture
3 x2 ¼ hh24 h h3
(1.2)
Turbine expansion process
Constant entropy
w1 ¼ h4 h5
(1.3)
h5 t ¼ hh44h 5s
(1.4)
W_ t ¼ m_ s wt
(1.5)
W_ e ¼ g W_ t
(1.6)
5 h6 m_ cw ¼ x2 m_ total hcDT
(1.7)
Condensing process
5
6 PART
I Basics of geothermal power plants
TABLE 1.2 Exergy and power plant efficiency [19]. Thermodynamic dimension
Equation number
Equation
Specific exergy
ex ¼ h(T, P) h(TO, PO) TO[s(T, P) s(TO, PO)]
(1.8)
Exergetic power
_ ¼ m_ total ex Ex
(1.9)
Entire power plant efficiency
u ¼ WE_net ¼ WE_e
_
_
(1.10)
geothermal fluid for a given pressure (P), temperature (T), ambient pressure (PO). and ambient temperature (TO). Eq. (1.9) shows the exergetic power, also known as the maximum theoretical thermodynamic power, which is the total geothermal mass flow rate times the exergy. Finally, the exergy efficiency of the entire power plant is given by Eq. (1.10) [19]. There are several environmental impacts of single-flash geothermal power plants [15]. Locations such as the cooling tower, the ejector vents, the pipeline drains, the steam tramps, the silencers, the mufflers, and the wellhead, which offers a structural interface between the wells and the production system, are the prime areas of pollution [18]. The blending of noncondensable gases such as methane (CH4), hydrogen sulfide (H2S), and carbon dioxide (CO2) from the steam of geothermal reservoirs is considered a main environmental concern. These gases are, however, subjected to treatment and isolation before being discharged into the atmosphere [20]. Further, in spite of its CO2 emissions, the GHG emissions from a single-flash geothermal power plant (0.06 kg CO2/kWh) are significantly lower than the traditional coal-fired (1.13 kg CO2/kWh) or natural-gasfired power plants (0.59 kg CO2/kWh) [13]. With respect to footprint, the land requirements of a coal-fired power plant (40,000 m2/MW) and a solar photovoltaic power plant (66,000 m2/MW) are much higher than the 1200 m2/MW required for a single-flash plant [15]. In general, the loss of characteristic beauty, ozone-depleting substances, land and water utilization, visual and noise pollution, and water contamination are some of the other environmental concerns associated with geothermal power plants [10]. Strategies to mitigate the environmental impacts of single-flash geothermal power plant as proposed in [20] include mufflers and silencers to abate noise pollution, air-cooled condensers, reinjection for surface water, and preventing expansion of geothermal projects into national parks. By and large, the emissions from geothermal power plants are inconsequential compared to fossil-fuel conventional power plants.
1.2.2
Double-flash steam power plants
The development of the double-flash steam geothermal power plant was to support power generation by the use of a mixture of vapor and liquid water generated in the geothermal production wells [21]. A double-flash power plant is considered more advantageous than a single-flash power plant, as the former can generate 25% more output power than the latter under the same geothermal fluid conditions [13]. However, double-flash steam power plant technology is more complex and its operation and maintenance are more expensive than single-flash power plants. Nevertheless, the efficient use of the geothermal resource is a pointer that a secondary flash process is valuable. The use of a second pressure drop in a secondary flash process (second separator), after the first pressure drop, supports the production of extra vapor from the separated liquid exiting the first separator. Further, the coupled turbine generator is able to produce additional power due to the supply of lower-pressure steam [22] or to a turbine depending on the configuration [23]. The double-flash power plant’s complete exergy performance is optimally boosted due to the separator of the geothermal steam-water, which is the principal technological development of this power plant technology. Upon comparison to a single-flash system (Fig. 1.1), a double-flash configuration (Fig. 1.3) uses a dual-admission turbine and a low-pressure separator. For the purpose of smooth combination with the expanded high-pressure steam, the low-pressure steam is supplied to the turbine at the right stage [15]. Fig. 1.3 presents the energy conversion process in a double-flash steam geothermal power plant. The double-flash steam power process commences at station 2; the source of the inlet temperature is the corridor of the geofluid. The first flashed-steam process occurs between stations 1 and 2, when the pressure drops and the fluids begin to boil (mixture steam-liquid) before reaching the separator at station 2. The fluid mixture is then separated into the brine and high-pressure steam. The mineral-laden brine hot water (station 3) is then downwardly controlled to low-pressure (station 8) and high-pressure steam (station 5) with the support of a separator. This prompts the second flashed-steam process. The second pressure drop at station 9 will lead to the production of a steam-brine mixture and the brine is collected by the low-pressure separator. The second steam is injected into the system at station 9 and the turbine collects the second fresh low-pressure steam. At station 5, the first high-pressure steam gains access to the turbine after the initial steam injection. The induced motion of the dualinjection turbine, connected to a generator, generates electrical energy. The condenser pressure at station 6, connected to the turbine, is the location where the steam expansion happens [15]. With respect to the process design, the first stage admission or injection of the turbine should have
Various cycle configurations for geothermal power plants Chapter 1
PW S WV BCV IW CWP HPCS
Production well Silencer Well valves Ball check valve Injection well Condensed water pump High-pressure cyclone separator
SV SE C CP HPT LPT LPFS
Stop valve Steam jet ejectors Condenser Condensate pump High-pressure turbine Low-pressure turbine Low-pressure flash separator
MR ST CV TV G EG WH
7
Moisture remover Steam tramp Control valve Throttle valve Generator Electric grid Wellhead
FIG. 1.3 Double-flash geothermal power plant with a dual admission turbine [15, 17].
the same pressure difference between the high and low separators [24]. The high-pressure stage mass flow is expected to be lower than the low-pressure stage mass flow. The residual hot fluids at station 6 are condensed by using an air-cooled condenser. Cool air is supplied at station c1 and exited at station c2. Finally, the residual brine from the second flashed process at station 10 and the condensed fluid from station 7 is reinjected into the system at station 4 (Fig. 1.3). The temperature-entropy (T-s) of a double-flash power plant is presented in Fig. 1.4. The two flash processes that exist in states 1–2 and 3–6 are significant, and they are studied separately as a single process [15]. To ascertain the quantity of steam generated in the separators at each flashed process (x2 in states 1–2 and x6 in states 3–6, the separation process), Eqs. (1.11)–(1.14) in Table 1.3 are applied. Further, the evaluation of the steam at state 2 and the brine at state 6, obtained from different separators at the high- and low-pressure stages, was achieved by using four equations, as shown in Eqs. (1.15)–(1.18). Again, Eqs. (1.15), (1.17) give the mass flow rate of steam generated at high pressure (m_ hps , at state 5) as well as at low pressure (m_ lps , at state 8), respectively. Also, Eqs. (1.16),
(1.18) are used to calculate the mass flow rate of brine produced at high pressure (m_ hpb , at state 3) and low pressure (m_ lpb , at state 7). The low-pressure turbine stage (at state 9) accommodates the high-pressure and low-pressure steams together. With the aid of Eqs. (1.15)–(1.18), four values are appraised: the disposed waste liquid, the
FIG. 1.4 Temperature-entropy process diagram for a double-flash power plant with a dual admission or injection turbine [15].
8 PART
I Basics of geothermal power plants
TABLE 1.3 Thermodynamic equations for double-flash geothermal power plants [14, 15]. State
Main characteristics
Equation
Equation number
Flash process 1
Constant enthalpy
h1 ¼ h2
(1.11)
Separation process 1
Constant pressure Mixture of liquid plus vapor
3 x2 ¼ hh24 h h3
(1.12)
Flash process 2
Constant enthalpy
h3 ¼ h6
(1.13)
Separation process 2
Constant pressure Mixture of liquid plus vapor
7 x6 ¼ hh38 h h7
(1.14)
Mass flow rate of steam generated
High pressure
m_ hps ¼ x2 m_ total ¼ m_ 4 ¼ m_ 5
(1.15)
Mass flow rate of brine produced
High pressure
m_ hpb ¼ ð1 x2 Þm_ total ¼ m_ 3 ¼ m_ 6
(1.16)
Mass flow rate of steam generated
Low pressure
m_ lps ¼ ð1 x2 Þx6 m_ total ¼ m_ 8
(1.17)
Mass flow rate of brine produced
Low pressure
m_ lpb ¼ ð1 x2 Þð1 x6 Þm_ total ¼ m_ 7
(1.18)
Turbine expansion process
High-pressure stage (Eqs. 1.22–1.23 are the Baumann rule)
whpt ¼ h4 h5
(1.19)
h5 hpt ¼ hh44h 5s
(1.20)
W_ hpt ¼ m_ hps whpt ¼ x2 m_ total whpt h7 h4 A 1 h8 h7 h5 ¼ A 1+ h8 h7
(1.21)
A ¼ 0.425(h4 h5s)
(1.23)
m_ 5 h5 + m_ 8 h8 ¼ ðm_ 5 + m_ 8 Þ h9
(1.24)
2 Þx6 h8 h9 ¼ x2 hx52 ++ ðð1x 1x2 Þx6
(1.25)
wlpt ¼ h9 h10
(1.26)
W_ lpt ¼ m_ 9 ðh9 h10 Þ h11 h9 A x9 h12 h11 h10 ¼ A 1+ h12 h11
(1.27)
Turbine expansion process
Low-pressure stage
condenser heat dissipated, the cooling water heat losses, and the turbine power production. This system is applied for the illustration of the two turbine expansion processes. Eq. (1.19) is the representation of the first turbine expansion process producing work (whpt) that occurs between states 4 and 5. Eq. (1.26) describes the second turbine expansion process producing work (wlpt).
(1.22)
(1.28)
A ¼ 0.425(h9 h10s)
(1.29)
h10 lpt ¼ hh99h 10s
(1.30)
W_ total ¼ W_ hpt + W_ lpt
(1.31)
W_ e , gross ¼ g W_ total
(1.32)
Thus, for a typical double-flash power plant with the high-pressure (hpt) and low-pressure (lpt) processes, two isentropic turbine efficiencies are derived by using Eq. (1.20), and a low-pressure process (lpt) by using Eq. (1.30). The first turbine stage power produced at high pressure (W_ hpt ) and the second stage power produced at low pressure (W_ lpt ) are defined by using Eq. (1.21) and
Various cycle configurations for geothermal power plants Chapter 1
Eq. (1.27), respectively. Eq. (1.31) provides the aggregate turbine power produced (W_ total ), which is an addition of individual turbine stage power generated. Lastly, as explained in Eq. (1.32), the efficiency of the generator (g) impacts the electrical power (W_ e, gross ). The efficiency of the overall power plant, the incoming geothermal fluid exergy, the environmental impacts, and the deployed equipment are similar to that of the single-flash process [24, 25].
1.2.3
Dry-steam power plants
Studies have shown that several locations around the globe are endowed with geothermal dry-steam, particularly in places such as the geysers in the United States and Larderello, Italy. Both places have the two largest dry-steam reservoirs. However, places such as Cove Fort, Utah, United States; Wairakei, New Zealand; Matsukawa, Japan; and Kamojang, Indonesia are characterized by limited dry steam [16, 20]. In the event that a geothermal reservoir dries, Zarrouk and Moon [16] discussed the possibility of
FIG. 1.5 Dry-steam geothermal power plant [9, 17].
9
converting the flashed generation systems into a dry-steam system. Considering high enthalpy system configurations, dry-steam geothermal power plants are the most efficient because the hydrothermal reservoirs support these configurations with vapor-dominant geothermal fluid at high temperatures [25, 26]. The coupled turbine generator is able to generate electrical energy using the steam supplied directly from the production well [10]. In dry-steam power generation, between 50% and 70% of the geothermal fluid available work (exergy) is converted into electrical power [25]. Upon comparison with the steamflashed process, the dry steam has a simpler concept and requires centrifugal cyclones to separate particulate matters such as rock chippings and dust [25]. Similarly, the condensate is eliminated by using drain pots, and the last moisture is eliminated to achieve high-grade steam in the turbine. A Venturi meter is also needed for the accurate calibration of the turbine steam flow rate [13]. The mechanism of conversion of energy in the dry-steam power plant process is described in Fig. 1.5. Both singleflash and dry-steam (geothermal fluid) processes share
10
PART
I Basics of geothermal power plants
similar features of the whole power generation cycle, from the production wells to the production turbine [15]. For both smaller or larger units of single flow or double flow, a single pressure is performed by the blading turbine of impulsereaction. Based on the graphs, Fig. 1.1 (single flash) shares the same similarity with Fig. 1.5 with a particulate remover rather than a cyclone separator. Fig. 1.6 is the T-s thermodynamic state of the dry-steam system. At state 1, saturated steam or lightly superheated steam is generated in the production wells. The expansion of the turbine occurs between states 1 and 2 while the cooling process is attained between states 2 and 3, where an emission of heat via the condenser occurs. The singleflash geothermal power plant system is equivalent to the thermodynamic study [18]. Table 1.4 shows the equations in analyzing the dry-steam power plants. In terms of impacts on the environment, the single-flash geothermal power plant process has a lower environmental impact than flashed power plants, as the system does not use mineral-laden brine [15].
FIG. 1.6 Temperature-entropy (T-s) process diagram for a dry-steam power plant with saturated steam at the turbine inlet [15, 18].
TABLE 1.4 Thermodynamic equations in the dry-steam process for turbine expansion process [15]. Equation
Equation number
wt ¼ h1 h2
(1.33)
h2 t ¼ hh11h 2s
(1.34)
W_ t ¼ m_ s wt ¼ m_ s ðh1 h2 Þ
(1.35)
W_ e ¼ g W_ t
(1.36)
1.2.4 Binary-organic Rankine cycle and Kalina cycle power plants The binary geothermal power plant (B-GPP) generates electrical energy from a secondary separated process. Preheating of the working fluid is involved and heat is lost upon contacting the geothermal fluid [10]. Geothermal resources with a temperature range of 20–150°C [25] or 85–170°C [27] are well suited for binary configurations [25]. A higher temperature range provide thermal stability of the working fluid while the lower temperatures are more feasible in terms of technoeconomic and financial factors. Further, the impacts of corrosion and scaling are not apparent at high temperature, as there is no contact between the power generation equipment and the geofluid. Under a conventional Rankine cycle, there is the functionality of the secondary fluid (working fluid) in the binary system [28], and the binary cycle is identified as the Organic Rankine Cycle (ORC) due to the organic nature of the working fluid. Binary power plants are versatile and the functionality of the power plant is decided by the secondary cycle. Different types of configurations of binary power plants are in operation, including B-GPP using an ORC with an internal heat exchanger (IHE), B-GPP with a regenerative ORC, and B-GPP with a regenerative ORC using IHE [29]. In 1982, Kalina patented a variation in B-GPP [30]. The working fluid used in the Kalina cycle consists of water and ammonia and can be used in different compositions to suit various configurations [31]. A thermal efficiency of 30%–40% is achievable and is considered more efficient than that of an ordinary B-GPP [28]. A closed loop of a thermodynamic Rankine Cycle used for the energy conversion system of a basic binary geothermal power plant is shown in Fig. 1.7. Harvesting the geothermal fluid via the production wells (PW) and then transporting it through various primary cycle components necessitates that pumping systems are deployed. The scouring and erosion of pipes and tubes can be prevented by extracting sand from the geofluid by employing sand removers (SR). Finally, an evaporator (E) and a preheater (pH) are used for continuous fluid flow while the geothermal fluid is reinjected into the reservoir near the injection well (IW) by using an injection pump (IP). Regarding the working cycle, two heating-boiling procedures are included in the working fluid. The PH is the location of the boiling point of the working fluid. Upon contact of PH with E, the working fluid becomes a saturated vapor. This results in the expansion and condensation of the working fluid in the turbine. The working fluid is then returned to the evaporator, thereby concluding the loop process and beginning the process again [31]. To prevent steam eruption and calcite scaling within the pipes, monitoring the geothermal fluid above the flash pressure point is necessary [13]. Thus, the low-temperature geothermal
Various cycle configurations for geothermal power plants Chapter 1
11
FIG. 1.7 Basic binary geothermal power plant [17, 18].
resources are enabled by this binary-type energy conversion setup and by relying upon specialized highlights to attain amazingly high plant performances [32]. Fig. 1.8 presents the pressure-enthalpy of a binary geothermal power plant. At state 1, the working fluid at the saturated vapor point accesses the turbine and facilitated the expansion and production of work, and this marks the beginning of the thermodynamic process. Based on the work produced, electrical energy is generated by the generator. At state 2, after the expansion process of the turbine, the temperature and pressure of the saturated vapor reduce. At state 3, a temperature reduction occurs as the working steam-fluid enters the condenser, and this eventually culminates into fluid condensation. The application of cooled water from the air-cooled tower initiates the cooling process of the working fluid between states 3 and 4. The transformation of the working fluid state into a saturated liquid is achieved by this cooling phase. At states 5 and 6, the working saturated liquid-fluid is pumped back to the preheater and evaporator, respectively. At state 1, the
emergence of the working fluid as saturated vapor initiates the process repeatedly [14, 15]. Regarding the condenser, the turbine, and the feed pump, the flash and dry-steam plants share similar
FIG. 1.8 Pressure-enthalpy diagram of a binary geothermal power plant [15].
12
PART
I Basics of geothermal power plants
TABLE 1.5 Thermodynamic equations for binary geothermal power plants [15]. State
Equation
Equation number
Turbine expansion process
w1 ¼ h1 h2
(1.37)
h2 t ¼ hh11h 2s
(1.38)
W_ t ¼ m_ wf wt ¼ m_ wf t ðh1 h2s Þ
(1.39)
W_ e ¼ g W_ t
(1.40)
Condensing process
Q_ c ¼ m_ wf ðh2 h3 Þ
(1.41)
Feed pump
W_ p ¼ m_ wf ðh4 h3 Þ
(1.42)
Heat exchange process at E and PH
m_ b ðha hc Þ ¼ m_ wf ðh1 h4 Þ
(1.43)
PH: m_ b c b ðTa Tc Þ ¼ m_ wf ðh5 h4 Þ
(1.44)
E: m_ b c b ðTa Tc Þ ¼ m_ wf ðh1 h5 Þ
(1.45)
Q_ E ¼ m_ b c b ðTa Tb Þ ¼ m_ wf ðh1 h5 Þ
(1.46)
Q_ PH ¼ m_ b c b ðTb Tc Þ ¼ m_ wf ðh5 h4 Þ
(1.47)
_
net th ≡ QW _
(1.48)
PH=E
W_ net ¼ Q_ PH=E Q_ c ; Q_ PH=E ¼ Q_ E + Q_ PH 3 th ¼ 1 hh21 h h4
(1.49)
(1.50)
thermodynamic analyses. The equation for the analysis of a binary geothermal power plant is captured in Table 1.5. The turbine expansion process work generated is computed in Eq. (1.37) while the isentropic turbine efficiency (t) is described in Eq. (1.38). Similarly, in Eqs. (1.39), (1.40), the power of the turbine (W_ t ) as well as the power of the generator (W_ e ) are evaluated as m_ wf connotes the mass flow rate of the working fluid and g is the efficiency of the generator. Q_ c is the description of the working fluid heat discarded due to cooling during the condensation process. Eq. (1.42) presents the power transferred to the working fluid from the feed pump (W_ p ). A steady flow, well-insulated PH and E as well as insignificant potential and kinetic energy are the three propositions useful in analyzing the heat exchange process [15]. Eq. (1.43) is administered to the thermodynamic system where a symbolizes the geothermal fluid inlet while b represents the geothermal fluid after E and c after PH.
The analysis of geothermal fluid and the working fluid evaporation heat transfer rate is shown in Eq. (1.46). The known brine inlet temperature is represented as Ta; Tb is obtained from the pinch-point temperature (minimum temperature difference between two fluids supplied by the manufacturer) and the known T5. Lastly, Eqs. (1.48)–(1.50) give the performance assessment parameters of the cycle. Based on the input of thermal power (Q_ PH=E ) and the thermal power rejected (Q_ c ), Eq. (1.48) is the presentation of the thermal efficiency of the entire cycle (th) [15, 19]. During the design process of a B-GPP, choosing the working fluid is critical and entails considering the geofluid and working fluid thermodynamics characteristics as well as safety, health, and the effect on the environment [15]. The economy and the efficiency of B-GPP are described by the working fluid adoption [33]. Table 1.6 shows different working fluids and explicitly explains how working fluid critical temperatures (CT) and critical pressures (CP) are extremely lower in contrast to water. Different contemporary binary technologies have emerged, promoting advancement at higher performances via the flexible adoption of a secondary cycle in B-GPP [24]. Studies by DiPippo [15] and Valdimarsson [34] describe various other binary configurations, including the dual-pressure binary cycle, the dual-fluid binary cycle, the Kalina binary cycles, and regenerative ORC. To adopt ORC-GPP, different methods have been investigated regarding working fluid. A work by Quoilin [35] put forward an approach for the selection of the working fluid and an expansion process in the same system. For any ORC process, the working fluid and expansion mechanism are adopted by the application of this method. Mikielewicz and Mikielewicz [25] studied 20 working fluids for an ORC and concluded that R123 and R141b possess the most suitability for small-scale operations. Regarding the adoption of the best applicable working fluid for an ORC, extensive
TABLE 1.6 Working fluids commonly used in binary geothermal plants [15]. PS @ 300 k MPa
CT (° C)
PC (MPa)
96.9
4.24
0.9935
i-C4H10
135.9
3.69
0.3727
n-Butane
C4H10
150.8
3.72
0.2559
i-Pentane
i-C5H12
187.8
3.41
0.0975
n-Pentane
C5H12
193.9
3.24
0.0738
Ammonia
NH3
133.6
11.63
1.061
Water
H2O
374.1
22.09
0.003536
Fluid
Formula
Propane
C3H8
i-Butane
Various cycle configurations for geothermal power plants Chapter 1
indicators are provided in [35], namely thermodynamic performance, isentropic saturation vapor curve, high vapor density, low viscosity, high conductivity, evaporating pressure, condensing gauge pressure, high-temperature stability, melting point, low ozone-depleting potential, low greenhouse warming potential, availability, and low cost. Astolfi [36] conducted a comprehensive study of binary ORC power plants focusing on harvesting low-medium temperature geothermal sources [37]. The above authors examined 54 working fluids in six dissimilar cycle configurations and concluded that the optimal fluid is decafluorobutane at a low temperature of 120°C while at a higher temperature of 180°C, R236ea has been the optimum fluid.
1.2.5 Advanced geothermal energy conversion systems—Hybrid configurations Concerning configurations of geothermal energy conversion processes, three innovative technologies were suggested: hybrid single-flash and double-flash systems, hybrid flash-binary systems, and hybrid fossil-geothermal systems [15]. Also, the amalgamation of geothermal technologies with biomass, waste-to-energy systems, fuel cells, and solar thermal systems is attracting significant attention [12, 38]. Thain and DiPippo [12] also stated the principal significance of hybrid geothermal as other clean energy technologies under the following factors: impacts on the environment, electrical energy cost, exergy, plant performances, the performance of turbo-machinery, the viability of the technoeconomic aspects, and the investment risk. With respect to the binary power plant’s environmental impact, the geothermal fluid is evacuated and returned to the reservoir. There is no chemical or physical interaction of the working fluid with the environment. The environmental impact is the only thermal pollution that occurs as a result of heat rejection along the cycle, and this is applicable for direct heating purposes [20]. The attention of many studies focuses on the hybrid configuration of various thermal and nonthermal clean energies with geothermal resources [12]; nevertheless, hybrid geothermal-fossil fuel systems have been in existence [39]. The current interest of the research community is the study of hybrid solar and geothermal energy configurations for power plants [26, 29]. Jiang [26] conducted thermodynamic research for a hybrid solar-enhanced geothermal system power plant, where the working fluid used was CO2 in a supercritical CO2 Brayton cycle. Based on their conclusion, the hybrid system has higher efficiency compared to the aggregate of the two independent off-grid systems. Supplementary electricity was produced during peak demand hours due to the increased capacity driven by the solar system, even as the geothermal cycle supports based-load electricity. Cardemil [29] analyzed the solar-
13
concentrating parabolic collector hybridized to a singleand double-flash geothermal power plant for various geothermal reservoir situations. Based on their findings, the hybrid single-flash configuration generates an increment of 20% extra power output, and the quantity of the geothermal fluid deployed from reservoirs in the hybrid double-flash configuration dropped by 19%. Ayub and Mitsos [38] integrated two existing simulations, an ORC geothermal model and a low-temperature solar technology, and the results showed how the hybrid system’s levelized energy cost was lowered by 2% while the ORC geothermal configuration’s levelized energy cost was reduced by 8%. Zhou [39] investigated hybrid geothermal energy systems. The result of the solar-geothermal viability investigation indicates an increment in the efficiency of the net electrical output of the power plant by 12.7% as well as an increment in the performance of the thermal plant by 7.5%. A related study in [40] showed the possibility of minimizing the cost of electrical energy generation by 20% via the deployment of a solar-geothermal power plant compared to using a standalone enhanced geothermal system. For a subcritical and supercritical ORC solar-geothermal plant, the yearly electricity production increases by 15% and 19% with the exergy of the solar fraction above 66% [39]. Recently, a novel solar-geothermal hybrid power plant was proposed based on the hybridization of an existing geothermal binary cycle with a solar-powered steam-Rankine topping cycle [41]. This hybridization can produce approximately 60% more electricity per day in hot seasons, decreasing the use of geothermal resources by 17% per year. Hybrid geothermal-fossil fuel power plant configurations provide substitutes to minimize the use of fossil fuels and greenhouse gas emissions for low-enthalpy geothermal resources. Zhou [39] examined a 500 MW hybrid geothermal-coal configuration with a 210°C reservoir temperature and a 400 kg/s brine flow rate, and he showed that 0.3 million tonnes of coal per year could be saved and approximately 0.72 million tonnes of GHE per year could be reduced [27]. In addition, in comparison with the standalone geothermal power plant, electrical energy can be reduced by between 33% and 87%; also, in comparison with a sole coal-fired power plant, electricity production can be increased by approximately 19% [39]. Power generation also depends on the potential of oil and gas fields. In a study, 349 abandoned onshore oil and gas wells were investigated by Reyes [42], and it was concluded that 1.7 109 kWh could be integrated into the electricity network of New Zealand. Likewise, a recent work in [30] investigated three gas reserves in Croatia, and the results showed the possibility of the economic viability of the unexploited gas fields. A study by Davis and Michaelides [43] showed that South Texas untapped oil wells have the potential to drive about 3 MW of electrical energy. Similar studies can be found in the literature [44].
14
PART
I Basics of geothermal power plants
Few studies have focused on the geothermal and biomass power plant hybrid systems. In order to achieve a higher power output, Thain and DiPippo [12] examined the hybrid system of a geothermal-biomass power plant. The authors’ findings showed an increase in net power by 32% in contrast with the standalone power plants. Similarly, Moret [45] presented the findings of a study on geothermal biomass. Overall, the deployment of geothermal heat to advance the biomass conversion system performances is proven by innovative hybrid geothermal-biomass systems that indicate positive synergies. A recent study also proposed a novel configuration of a hybrid binary geothermal-biomass power plant [46]. This proposed configuration increased the temperature of the geothermal fluid by 28%, minimizing the operational cost associated with the biomass.
Acknowledgments Diego Moya has been funded by the Ecuadorian Secretariat for Higher Education, Science, Technology, and Innovation (SENESCYT), Award No. CZ03-35-2017; The Technical University of Ambato (UTA), Award No. 1895-CU-P-2017 (Resolucio´n HCU); and supported by The Science and Solutions for a Changing Planet Doctoral Training Partnership, Grantham Institute, at Imperial College London. The Institute for Applied Sustainability Research (IIASUR) supports international research on global sustainability applied to the Global South. We acknowledge the important comments and suggestions made by anonymous reviewers to improve the quality, clarity, and strictness of this article. Andrea Morales and Rafaela Moya are highly appreciated for their support during the development of the manuscript.
REFERENCES 1.3
Closing remarks
This study reveals other measures of minimizing global fossil fuel consumption and its environmental effects by using a geothermal energy system in a range of power plant configurations. An assessment study by Stefansson [47] shows that a geothermal potential lower limit is expected to generate 1.5 TW of electricity from the global geothermal resource. Further, the author discovered that around 68% of the aggregate geothermal resources have temperatures below 130°C (mainly for direct heat uses), and the extra resources are temperatures above 130°C (appropriate for power generation). The application of low-temperature geothermal resources might be achieved by using Binary Organic Rankine Cycle power plants. In addition, studying a hybrid geothermal-biomass configuration is expedient despite the in depth research on hybrid geothermal-solar thermal configurations. The increase of energy output, the thermal performance, and the increase of the geothermal reservoir lifespan are features of both solar and biomass. In contrast, favorable opportunities for increasing geothermal project income are attained by the direct applications of geothermal heat. The application of numerous direct uses is achievable subject to the geographic zone. The applications of the geoheat contribute to increasing the living conditions of the regions situated at the geothermal resource due to the contribution of the cascade configurations (for example) to maximize the use of geoheat. Cascade applications allow the use of different temperatures of the same geothermal fluid in staging and successive applications in a sequence that require lower temperatures downstream. Geothermal hybrid configurations along with direct uses (heat, air conditioning, refrigeration, drying, evaporation, and district heating, among others)—cascade applications—are some of the poly-generation hybrid configurations identifiable for future research opportunities.
[1] Christensen JL, Hain DS. Knowing where to go: the knowledge foundation for investments in renewable energy. Energy Res Soc Sci 2017;25:124–33. [2] Shortall R, Davidsdottir B, Axelsson G. Geothermal energy for sustainable development: a review of sustainability impacts and assessment frameworks. Renew Sust Energ Rev 2015;44:391–406. [3] Budisulistyo D, Krumdieck S. Thermodynamic and economic analysis for the pre-feasibility study of a binary geothermal power plant. Energy Convers Manag 2015;103:639–49. [4] DiPippo R. Geothermal power plants: evolution and performance assessments. Geothermics 2015;53:291–307. [5] Axelsson G, Stefa´nsson V, Xu Y. Sustainable management of geothermal resources. In: Proceedings of the international geothermal conference; 2003. [6] Chen J, Jiang F. Designing multi-well layout for enhanced geothermal system to better exploit hot dry rock geothermal energy. Renew Energy 2015;74:37–48. [7] Bertani R. Geothermal power generation in the world 2010–2014 update report. Geothermics 2016;60:31–43. [8] Eslami-Nejad P, Ouzzane M, Aidoun Z. Modeling of a two-phase CO2-filled vertical borehole for geothermal heat pump applications. Appl Energy 2014;114:611–20. [9] Zarrouk SJ, Purnanto MH. Geothermal steam-water separators: design overview. Geothermics 2015;53:236–54. [10] Braun GW, McCluer PH. Geothermal power generation in United States. Proc IEEE 1993;81(3):434–48. [11] Kipsang C. Cost model for geothermal wells. vol. 11. Reykjavik, Icelan: United Nations University; 2014. Geothermal training programme reports 2013. [12] Thain I, DiPippo R. Hybrid geothermal-biomass power plants: applications, designs and performance analysis. In: Proceedings of the world geothermal congress 2015; 2015. [cited 2017 February 2]; Available from: http://events.bioenergy.org.nz/documents/events/ Thain-DiPippo-WGC2015-Geothermal-Biomass-Hybrid-PowerPlant-FINAL-VERSION.pdf. [13] Bertani R. Geothermal power Generation in the world 2010–2014, update report. In: Proceedings of the world geothermal congress 2015. International Geothermal Association; 2015. [14] Valdimarsson P. Geothermal power plant cycles and main components. In: Presented at “short course on geothermal drilling, resource
Various cycle configurations for geothermal power plants Chapter 1
[15]
[16] [17] [18]
[19] [20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
development and power plants” organized by UNU-GTP and LaGeo, in Santa Tecla, El Salvador; 2011. DiPippo R. Geothermal power plants: Principles, applications, case studies and environmental impact. 3rd ed. Butterworth-Heinemann; 2012. Zarrouk SJ, Moon H. Efficiency of geothermal power plants: a worldwide review. Geothermics 2014;51:142–53. SmartDraw. Engineering library. San Diego, CA: Smart Draw Software LLC; 2015. Chamorro C, et al. World geothermal power production status: energy, environmental and economic study of high enthalpy technologies. Energy 2012;42(1):10–8. C¸engel YA, Boles MA. Thermodynamics: An engineering approach. New York: McGraw-Hill; 2015. Bayer P, et al. Review on life cycle environmental effects of geothermal power generation. Renew Sust Energ Rev 2013;26:446–63. Aikawa K, Soda M. Advanced design in Hatchobaru geothermal power station. In: Proceedings second UN symposium development and use of geothermal resources. Washington, DC: US Government Printing Office; 1976. DiPippo R. Geothermal energy electricity generation and environmental impact. Energy Policy 1991;19(8):798–807. Yildirim N. Modelling, simulation and optimization of flashed-steam geothermal power plants from the point of view of non-condensable gas removal systems. In: Graduate School of Engineering and Sci_ ences of. Turkey: Izmir Institute of Technology; 2010. p. 178. Valdimarsson P. New development in the ORC technology; 2014. p. 1–10. [cited 2015 14 May]; Available from: http://www.os.is/ gogn/unu-gtp-sc/UNU-GTP-SC-18-18.pdf. Mikielewicz D, Mikielewicz J. A thermodynamic criterion for selection of working fluid for subcritical and supercritical domestic micro CHP. Appl Therm Eng 2010;30(16):2357–62. Jiang P-X. Thermodynamic analysis of a solar–enhanced geothermal hybrid power plant using CO2 as working fluid. Appl Therm Eng 2017;116:463–72. Zhou C, Doroodchi E, Moghtaderi B. Assessment of geothermal assisted coal-fired power generation using an Australian case study. Energy Convers Manag 2014;82:283–300. Kopunicova´ M. Feasibility study of binary geothermal power plants in Eastern Slovakia, analysis of ORC and kalina power plants. In: The School for Renewable Energy Science. University of Iceland & the University of Akureyri; 2008. Cardemil JM. Thermodynamic evaluation of solar-geothermal hybrid power plants in northern Chile. Energy Convers Manag 2016;123:348–61. Kurevija T, Vulin D. High enthalpy geothermal potential of the deep gas fields in Central Drava Basin, Croatia. Water Resour Manag 2011;25(12):3041–52.
15
[31] Zhang X, He M, Zhang Y. A review of research on the Kalina cycle. Renew Sust Energ Rev 2012;16(7):5309–18. [32] DiPippo R. Second law assessment of binary plants generating power from low-temperature geothermal fluids. Geothermics 2004;33 (5):565–86. [33] Bao J, Zhao L. A review of working fluid and expander selections for organic Rankine cycle. Renew Sust Energ Rev 2013;24:325–42. [34] Valdimarsson P. Short course: electricity generation from low temperature geothermal resources. In: Presented at ’WGC2015’ World geothermal congress 2015, Melbourne, Australia; 2015. [35] Quoilin S. Working fluid selection and operating maps for organic Rankine cycle expansion machines. In: Proceedings of the 21st international compressor conference at Purdue; 2012. [36] Astolfi M. Binary ORC (organic Rankine cycles) power plants for the exploitation of medium–low temperature geothermal sources—part A: thermodynamic optimization. Energy 2014;66:423–34. [37] Astolfi M, et al. Binary ORC (organic Rankine cycles) power plants for the exploitation of medium–low temperature geothermal sources—part B: techno-economic optimization. Energy 2014;66:435–46. [38] Ayub M, Mitsos A, Ghasemi H. Thermo-economic analysis of a hybrid solar-binary geothermal power plant. Energy 2015;87:326–35. [39] Zhou C. A fundamental study on hybrid geothermal energy systemsj NOVA. The University of Newcastle’s Digital Repository, in School of Engineering, Chemical Engineering. Australia: The University of Newcastle; 2014. [40] Zhou C, Doroodchi E, Moghtaderi B. An in-depth assessment of hybrid solar–geothermal power generation. Energy Convers Manag 2013;74:88–101. [41] Bonyadi N, Johnson E, Baker D. Technoeconomic and exergy analysis of a solar geothermal hybrid electric power plant using a novel combined cycle. Energy Convers Manag 2018;156:542–54. [42] Reyes AG. Abandoned oil and gas wells: A reconnaissance study of an unconventional geothermal resource. GNS Science; 2007. [43] Davis AP, Michaelides EE. Geothermal power production from abandoned oil wells. Energy 2009;34(7):866–72. [44] Liu J, et al. Feasibility of combination of CO2 geological storage with geothermal-type water-soluble gas recovery in Yinggehai Basin, China. Int J Greenhouse Gas Control 2016;45:139–49. [45] Moret S. Integration of deep geothermal energy and woody biomass conversion pathways in urban systems. Energy Convers Manag 2016;129:305–18. [46] Briola S, Gabbrielli R, Bischi A. Off-design performance analysis of a novel hybrid binary geothermal-biomass power plant in extreme environmental conditions. Energy Convers Manag 2019;195:210–25. [47] Stefansson V. World geothermal assessment. In: Proceedings of the world geothermal congress; 2005.
Chapter 2
Global value chain and manufacturing analysis on geothermal power plant turbines Sertaҫ Akar, Chad Augustine, and Parthiv Kurup National Renewable Energy Laboratory (NREL), Golden, CO, United States
2.1
Global geothermal energy market
The global geothermal power market grew by 13% between 2015 and 2019, which increased the total installed geothermal power capacity from 13.65 GWe in 2015 [1] to 15.41 GWe in 2019 [2]. The countries with the highest geothermal installed capacity by the end of 2019 were the United States, Indonesia, the Philippines, Turkey, New Zealand, Mexico, Italy, Kenya, Iceland, and Japan (Fig. 2.1). Between 2005 and 2015, 190 new geothermal power plants were installed around the world, where 62% were binary-cycle plants, 31% were flash-cycle plants, and 7% were dry-steam plants [3]. The number of installed geothermal power plants is expected to grow and reach about 18.4 GWe by 2021, based on forecasts [1] and pipeline projects [4] (Fig. 2.2). This could create a market demand for a diverse mix of geothermal turbine types. Recently, it is unclear whether the expected additional capacity increases and the demands are enough to increase the manufacturing volume of both binary-cycle and flash-cycle power plant turbines. However, based on the information about proposed projects and resource assessments, a significant annual power capacity addition between 0.75 and 1 GWe can be expected in the global market. It is very promising that this growth in the geothermal market will allow standardization in turbine design, rather than the customized turbines that are being manufactured today. This increase will create a modularity in the geothermal market, which will be adapted to the geographic diversity of projects to offer an economy of scale.
2.1.1
Global value chain and trade flow
Geothermal project developers customize the size of the power plant to fit the resource being developed. Geothermal power plant turbines are designed to optimize efficiency. The best utilization of geothermal resources such as HX,
WCCT, or ACC is then chosen to complement the turbine size and design. As an example, the Salton Sea Unit 5 geothermal steam turbine in Imperial Valley, Southern California, is designed and optimized for 58.32 MWe [5]. Custom design turbines have relatively higher manufacturing set-up costs, longer lead times, and higher capital costs than the standard design turbines manufactured in larger volumes. However, turbines produced in standard increments and in larger manufacturing volumes could result in lower costs per turbine. The current manufacturing process for geothermal turbines is made to order. In other words, every order is a custom design based on geothermal fluid properties. The challenges of geothermal fluid chemistry force designs to use special corrosion-resistant metals that are more expensive than standard metals used in fossil fuel-powered turbines. Additionally, the high fixed capital costs of resource development and low power purchase agreement (PPA) prices lead developers to maximize resource utilization by customizing their turbines. The customization design factors result in greater manufacturing set-up costs, extensive engineering, and a higher lead time (up to 18 months) from initial design to installation. In turn, these factors may impact developers’ returns and decrease the attractiveness of deploying geothermal energy. There are two major geothermal turbine technologies: flash cycle steam turbines and binary cycle turboexpanders. The global steam turbine market is expected to increase from $14.5 billion in 2013 to $17.4 billion by 2020, with an annual growth rate of 2.6% over this period [6]. However, the market share of geothermal power plants constitutes a small portion of the global steam turbine market. Annual global orders for steam turbines are broadly stable at around 100 GW, and geothermal steam turbines constitute only 1%–2% of the total annual demand [6]. Usually, large coal-fired, natural gas-fired, and nuclear power plants drive the market. Major manufacturing
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00007-X Copyright © 2021 Elsevier Inc. All rights reserved.
17
18
PART
I Basics of geothermal power plants
FIG. 2.1 Global geothermal installed capacity for the top 10 countries by the end of 2019. (Data Source: Think GeoEnergy. The Top 10 Geothermal Countries 2019—based on installed generation capacity (MWe). Think GeoEnergy - Geothermal Energy News, Jan. 27, 2020. http://www.thinkgeoenergy. com/the-top-10-geothermal-countries-2019-based-on-installed-generation-capacity-mwe/ (Accessed 28 January 2020).)
Total installed Capacity (MWe)
20,000
15,000
Turboexpander (Binary) MW Steam Turbine (Flash + Dry Steam) MW
10,000
5,000
0 2007 2009 2011 2013 2015 2017 2019 2021P 2005 FIG. 2.2 Historical, current, and projected global installations of geothermal power plant turbines. P, projection. (Data displayed to represent the median figures that have been compiled from GEA. Annual U.S. & global geothermal power production report. Geothermal Energy Agency; 2016, Bertani R. Geothermal power generation in the world 2010–2014 update report. Geothermics 2016; 60:31–43. https://doi.org/10.1016/j.geothermics.2015. 11.003, BNEF. Geothermal market outlook report. Bloomberg New Energy Finance; 2016.)
locations for geothermal steam turbines are Japan, Italy, the United States, France, Mexico, Russia, India, and China, where Japan accounted for 82% of the global manufacturing market between 2005 and 2015 [1, 4, 7–10]. The second type of turbine technology is turboexpanders, which are mostly utilized in organic Rankine cycle (ORC) binary cycle geothermal plants. Apart from geothermal energy applications, the ORC technology has also been used for other commercial applications, such as waste heat recovery (WHR), bioenergy production
(from biogas and landfill gas), and concentrating solar power (CSP), over the last decade. While bioenergy has the greatest number of ORCs installed (for WHR with smaller installed sizes), geothermal power plants contributed 71% of all ORC installed capacity in the world between 2005 and 2016 (Fig. 2.3), as bioenergy and WHR followed with 15% and 13.7%, respectively [11]. The main manufacturing locations for binary cycle turboexpanders are Israel, the United States, Italy, and Germany; Israel accounts for 74% of the geothermal
Global value chain and manufacturing analysis Chapter 2
19
FIG. 2.3 Overview of global ORC turboexpander market 2005 and 2016 (lab-scale prototypes and installed capacity lower than 50-kilowatt electric (kWe) have not been included). (Data modified from Tartiere T. World overview of the organic Rankine cycle technology. 2016. https://orc-worldmap.org/index.html (Accessed 22 January 2020).)
binary cycle turboexpander manufacturing market. Italian turboexpander manufacturers have increased their market share significantly in the last couple of years [1, 4, 7–10]. The global trade flow of both geothermal steam turbines and ORC turboexpanders between 2005 and 2015 can be seen in Fig. 2.4. The United States has the highest proven geothermal resource capacity, and it is one of the major players in the geothermal power plant turbines and technologies market. The current installed geothermal capacity of the United States is 3.68 GWe, with an additional 23 MW just added before the end of 2019. The geothermal installed capacity has been growing at a rate of about 2% per year and is projected to exceed 3.9 GWe by 2022 [12]. A comprehensive study of the US geothermal market by the National Renewable Energy Laboratory (NREL) suggests that additional power plants may come online in the next 5 years if existing barriers can be removed to expedite project development [13]. Indonesia has the second-highest installed geothermal capacity, and it also has a fast-growing demand for electricity. Indonesia’s current installed geothermal power capacity is 2.13 GWe, and the government has ambitious plans for 6.50 MWe of geothermal development by 2025 [14]. Indonesia also has a high feed-in-tariff (FIT) policy, which ranges from 12.6 to 26.2 ¢/kWh [14].
Turkey has been the fastest-growing market since the last decade. The total installed geothermal power capacity is 1.53 GWe as of 2019 and it has a capacity target of 2.0 GWe, including projects in the pipeline [15]. Turkey implemented a renewable energy law in 2010 to reach its target for increasing the share of renewables up to 30% of the energy mix by 2023 [16]. The Turkish FIT for geothermal power plants is 10.5 ¢/kWh. The FIT applies for 10 years of power generation, and producers also benefit from an 85% discount on transmission costs for the 10 years [16]. The 2010 Renewable Energy Law also includes bonus payments for hardware components made in Turkey to support and boost the national manufacturing sector. Companies that rely on locally produced equipment or components receive a bonus FIT, which is fixed at 1.3 ¢/kWh for turbines, 0.7 ¢/kWh for generators, and 0.7 ¢/kWh for pumps and compressors [16] The FIT has increased the interest of developers and manufacturers in domestic manufacturing. The total FIT for geothermal could reach 13.2 ¢/kWh with 10 years of a purchasing guarantee. The FIT is applied to all geothermal power plants that come online through the end of 2020. Another important market is Kenya, which reached 0.86 GWe of installed geothermal power capacity in 2019 by adding 193 MWe of extra capacity [2]. Kenya is currently in a very ambitious phase of development with an aggressive
FIG. 2.4 Global trade flow map of geothermal turbines between 2005 and 2015. (Data Source: NREL industry outreach, GEA. Annual U.S. & global geothermal power production report. Geothermal Energy Agency; 2016, Bertani R. Geothermal power generation in the world 2010–2014 update report. Geothermics 2016; 60:31–43. https://doi.org/10.1016/j.geothermics.2015.11.003, BNEF. Geothermal market outlook report. Bloomberg New Energy Finance; 2016, BNEF. Q2 2013 geothermal market outlook report. Bloomberg New Energy Finance; 2013, BNEF. H2 2014 geothermal market outlook report. Bloomberg New Energy Finance; 2014, BNEF. H1 2015 geothermal market outlook report. Bloomberg New Energy Finance; 2015, GEA. Annual U.S. & global geothermal power production report. Geothermal Energy Agency; 2015, Graphic Credit: Billy Roberts (NREL).)
Global value chain and manufacturing analysis Chapter 2
construction pipeline of new projects in several geothermal resource areas. The total estimated resource potential of the country is around 10GWe [1].
21
2.2.1 Methodology for manufacturing analysis 2.2.1.1 Manufacturing process flow
2.2
Manufacturing analysis
A bottom-up cost model includes a mapping of all components that make up a system (i.e., labor, material, processes, machining, and the balance of the system). The bottom-up manufacturing cost model developed (and highlighted next) for geothermal turbines considers the materials, manufacturing steps, equipment, and assembly of turbine subcomponents. The process flow diagram in Fig. 2.5 highlights the raw materials, the required manufacturing processes and equipment, and the utility requirements that are inputs to the cost model. The raw metals required for preprocessing are iron ore, carbon, chromium, molybdenum, nickel, titanium, and aluminum while the processed metals are stainless steel, Inconel (nickel) alloys, and titanium alloys [17, 18]. In addition to metals, epoxy-based refined plastics are used for insulation and sealing purposes.
The manufacturing cost model includes three main steps: (1) Materials (used as raw and processed), (2) manufacturing (inhouse machining and outsourced parts), and (3) final assembly. The final product could be either an ORC turboexpander or a geothermal steam turbine.
2.2.1.2 Materials The most common corrosion-resistant materials used for machining the impellers are titanium or stainless steel; the shaft is produced from a stronger material such as a forged nickel alloy or Inconel. Geothermal fluids contain dissolved carbon dioxide (CO2), hydrogen sulfide (H2S), ammonia (NH3), and chloride (Cl) ions that can cause corrosion of metallic materials. The main corrosion problems are pit corrosion, cracking corrosion, breaking with stressed corrosion, breaking with sulfur stressed corrosion, corrosion between the particles, and wearing corrosion [18].
FIG. 2.5 Manufacturing process flow diagram for geothermal power plant turbines.
22
PART
I Basics of geothermal power plants
FIG. 2.6 World steel production, units are a million metric tons per year. (Data Source: World Steel information system Steel Dynamics. 2015 Annual report. Steel Dynamics Inc.; 2015. Available from: https://s3.amazonaws.com/b2icontent.irpass.cc/2197/165986.pdf? AWSAccessKeyId¼1Y51NDPSZK99KT3F8VG2&Expires¼1580257699&Signature¼9zZjMvMMB3ob07MYrZq%2FbCQdvl4%3D (Accessed 28 January 2020).)
Stainless steel material decreases the probability of uniform corrosion formation in a geothermal fluid environment. AISI 400 series stainless steels contain 12%–18% chrome, which is more suitable for turbine blades. AISI 430 (Ferrite) and AISI 431 (Martensitic) stainless steels are often used for valve and pump components in geothermal systems. Stainless steel production is widespread throughout the world (Fig. 2.6). Based on the World Steel Dynamics 2015 data, China, Japan, and the United States are the top three countries in stainless steel production [19]. Titanium (Ti) and titanium alloys are more resistant to corrosion. In addition, titanium is resistant to cavitation and impact damage. Ti alloys are much more resistant to local corrosion than pure titanium. Ti-code-7 (Ti-0.15 Pd), Ti code-12 (Ti-0.3 Mo-0.8 Ni), and Ti-code-29 (Ti-6 Al-4 V0.1 Ru) show good corrosion resistance [20] when they are compared based on cost and performance. The critical places for using titanium alloys can be impellers, wellhead valves, pressure gauges, pipes, and blow-out preventers. The world’s titanium production is limited to certain areas (Fig. 2.7). Based on data from the US Geological Survey (USGS) Minerals Yearbook 2015, Canada, Australia,
China, South Africa, Vietnam, the United States, Brazil, India, Mozambique, Madagascar, Norway, Ukraine, Kenya, Kazakhstan, Indonesia, Malaysia, and Sri Lanka are the main countries for titanium production [21]. Inconel, a nickel (Ni) alloy, is another important material for turbine manufacturing. There are various types of Inconel available on the market, and the mineral content defines the strength and corrosion resistance (Table 2.1). For high-temperature geothermal systems, it is suitable to use nickel, chromium, and molybdenum (Ni-Cr-Mo) alloys as a material (Kaya and Hos¸ han, 2005). Inconel-625 and Hastelloy C-256 are especially strong in combatting corrosion. Other nickel alloys that have iron elements can also be used in some applications [20]. These alloys are much stronger than stainless steel. Forged Inconel is mostly used for turbine shafts because of its strength against rotational force.
2.2.1.3 Machine inventory and factory model The factory model includes the minimum workspace required for the machines in addition to machine-related
Global value chain and manufacturing analysis Chapter 2
23
FIG. 2.7 World titanium ore production, units are thousand metric tons per year. (Data Source: USGS. Minerals yearbook—Metals and minerals. United States Geological Survey; 2015. Available from: https://www.usgs.gov/centers/nmic/minerals-yearbook-metals-and-minerals (Accessed 22 January 2020).)
TABLE 2.1 Inconel alloy element compositions by weight. Elements by mass (%) Inconel alloys
Nickel (Ni)
Chromium (Cr)
Iron (Fe)
Molybdenum (Mo)
Niobium (Nb)
600
72.00
16.00
10.00
0.00
0.00
617
44.00
24.00
3.00
10.00
625
58.00
20.00
5.00
690
60.00
30.00
9.00
718
55.00
21.00
X-750
70.00
14.00
Magnesium (Mn)
Copper (Cu)
Aluminum (Al)
Titanium (Ti)
Others
0.00
1.00
0.50
0.00
0.00
0.50
0.00
15.00
0.50
0.50
1.00
0.50
0.50
10.00
4.00
1.00
0.50
0.00
0.40
0.40
0.70
0.00
0.00
0.00
0.35
0.01
0.02
0.00
0.62
12.00
3.00
5.00
1.00
0.30
1.00
1.00
0.20
0.50
9.00
0.00
1.00
1.00
1.00
0.50
0.50
2.50
0.50
labor requirements. The machine inventory includes heavy machining and precise computer numerical control (CNC) machining processes [22] (Table 2.2). Heavy machining includes electric arc furnace casting and forging operations. CNC machining includes a five-axis CNC machine, a threeaxis CNC machine, a CNC horizontal lathe, and a CNC grinding machine. The quality control (QC) pieces are a coordinate measuring machine (CMM) and overspeed testing and dynamic balancing (OSTB). The proposed manufacturing model has not only been tested for geothermal turbines, but has also been validated against other industries such as solar PV [23, 24]. One other important parameter in a factory model is the annual maximum allowable working hours (MAWH). MAWH can be defined by the annual total labor hours and number of shifts as well as production-up times. As an example, 250 annual labor days and 8 working hours with 2 shifts per day with 85% production-up time, make a total
Cobalt (Co)
of 3400 MAWH. The machining rate for each machine is based on the MAWH and operation hours with and without set-up time for the factory model. In a dedicated factory model, machines are utilized as much as possible across several different projects to fulfill the MAWH because the capital cost associated with the facilities and equipment is applied over the total time of MAWH. However, in a shared factory model, the capital cost associated with buildings, facilities, space, and the depreciation of machines is proportionally distributed over the time when the machine is utilized for manufacturing the specific turbine parts. The amount of required machinery is selected based on total operational hours for different volumes of manufacturing and MAWH. With up to 100 units per year of manufacturing volume, one of each machine would be enough to fulfill the target manufacturing volume. For more than 100 units per year, additional machines would be required (Table 2.3). The manufacturing volume of 50 units per year
24
PART
I Basics of geothermal power plants
TABLE 2.2 Machine inventory for the custom factory model. Energy consumption (kW)
Machine type
Price ($)
Footprint (m2)
Five axis CNC machine
$150–$300 k
10–15
20–30
Three axis CNC machine
$100–$200 k
10–15
20–30
CNC horizontal lathe
$60–$150 k
12–18
30–40
CNC grinding machine
$80–$150 k
35–40
10–20
Casting
$500 k–$1 M
1000
500
Forging
$400–$500 k
1000
500
Over speed testing and balancing machine (OSTB)
$10–$20 k
10
5–7
CMM dimensioning
$8–$10 k
10
1–3
Assembly line
$50–$300 k
50–60
5–10
calculated based on the average power consumption of each machine, operating for a given number of operational hours. The storage and shipping costs of the turbine parts/components are not included in the factory model.
2.2.1.4 Machining cost analysis The machining costs of the key components of turboexpanders, including impellers and shafts, are calculated by using design for manufacture and assembly (DFMA) software. DFMA allows the user to produce a detailed projected cost of the component based on the volume of material needed, the machines and process steps, the machine setup time, and tooling if needed [25]. Fig. 2.9 shows the representative material and machining cost estimates of a typical impeller for both a custom design and a standard design (at a volume of 10 units per year) 5 MWe turboexpander. Tooling investment considers tool wear and lifetime, and it is calculated as an additional cost element for processes that require part-specific dies and tools. A custom design 5 MWe impeller could be $4000/unit, compared to $1000/unit with the standard design (Fig. 2.8). Assuming the same yield rate, the standard design impellers can lead to a cost savings of between 25% and 30% compared to a custom design (single unit) because of the set-up times for machining the impeller. A similar approach is applied to other subcomponents of a turboexpander such as shafts, nozzles, inlet guide lanes, disks, and casings to calculate machining costs.
Data Source: NREL Industry outreach.
can be set as a threshold based on manufacturers’ annual manufacturing capacities and project portfolio. Annual straight-line depreciation is selected for capital costs associated with machinery, as is handled in accounting procedures. Facility cost is defined based on the minimum required working area for each machine. Energy cost is
2.3
Definition of minimum sustainable price
The minimum sustainable price (MSP) is the minimum price that a company would have to charge for a good or service to cover all variable and fixed costs and make enough profit to repay investors at their minimum required
TABLE 2.3 Number of required machines for different volumes of manufacturing at MAWH.
#Units
Five axis CNC machine
Three axis CNC machine
CNC horizontal lathe
CNC grinding machine
CMM
OSTB
Assembly line
10
1
1
1
1
1
1
1
25
1
1
1
1
1
1
1
50
1
1
1
1
1
1
1
100
1
1
1
1
1
1
1
150
1
2
1
1
1
1
1
200
1
2
1
2
1
1
1
500
2
5
3
4
1
1
2
1000
3
9
5
7
1
2
3
Global value chain and manufacturing analysis Chapter 2
Part
Material
Procurement
Impeller
Titanium
Plate
Part
Machining Process
Raw Material Purchased MaterialEstimated Volume Unit Price ($/kg)Unit Price ($/kg) (m3)
22.82
Process Setup Time/Unit Time/Unit (hours) (hours)
39.04 Machine Rate/Unit ($/hour)
0.037
Material Estimated Density (kg/m3) Weight (kg)
4,500
Manufacturing Machining Subtotal Cost/Unit Cost/Unit ($) ($)
25
Material Cost ($)
167
6,500
Center Hole Drilling
1 Unit (Custom Design) Drilling
Impeller
0.8
35
0.2
35
5-Axis CNC Roughing
25.0
5.0
35
1,055
5-Axis CNC Rest Milling
42.0
8.0
35
1,760
5-Axis CNC Finishing
10.0
2.0
35
422
2.5
0.5
27
80
20.0
4.0
25
648
35
9
Blade Roughing
Highly flexible simultaneous 5-axis roughing
4,000 Hub Finishing
QC Balancing
Optimized tool paths for finishing hubs
10 Units (Standard Design) Drilling
Impeller
0.3
0.2
5-Axis CNC Roughing
2.5
5.0
35
264
5-Axis CNC Rest Milling
4.2
8.0
35
440
5-Axis CNC Finishing
1.0
2.0
35
105
QC
0.3
0.5
27
20
Balancing
2.0
4.0
25
162
Rest Milling
Automate removal of remaining material
1,000 Blade / Splitter Finishing
Automate finishing of blades and splitters
FIG. 2.8 Representative material and machining cost estimates of a typical impeller for both a custom design and a standard design (at a volume of 10 units per year) 5 MWe turboexpander.
rates of return [24]. The MSP is computed by setting the net present value (NPV) of an investment equal to zero with the internal rate of return (IRR) equal to the weighted average cost of capital (WACC). The US capital assets pricing model is used to derive these debt and equity ratios, and to weight them by their relative contribution to the overall capital structure of the firm to estimate WACC values [26]. The purpose of the discounted cash flow (DCF) is to create a detailed financial model to provide the necessary framework for deriving the MSP for each product in a manufacturing facility. Within the DCF, there can be several considerations for manufacturing, such as capital costs, fixed operating costs (labor, depreciation, inflation, taxes, insurance, and rent), typical sales, general and administrative (SG&A) expenses, typical design and engineering (D&E) costs, and warranty coverage [24]. DCF model uses a simple straight-line depreciation for expenditures such as equipment and facilities, and the discount rate is calculated from the required rate of return (ROR). A summary of the financial input parameters required for DCF analysis can be found in Table 2.4. The MSP is derived by an iterative algorithm that runs until the NPV of the cash flow equals the total initial capital expenditure.
2.4
Manufacturing analysis case studies
The manufacturing cost and MSP for three different scenarios are calculated for three case studies, where each scenario had five volumes of manufacturing (1, 5, 10, 25, and 50): (1) 1 MWe ORC turboexpander. (2) 5 MWe ORC turboexpander. (3) 20 MWe steam turbine. All three scenarios assume US production facilities and costs. The generator is a separate piece and is not included in the manufacturing cost analysis. Increasing volumes of manufacturing effectively decreased the manufactured cost per unit, as we spread the capital expenditures (CAPEX) over more units. Machine set-up times and D&E costs are the cost components that are most impacted by volume manufacturing, as these are essentially one-time charges that are not volume-dependent. In Case 1, the results show that MSP decreases significantly when the volume of manufacturing is increased from 1 unit (custom design) to 5 units (standard design). The MSP of a single custom design 1 MWe turboexpander is found to be 893 $/kW, whereas a standard design 1 MWe
26
PART
I Basics of geothermal power plants
TABLE 2.4 Summary of input parameters for DCF analysis. Inputs for DCF calculations
Values
Units
Inflation on cost of goods sold (COGS)
3
%
Corporate interest rate
3.3
%
Initial loan (or bond) maturity
10
Years
Corporate tax rate
30
%
Dividend payout rate
0
%
Cost of equity
10.6
%
Cash flow analysis period
20
Years
Working capital collection period
10
Years
Calculated WACC
5.3
%
Working capital inventory turnover
4
Years
Working capital payable period
10
Years
CAPEX Initial target capital structure (% of debt)
64
%
Replacement equip. target capital structure
50
%
Depreciable life for plant
25
Years
Capital replacement loan maturity
10
Years
Equipment depreciation type
7 years straight-line
N/A
Tooling depreciation type
3 years straight-line
N/A
Building depreciation type
15 years straight-line
N/A
turboexpander has an MSP of 226 $/kW at a manufacturing volume of five (Fig. 2.9). Effectively, a standard turboexpander design, even at low manufactured volumes, could save approximately 75% of the turboexpander $/kW. In Case 2, the results show that MSP decreases significantly when the volume of manufacturing is increased from 1 unit (custom design) to 5 units (standard design). The MSP of a single custom design 5 MWe turboexpander was found to be 216 $/kW, whereas a standard design 1 MWe turboexpander has an MSP of 66 $/kW at a manufacturing volume of five (Fig. 2.10). In Case 3, a manufacturing volume of up to 5 units per year is selected based on the annual demand for geothermal steam turbines and the manufacturing capacities. The MSP of a single custom design 20 MWe geothermal steam turbine is found to be 361 $/kW, whereas the MSP of a standard design 20 MWe steam turbine is calculated as 135 $/kW at an annual production rate of 5 units per year (Fig. 2.11). Effectively, a standard steam turbine design, even at low manufactured volumes, could save approximately 63% of the steam turbine $/kW. A comparison of the MSP analysis for all three cases can be found in Table 2.5. The manufacturing cost of a custom design 5 MW ORC turboexpander is only $187,000 more than that of a custom design 1 MW ORC turboexpander. This shows that the size of the turbine does not have a significant effect on the total cost of the turbine/turboexpander. However, if the unit costs per MW for both custom and standard design cases are considered, the manufacturing cost savings are significant (667 $/kW for a 1 MW turboexpander and 150 $/kW for a 5 MW turboexpander).
2.4.1
Sensitivity analysis
Sensitivity analysis determines how the target manufacturing cost model is affected based on changes in cost factor
FIG. 2.9 Calculated MSP and manufacturing cost breakdown for a 1 MWe ORC turboexpander in different volumes of manufacturing in the United States.
Global value chain and manufacturing analysis Chapter 2
27
FIG. 2.10 Calculated MSP and manufacturing cost breakdown for a 5 MWe ORC turboexpander in different volumes of manufacturing in the United States.
FIG. 2.11 Calculated MSP and manufacturing cost breakdown for a 20 MWe geothermal steam turbine in different volumes of manufacturing in the United States.
TABLE 2.5 Comparison of MSPs for standard and custom design turbines. MSP
Custom design Single unit
Standard design Volume of 5 units
1 MW turboexpander
$893,000
893 $/kW
226,000 $
226 $/kW
$74,000
74 $/kW
5 MW turboexpander
$1,080,000
216 $/kW
332,000 $
66 $/kW
$152,000
30 $/kW
20 MW steam turbine
$6,350,000
361 $/kW
2,790,000 $
135 $/kW
N/A
N/A
variables (input variables). The impact of each input on the calculated MSP can be calculated by varying one input variable while keeping the others constant. Each cost factor in the overall cost model has a different weight based on the relative importance, and a change in one input variable would
Standard design Volume of 50 units
thus have proportional effects relative to the weight on the manufactured cost. For the sensitivity analysis, a custom design single unit 5-MWe ORC turboexpander and a custom design single unit 20-MWe steam turbine were evaluated with respect to their standard design higher manufacturing
28
PART
I Basics of geothermal power plants
volume alternatives. The manufacturing volume is set at 10 units per year for the ORC turboexpander and 5 units per year for the steam turbine. The D&E time for a custom design 5 MWe ORC turboexpander is assumed to take 9 months of labor from two full-time employees (FTEs). Thus, D&E is the most important cost factor for a custom design unit due to the time spent on a tailor-made design for each custom unit (Fig. 2.12). Custom design turbine manufacturing has a longer set-up time with respect to high volume standard design turbines. The machining set-up time constitutes 51% of the total machining cost for a custom design unit.
This makes the manufacturing labor the second most important cost factor in a custom design unit. Other important cost factors are sales and general administration (SG&A), equipment cost, energy cost, and material cost. Their effect is less important on manufacturing costs for a custom design unit. However, when the manufacturing model has changed to a standard design at a volume of 10 units year, the material and the labor costs become the dominant cost factors with shares of 46% and 31%, respectively. On the other hand, D&E and SG&A costs become less important. The cost drops by the cost factor are also presented on cost waterfall charts (Fig. 2.13).
Design & Engineering Labor Selling, General & Administration Materials Capital Energy Outsourced Parts Maintenance –50000
–40000
Maintenance
–30000
–20000
Outsourced Parts
–10000
0
Energy
10000
Capital
20000
30000
Materials
40000
50000
Selling, General & Administration
25.00%
$221
$1,288
$3,545
$11,602
$11,743
$16,646
–25.00%
–$221
–$1,288
–$3,545
–$11,602
–$11,743
–$16,646
Labor
Design & Engineering
$29,394 –$9,394
$39,658 –$39,658
Design & Engineering Labor Selling, General & Administration Materials Capital Energy Outsourced Parts Maintenance –50000
–40000
Maintenance
–30000
–20000
Outsourced Parts
–10000
Energy
0
10000
Capital
20000
30000
Materials
40000
50000
Selling, General & Administration
Labor
Design & Engineering
25.00%
$55
$1,213
$1,358
$2,900
$11,061
$1,665
$7,348
$3,966
–25.00%
–$55
–$1,213
–$1,358
–$2,900
–$11,061
–$1,665
–$7,348
–$3,966
FIG. 2.12 Sensitivity analysis for 5 MWe turboexpander based on (A) Manufacturing volume of 1 unit per year (custom design) and (B) Manufacturing volume of 10 units per year (standard design) in the United States.
Global value chain and manufacturing analysis Chapter 2
29
FIG. 2.13 Manufacturing cost drop by cost factor for a standard design (10 units) 5 MWe ORC turboexpander.
The custom design 20 MWe steam turbine manufacturing has high labor requirements during assembly. This makes labor cost the most important cost factor for a custom design unit (Fig. 2.14). Labor includes set-up time, which is 49% of the total machining cost for a custom design single unit. Capital cost, including the equipment and facilities cost, is the second most important cost factor. D&E time for a custom design 20MWe steam turbine is assumed to take 12months and four FTEs due to time spent on tailor-made parts for each unit. The design of steam turbines is more detailed than ORC turboexpanders because they are in direct contact with saturated steam as well as noncondensable gases (NCG) such as H2S and CO2, and they have multiple pressure stages. SG&A, capital (equipment and facilities), and materials are the other important factors that have a moderate effect on the manufacturing cost for a custom design unit. For one-off design turbines at a volume of 5 units, while the impact factor of labor and material stays almost the same, the D&E and SG&A costs becomes less important. The cost drops by the cost factor are also presented on cost waterfall charts (Fig. 2.15).
2.5 Power plant design and performance analysis The purpose of the turbine performance analysis is to determine the commercially favorable operating range of a standard ORC compared to custom designed ORC equipment. A typical example of the process flow model for an ORC geothermal power plant at a given design point of a standard size (5 MWe) turbine can be seen in Fig. 2.16.
The balance of plant (BOP) is optimized to maximize power generation. In other words, the BOP, including the heat exchanger, the air-cooled condenser, pumps, and piping, can be designed to optimize turbine output. The design assumptions for the optimized system include the pinch point temperature. The design point is set at 175°C for the inlet brine temperature and 80 kg/s for the brine mass flow rate for the standard turbine. An optimization algorithm is developed to optimize BOP and operating conditions by adjusting the pressure before and after the turbine for maximum turbine output at given geothermal inputs. The performance of the standard turbine is compared to a custom design turbine by running off-design models for varying geothermal resource temperatures (between 160°C and 190°C) and brine flow rates (between 40 and 120 kg/s). A turbine off-design efficiency curve provided by a reliable manufacturer as a function of the mass flow rate of the working fluid is used to evaluate the impact on power generation of the standard versus custom design (Fig. 2.17). The design point isentropic efficiency is selected as 80%. The turbine efficiency curve shows relative efficiency as a function of relative working fluid mass flow rate at a constant isentropic enthalpy drop across the turbine. The curve does not account for changes in isentropic enthalpy drop. In IPSEpro modeling, both the working fluid mass flow rate and the isentropic enthalpy drop across the turbine vary. However, the turbine model only considers the working fluid mass flow rate when adjusting the turbine isentropic efficiency. The resulting efficiency curve is likely not representative of actual turbine performance and is used only for illustrative purposes in this report. Turbine manufacturers and project developers have access
FIG. 2.14 Sensitivity analysis for 20 MWe turboexpander based on (A) manufacturing volume of 1 unit per year (custom design) and (B) manufacturing volume of 5 units per year (standard design) in the United States.
FIG. 2.15 Manufacturing cost drop by cost factor for a standard design (5 units) 20 MWe steam turbine.
Global value chain and manufacturing analysis Chapter 2
GTO CEMAC ORC Basic Organic Rankine Cycle
Cycle Characteristics
5
70.35 7.721
478.51 112.18
70.35 7.521
1
478.51 111.79
Turbine Enthalpy Drop
4706.83
[kW]
Cycle El. Power Output
4520.44
[kW]
3750.55
[kW]
34655.48
[kW]
Net El. Power Output Total heat Input
80 741.73 175 20
Gross Cycle Efficiency
13.04
Net Cycle Efficiency
12.70
Condenser Duty 70.35 7.821 70.35 7.821
198.74 112.8 198.74 112.8
70.35 0.817
411.6 65.306
80 495.71 19.9 117.8
9.21
Turbine Mass Flow Turbine Inlet Volumetric Flow
[-] [kg/s]
70.35
Turbine Outlet Volumetric Flow
3411.57
[I/s]
32726.19
[I/s]
12
80 308.54 19.8 73.326
% [kW]
30051.18
Turbine Pressure Ratio 2
%
3066 25.157 1.013 24.899
4 11
3
70.35 -14.101 7.921 29.827
70.35 -15.559 0.807 29.455 3066 15.152 15 1.013
-625.5 #
p [bar abs]
1 2 3 4 5 11 12
7.52 0.82 0.81 7.92 7.72 1.01 1.01
x
h
t [°C] 111.79 65.31 29.45 29.83 112.18 15.00 24.90
[kJ/kg] 478.51 411.60 -15.56 -14.10 478.51 15.15 25.16
fluid [-] 1.00 1.01 0 0 1.18 0 0
n-Pentane n-Pentane n-Pentane n-Pentane n-Pentane Cooling Water Cooling Water
FIG. 2.16 Process flow diagram of standard-size ORC power plant [27].
FIG. 2.17 Off-design turbine efficiency curve.
ORC Fluid mass[kg/s] h[kJ/kg] p[bar] t[°C]
Geothermal Brine mass[kg/s] h[kJ/kg] p[bar] t[°C]
Air mass[kg/s] z[kg/kg] p[bar] t[°C]
31
32
PART
I Basics of geothermal power plants
FIG. 2.18 Actual plant brine effectiveness.
to actual turbine performance curves and can use the methodology in this report to assess the potential benefits of standard turbine design. One other important parameter in the plant performance analysis is the brine effectiveness (BE). Simply, BE is the amount of energy that you can extract per pound of geothermal brine or steam, which is defined as net plant output divided by the brine flow rate (w-h/lb). The use of BE to describe plant performance comes from the Geothermal Technology Evaluation Model (GETEM, 2016) on which the System Advisor Model (SAM) geothermal module is based. In SAM, BE is set by adjusting the plant efficiency input. According to IPSEpro modeling results, the BE of binary plants studied varies between 3.3 and 7.5 w-h/lb (Fig. 2.18). This value is 5.9 w-h/lb for the standard turbine at its design point in IPSEpro. The BE value determines the more conventional thermal to electric conversion efficiency (TE) of the plant. TE varies as a function of inlet geothermal brine temperature and mass flow rate (Fig. 2.19). TE is calculated as 10.83% at the design point for the base case IPSEpro model.
2.6
Economic analysis
Monetizing the processes developed in power plant performance modeling helps project developers to convert performance calculations into a representative technoeconomic model of a total geothermal power plant. NREL’s SAM is used to perform a DCF analysis of standard and custom
design turbines using results from IPSEpro over the range of geothermal resource temperatures and flow rates of interest. The base case inputs for the geothermal resource are applied to SAM inputs and a base case model is established (Table 2.6). To compare the projects and results on a common basis, the “exact number of wells” option is chosen in SAM, and the number of production wells is set at one. For the base case, this results in a gross turbine output power capacity (nameplate capacity) of 5 MW, so that the power plant cost values from the MSP analysis can be used. SAM allows the user to set plant efficiency (%), which sets the plant BE as a percentage of the maximum brine effectiveness [28]. Setting the plant efficiency to 100% gives a plant with BE equal to the maximum brine effectiveness. Setting the plant efficiency to 50% provides a plant with brine effectiveness equal to 50% of the maximum BE. Using these data, a reverse calculation of the plant efficiency is needed to match BE values from IPSEpro runs. The binary plant efficiency is set to 65.1% to match the IPSEpro BE results in w-h/lb for the base case. The system cost scenarios are developed for custom design and standard design turbines. The SAM version of GETEM does not currently include the ability to automatically estimate plant costs, but the Excel version of GETEM does. Therefore, GETEM is used to estimate the plant costs and those values are imported in SAM. For the custom design scenarios, the plant size and efficiency results from the IPSEpro model are used as inputs to GETEM to estimate the plant costs. Plant costs in GETEM are determined by
Global value chain and manufacturing analysis Chapter 2
33
FIG. 2.19 Thermal to electric conversion efficiency for 5 MWe ORC turbine.
TABLE 2.6 Base case geothermal resource characterization for SAM financial model.
TABLE 2.7 Financial parameters for SAM model. Parameter
Unit
Value
PPA price
/kWh
10.00
Annual escalation rate
%
1.00
IRR target
Years
20.00
Project debt ratio
%
60.00
Real discount rate
%/ year
5.5
80
Inflation rate
%
2.5
1
Nominal discount rate
%/ year
8.15
Annual interest rate
%
7.00
Incentives (PTC/ITC)
$
0.00
Depreciation structure (5 years MACRS)
%
100.00
Parameter
Unit
Value
Resource temperature
°C
175
Reservoir pressure change per 1000 lb
psi-h
0.35
Reservoir depth
m
2000
Temperature decline rate
%/year
0.3
Number of production wells
–
1
Production well flow rate
kg/s
Number of injection wells
–
estimating the individual costs for the major plant components (turbine, heat exchangers, condenser, and working fluid pump) and using a direct-cost multiplier to account for the piping, instrumentation, construction costs, etc. This value is then used as the input for the specified plant cost in SAM. For the standard design scenarios, the same individual component costs and direct cost multiplier are used, but the turbine cost is decreased by $150/kW to reflect the cost savings from using a standard turbine design (see Table 2.5). Results from IPSEpro are used as the BE (plant efficiency) inputs in SAM for the custom and standard scenarios to account for the reduced efficiency of the standard turbine (compared to the custom turbine) when it operates at off-design conditions.
For the DCF analysis, a business model is developed with standard financial assumptions for all scenarios (Table 2.7). Changes in financial parameters would affect the NPV of costs. The simplest business model is a 100% equity model in which the developer pays cash for the project at the start of operations. In this case, the standard turbine is not as competitive as a custom turbine. Realistically, the more you defer costs to the future (debt) or offset costs in the future (depreciation, tax advantages), the more the custom turbine design is favored.
34
PART
I Basics of geothermal power plants
2.6.1 Decision criteria used in SAM financial model The decision criteria of the SAM financial model are functions of: l l l l l l
Electricity generated. Power purchase agreement (PPA) price. Analysis period/project life. Project equity investment amount. Annual project costs. Discount rate.
and debt-related costs. Real discount rate is a measure of the time value of money expressed as an annual percentage. SAM’s financial model results are very sensitive to the real discount rate input [29, 30]. The NPV is calculated as: X n Fn NPV ð$Þ ¼ 1 ð1 + d Þn where l l
The power purchase agreement (PPA) price is the bid price that the project receives for each unit of electricity that the system generates. Levelized PPA uses the discount rate to determine the present value of the project’s PPA revenue over its lifetime. For PPA models, SAM assumes the project sells all the electricity generated by the system at a price negotiated through a PPA. A financially viable project is likely to have a levelized cost of electricity (LCOE) that is less than the levelized PPA price to cover project costs and meet IRR requirements. If there is no profit margin between the price of supply (PPA price) and the cost of production (LCOE), a project will be financially unviable. The IRR is a measure of the project’s profitability and is defined as the nominal discount rate that corresponds to an NPV of zero [29, 30]. SAM uses a search algorithm to find the PPA price required to meet the target IRR and reports NPVs for the project. The NPV is the net present value of the after-tax cash flow discounted to year one using the nominal discount rate. The PPA price determines annual revenue. The net capitalized cost is the sum of the total installed cost and debt, other financing fees, and reserve funding from the financial parameters, less investment-based and capacitybased incentives. SAM also allows users to specify parameters such as up to five construction loans to approximate interest during construction (IDC) that SAM considers a cost to the project. The project term debt input variables determine the size of the debt or the amount borrowed
Net cash flow in year n, $ (Fn). Annual discount rate (d).
The simplified LCOE calculation uses the user-defined installation cost, operating costs, and a fixed charge rate as input, and the model calculates the LCOE based on the annual energy generated by the system. The calculator can also calculate the fixed charge rate when users provide basic financial parameters. The list of financial parameters required to calculate financial outputs can be found in Table 2.8. The LCOE is calculated as: LCOE
$ ¼ kWh
Ccap + FOC Xn ðAEP er Þ + VOC ð1 + r Þn
1
where l l l l l l
Project lifetime, years (n). Capital cost, $ (Ccap). Fixed annual operating cost, $ (FOC). Variable operating cost, $/kWh (VOC). Nominal discount rate (r). Annual electricity production, kWh (AEP).
2.6.2
SAM results and discussion
To start with, SAM scenarios are created for custom design and standard design turbines for the base case (175°C temperature and 80 kg/s mass flow rate), where it is assumed that the standard and custom turbine designs have identical performance. The net electricity generation capacity is used to calculate annual revenue from electricity sales. The
TABLE 2.8 Summary of financial parameters used to calculate financial outputs. PPA (revenue)
Discount rate
Project costs
Expenditures
X
X
IRR
X
NPV
X
X
X
X
LCOE
X
X
X
X
Levelized PPA
X
X
Electricity generation
IRR target year
Analysis period
X X X
X
X
X
Global value chain and manufacturing analysis Chapter 2
TABLE 2.9 Comparison of SAM financial model results for custom and standard design scenarios.
Metric
Unit
Custom design (base case)
Levelized COE (nominal)
¢/kWh
10.49
9.82
Levelized COE (real)
¢/kWh
8.13
7.61
Net present value (NPV)
$
$1,346,430
$2.786.840
Internal rate of return (IRR)
%
7.20%
11.99%
Year IRR is achieved
Year
20
20
IRR at the end of the project
%
10.03%
13.66%
Net capital cost (NCC)
$
$24,456,800
$22,144,500
Equity
$
$9,782,720
$8,857,800
Size of debt
$
$14,674,080
$13,286,700
NCC difference
$
+$2,312,300
NPV difference
$
+$1,440,410
Standard design (base case)
l
l
35
Standard turbines are more cost-effective than custom turbines near the design point and less cost-effective away from it. This is because the standard turbine cannot perform at a higher isentropic efficiency than the custom turbine; it can only be equal or less. The NPV differences between standard and custom design scenarios show 45 of 63 test cases that resulted in positive values where standard design turbines are favorable (Fig. 2.22).
Using a standard turbine design results in an NPV that is higher than when using a custom turbine design over a large range of geothermal brine temperatures and flow rates, as shown in Fig. 2.22. The highest NPV results tend to be at elevated geothermal brine temperatures and flow rates. The figure does not consider practical limitations on the power output from the standard turbine. The actual output from the model can be much larger than the design output of 5000 kW, as shown in Fig. 2.23. In practice, a turbine would be unable to operate at this high an output above its design point. The cut-off output for the standard design would change based on the technical specifications of different turbine designs, but a large portion of the upper right part of Fig. 2.22 is not in the practical operating range of the standard turbine design. Turbine manufacturers and project developers should keep these limitations in mind when evaluating this chart.
2.6.2.1 Sensitivity analysis
results show that the standard design turbines provide savings at the net capital cost and result in a higher NPV and IRR for the project at the given base case conditions (Table 2.9). While the net capital cost saving may reach +$2,312,300, the difference between the NPV of standard design and custom design turbines could reach +$1,440,410. Then, the financial analysis is conducted over 63 offdesign cases by changing the inlet geothermal brine temperature (between 160°C and 190°C) and the inlet mass flow rate (between 40 and 120 kg/s). The standard turbine power generation capacity is taken as 5 MW with off-design power outputs ranging between 1.4 and 6.9 MW gross. The results for standard turbines operating at off-design conditions show that: l
l
Net capital cost in $/kW significantly decreases with respect to increasing geothermal brine temperature and mass flow rate (Fig. 2.20). The standard turbines are competitive over a wide range of temperatures and flow rates, and they give positive NPV for cases near the design point (Fig. 2.21). (Areas at the right side of the economic boundary curve represent positive NPV.)
As described in Section 2.5, a full turbine performance curve is needed for detailed analysis. Therefore, the results above are only illustrative of the relative costs and performance of standard and custom turbine designs. Although the data are inadequate to accurately model off-design turbine performance, they give the information needed to determine the relative efficiency at which a standard turbine design is cost-competitive with a custom turbine design. The sensitivity analysis is conducted based on the impact of turbine performance on the NPV of the power plant by iteratively varying geothermal brine temperature and flow rate to calculate the isentropic turbine efficiency at the break-even NPV point (Fig. 2.23). In other words, the relative isentropic efficiency of the turbine is set to achieve the NPV of the custom plant equal to the NPV of the plant with a standard turbine. This is the economic boundary between the standard design and custom design turbines. For the sensitivity analysis, 63 test case scenarios are taken, and 1008 observation points are generated for different relative isentropic efficiencies with respect to the design point ranging between 85% and 100%. The results for select cases (minimum, design, and maximum geothermal brine temperature and flow rates) are shown in Fig. 2.24 and for all cases in Fig. 2.25. In these figures,
36
PART
I Basics of geothermal power plants
FIG. 2.20 Net capital cost per kilowatts for different off-design cases of the standard turbine.
FIG. 2.21 NPV after tax for different off-design cases of the standard turbine.
Global value chain and manufacturing analysis Chapter 2
37
FIG. 2.22 NPV difference between custom and standard design scenarios for given resource conditions. The black solid line represents the economic boundary of standard turbines where the NPV difference is zero.Areas at the right side of the economic boundary curve represent positive NPV cases where standard design turbines are favorable.
FIG. 2.23 Standard turbine design gross turbine output in kilowatts as a function of geothermal brine temperature and flow rate. Standard turbine design output (nameplate capacity) is 5000 kW.
38
PART
I Basics of geothermal power plants
FIG. 2.24 Sensitivity analysis for NPV difference with respect to relative isentropic efficiencies for select cases.
FIG. 2.25 Sensitivity analysis for NPV difference with respect to relative isentropic efficiencies for all cases. Gray dashed line, gray dotted line, and black solid line represent the lower limit, median, and upper limit, respectively.
Global value chain and manufacturing analysis Chapter 2
39
FIG. 2.26 The required isentropic efficiency of the standard turbine relative to a custom turbine to get a break-even NPV.
the standard turbine design is cost-competitive at a given relative isentropic efficiency if the NPV difference (standard design NPV minus custom design NPV) is positive. There is a large range of relative isentropic efficiencies over which the standard turbine design is cost-competitive for the maximum and design geothermal brine temperatures and flow rates (Fig. 2.24). For the lowest geothermal brine flow rate and temperature, the standard design is not competitive, even at 100% relative efficiency. The correlation between the NPV difference versus relative efficiency is linear (Fig. 2.26). By fitting a linear curve to each case and calculating the relative efficiency where the NPV is zero, the break-even isentropic efficiency is determined for each case, or the relative isentropic efficiency of the standard turbine necessary to make the project cost-competitive with a custom turbine design. The results of this analysis are shown in Fig. 2.26. The results show that the NPV of the project is sensitive to turbine isentropic efficiency. The results also imply that a detailed turbine efficiency analysis is needed for more precise economic analysis, and this is done at the project level. Fig. 2.26 shows that for lower temperature and flow rates, a standard turbine requires an isentropic efficiency greater than zero to be cost-competitive. The reason for this is illustrated in Fig. 2.27, which shows the total plant cost savings from using a standard turbine design versus a
custom turbine design for each case. The standard turbine cost is fixed for each case while the custom turbine cost depends on its size and efficiency. At low geothermal brine temperatures and flow rates, where the plant power output is lower, the plant cost for the custom turbine is lower than for the standard turbine because of the small turbine size. To compensate, the standard turbine would have to operate at higher efficiency and generate more electricity than the custom turbine to be cost-competitive. This illustrates that at some point, building a smaller custom turbine at a higher cost ($/kW) offsets the cost savings from a standard (but oversized) turbine. This is the type of information a manufacturer would need to consider when deciding which sizes or design power generation capacities to choose for a series of standard turbine designs.
2.7
Closing remarks
The current global geothermal turbine market is driven by custom turbines that are designed specifically to fit a developer’s demand for plant efficiency and varying geologic conditions encountered at different geographies and geothermal systems. The global manufacturing and value chain of geothermal turbines is dominated by a small number of players in the market and in certain geographic locations, which are selected by the manufacturer’s country of origin and their strategic decisions. Most of the time, the
40
PART
I Basics of geothermal power plants
FIG. 2.27 Plant cost savings (standard minus custom) as a function of geothermal brine temperature and flow rate.
manufacturing locations are not the same as the location of geothermal areas, which creates a high volume of trade flow between the country of the manufacturer and the country of installation. The results of MSP analysis show that standard design turbines can have significant cost savings with respect to the custom design single unit. The MSP calculations and sensitivity analysis also show that MSP could highly vary based on turbine size, standardization, and the volume of manufacturing, and the economy of scale applies both to the size of the turbine and the number of units manufactured in a single run. Even though the standardization of turbines makes a significant change in the upfront capital, some degree of custom design components may still be required. As an example, geothermal steam turbines often require custom metals because the geothermal fluid is highly corrosive and the level of corrosion changes by the nature of geothermal reservoir conditions and fluid chemistry at different geothermal sites. Sensitivity analysis for manufacturing shows that the labor and D&E costs are the main cost factors for a custom-designed unit. Manufacturing costs decrease significantly with increasing manufacturing volume due to shorter set-up times and spreading the D&E cost among the total number of units manufactured per year. This creates an opportunity for turbine manufacturers to realize manufacturing cost savings due to labor and D&E by switching from custom to standard design at larger volumes. If manufacturers at all steps of the supply chain could
successfully operate their facilities like the presented manufacturing model, it could result in up to 60% in manufacturing cost savings. In practice, a standard turbine design would likely operate at off-design conditions, resulting in lower efficiencies, less electricity generation, and less revenue than a custom turbine design. It is important to know whether and under what conditions the upfront capital cost savings from a standard turbine design could offset future revenue losses. The trends show that standard turbine designs could be competitive over a wide range of temperatures and flow rates. A calculation of the standard turbine efficiencies at off-design conditions that give the same NPV as a project using a custom turbine show that the range of off-design efficiencies supports this conclusion. Developing and using standard turbine designs may be an effective strategy for lowering geothermal power project costs if a pipeline of turbines can be set. Ideally, these turbines would be designed to be flexible and operate over a wide range of conditions with minimal efficiency decreases. The strategy requires that multiple turbines be built at once and then warehoused until sold. A significant barrier to implementing this strategy is the demand for these technologies at high volumes. However, as the global geothermal market continues to grow, opportunities in new markets will continue to increase. The emerging geothermal markets discussed in this chapter show that there may be an opportunity for using standardized turbines to reduce plant capital costs.
Global value chain and manufacturing analysis Chapter 2
Acknowledgments This work was supported by the US Department of Energy (DOE), the Office of Energy Efficiency and Renewable Energy (EERE), Geothermal Technologies Office (GTO) under Contract No. DE-AC3608-GO28308 with the National Renewable Energy Laboratory (NREL). The authors wish to thank reviewers for their comments and suggestions, including Doug Arent, Jill Engel-Cox, Margaret Mann, Emily Newes, and Samantha Reese of NREL. The authors also thank Billy Roberts of NREL for his help on mapping. All errors and omissions are the responsibility of the authors.
References [1] GEA. Annual U.S. & global geothermal power production report. Geothermal Energy Agency; 2016. [2] Think GeoEnergy. The Top 10 Geothermal Countries 2019—based on installed generation capacity (MWe). Think GeoEnergy - Geothermal Energy News 2020;(Jan. 27). http://www.thinkgeoenergy. com/the-top-10-geothermal-countries-2019-based-on-installed-gener ation-capacity-mwe/. [Accessed 28 January 2020]. [3] Bertani R. Geothermal power generation in the world 2010–2014 update report. Geothermics 2016;60:31–43. https://doi.org/10.1016/ j.geothermics.2015.11.003. [4] BNEF. Geothermal market outlook report. Bloomberg New Energy Finance; 2016. [5] DiPippo R. Geothermal power plants: Principles, applications, case studies and environmental impact. Butterworth-Heinemann; 2008. [6] Frost & Sullivan. Global gas and steam turbine markets: Conventional thermal power expansion driven by emerging markets and rising natural gas availability. Frost & Sullivan; 2014. M96C-14. [7] BNEF. Q2 2013 geothermal market outlook report. Bloomberg New Energy Finance; 2013. [8] BNEF. H2 2014 geothermal market outlook report. Bloomberg New Energy Finance; 2014. [9] BNEF. H1 2015 geothermal market outlook report. Bloomberg New Energy Finance; 2015. [10] GEA. Annual U.S. & global geothermal power production report. Geothermal Energy Agency; 2015. [11] Tartiere T. World overview of the organic Rankine cycle technology, https://orc-world-map.org/index.html; 2016. [Accessed 22 January 2020]. [12] DOE. GeoVision: Harnessing the heat beneath our feet. Washington, DC: U.S. Department of Energy; 2019. DOE/EE-1306. [13] Wall A, Young K. Doubling geothermal generation capacity by 2020: A strategic analysis. Golden, CO: National Renewable Energy Lab (NREL); September 2018. NREL/TP-6A20-64925. Available from: https://www.nrel.gov/docs/fy16osti/64925.pdf. [14] Poernomo A, Satar S, Effendi P, Kusuma A, Azimudin T, Sudarwo S. An overview of Indonesia geothermal development—Current status and its challenges; 2015. p. 11. [15] Enerji Atlasi. Country update for geothermal power plants in Turkey. Enerji Atlası; 2019. https://www.enerjiatlasi.com/jeotermal/. [Accessed 22 January 2020].
41
[16] MENR. Turkish Ministry of Energy and Natural Resources Renewable Energy Law 2010; 2010. [17] Ellis PFI, Conover MF. Materials selection guidelines for geothermal energy utilization systems. Austin, TX: Radian Corp.; January 1981. https://doi.org/10.2172/6664808. DOE/RA/27026-1. [18] Kaya T, Hoshan P. Corrosion and material selection for geothermal systems. In: Proceedings world geothermal congress 2005; 2005. p. 5. [19] Steel Dynamics. 2015 Annual report. Steel Dynamics Inc.; 2015. Available from: https://s3.amazonaws.com/b2icontent.irpass.cc/ 2197/165986.pdf?AWSAccessKeyId¼1Y51NDPSZK99KT3F8VG2& Expires¼1580257699&Signature¼9zZjMvMMB3ob07MYrZq%2Fb CQdvl4%3D. [Accessed 28 January 2020]. [20] Yadav M, Misra A, Malhotra A, Kumar N. Design and analysis of a high-pressure turbine blade in a jet engine using advanced materials. Mater Today Proc 2019. https://doi.org/10.1016/j.matpr.2019.07.530. [21] USGS. Minerals yearbook—Metals and minerals. United States Geological Survey; 2015. Available from: https://www.usgs.gov/centers/ nmic/minerals-yearbook-metals-and-minerals. [Accessed 22 January 2020]. [22] Klocke F, et al. Turbomachinery component manufacture by application of electrochemical, electro-physical and photonic processes. CIRP Ann 2014;63(2):703–26. https://doi.org/10.1016/ j.cirp.2014.05.004. [23] Horowitz KAW, Woodhouse M, Lee H, Smestad GP. A bottom-up cost analysis of a high concentration PV module. AIP Conf Proc 2015;1679(1):100001. https://doi.org/10.1063/1.4931548. [24] Goodrich A, et al. A wafer-based monocrystalline silicon photovoltaics road map: utilizing known technology improvement opportunities for further reductions in manufacturing costs. Sol Energy Mater Sol Cells 2013;114:110–35. https://doi.org/10.1016/j. solmat.2013.01.030. [25] Dewhurst B. Inc. DFA: Product simplification and DFM: Concurrent costing, http://www.dfma.com/software/index.html; 2016. [Accessed 17 May 2017]. [26] Ross SA, Westerfield R, Jordan BD. Fundamentals of corporate finance. New York, NY: McGraw-Hill, Irwin; 2009. [27] SimTech. IPSEpro process simulation environment. In: IPSEpro User Guide; 2014. [28] GETEM. Geothermal electricity technology evaluation model. Energy.gov; 2016. https://www.energy.gov/eere/geothermal/ geothermal-electricity-technology-evaluation-model. [Accessed 22 January 2020]. [29] Mendelsohn M, Kreycik C, Bird L, Schwabe P, Cory K. Impact of financial structure on the cost of solar energy. Golden, CO: National Renewable Energy Lab (NREL); March 2012. NREL/TP-6A2053086. Available from: https://www.nrel.gov/docs/fy12osti/53086.pdf. [30] Short W, Packey DJ, Holt T. A manual for the economic evaluation of energy efficiency and renewable energy technologies. Golden, CO: National Renewable Energy Lab (NREL); March 1995. NREL/TP462-5173. Available from: https://www.nrel.gov/docs/legosti/old/ 5173.pdf.
Chapter 3
CO2 emissions from geothermal power plants and state-of-the-art technical solutions for CO2 reinjection Joseph Bonafin and Arianna Bonzanini Turboden S.p.A., Brescia, Italy
3.1
Introduction
Among several chemical components, geothermal fluids commonly contain different gases that may be found in the liquid phase dissolved in the geothermal brine or in the gaseous phase mixed with geothermal steam. The natural presence of these gases affects the physical and thermodynamic properties of the geothermal fluids, mainly of the geothermal steam. The knowledge of the gas concentration is a crucial parameter in the design of geothermal power plants because it affects the energy conversion efficiency of the power plant. The contained gases are usually carbon dioxide (CO2), hydrogen (H2), methane (CH4), hydrogen sulfide (H2S), and ammonia (NH3), but also small fractions of other gases can be found at particular sites. These gases are referred to as noncondensable gases (NCG) because they do not condensate at the same condition as water vapor, but remain in the gaseous phase. CO2 is the most abundant gas, typically more than 95%, so the exploitation of geothermal resources may lead to greenhouse gas (GHG) emissions. This is because NCGs are often released in the atmosphere after passing through the power plant together with the geothermal fluid. Even if methane is generally present in smaller concentrations, usually a few percentage points compared to CO2, it is a strong climate-changing gas and may give a nonnegligible contribution to the GHG emissions of a power plant. However, because the most common gas is CO2, the focus will be mainly on this component. The chemical composition of the geothermal fluid is dictated mainly by the composition of the host rocks of the geothermal reservoir. These are the rocks in contact with the water infiltrating the Earth’s crust, which dissolves the carbonates composing the rocks. One of the parameters that influences the CO2 concentration of the geothermal fluid is the type of host rock. If it contains a large fraction of carbonate, as in the case of carbonate rocks, the CO2
concentration will be higher. In igneous rocks, the carbonate fraction is small, and so is the CO2 concentration in the geothermal fluid. Sedimentary rocks may contain different percentages of carbonates, and thus the CO2 concentration may vary strongly. In some cases, CO2 and other gases are derived from magma bodies, which may be located under the geothermal reservoir; these magmatic intrusions can release gases during the natural degassing process. The CO2 dissolved in the brine can precipitate in carbonate or silicate minerals, and the CO2 mixed in the steam can be vented through fumaroles. These two processes reduce the CO2 concentration in the geothermal fluids and can be defined as CO2 sinks. The balance of the sources and sinks of CO2 determines the chemical composition of the geothermal fluids, which may vary in time due to the different contributions of sources and sinks in the chemical reactions. The GHG emission is generally smaller if compared to traditional power plants, such as oil, gas, or coal-fired plants. However, the geothermal sector has expanded in recent years, and geothermal reservoirs with high NCG content have been brought into use. The attention paid to GHG emissions has also put geothermal power plants in the discussion to evaluate the environmental footprint of the kWh produced by means of geothermal sources. Especially, banks and financial institutions such as the World Bank are interested in the assessment of geothermal power plant CO2 emissions to evaluate whether the investment can be considered environmentally friendly and if the investment can benefit from government incentives for green energy. To demonstrate the interest on the topic, the World Bank encouraged, in the ESMAP Technical Report [1], estimating ex ante the CO2 emissions of geothermal projects under development as well as geothermal power plants that already exist. Meanwhile, an ongoing debate has been opened to clarify whether geothermal energy can be classified as
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00012-3 Copyright © 2021 Elsevier Inc. All rights reserved.
43
44
PART
I Basics of geothermal power plants
“green” because it may be linked to GHG emissions. It is not generally clear if and to what extent the exploitation of geothermal reservoirs to produce electricity affects the natural emission of CO2. In fact, in some geothermal areas, CO2 is emitted naturally and continuously from the soil, steam vents, fumaroles, and in general from natural geothermal surface activities. However, in other geothermal fields, the CO2 stored in the reservoir would not be released without the installation of the power plant, which may vent to the atmosphere the mined geothermal steam together with CO2. In some geothermal fields, the presence of the geothermal power plant may enhance or even reduce the CO2 emissions of the site. In the end, the CO2 emissions from geothermal energy production should be evaluated on a case-by-case basis to assess the real quantity of CO2 emitted per kWh produced. Due to this uncertainty, carbon emissions from geothermal power plants are treated differently by national regulation from country to country. In some countries, CO2 emissions from geothermal power plants are not considered anthropogenic while in other nations, the CO2 emissions must be monitored and controlled. CO2 emissions from a geothermal power plant may also vary in time because the geothermal field exploitation may affect the natural balance between sources and sinks of CO2, often in an unpredictable way. In a few cases, the gradual decline of carbon emissions has been observed, namely by means of two phenomena: the reinjection of geothermal fluid depauperate of noncondensable gases and the formation of steam caps. In the first case, the reinjection of geothermal brine or steam condensate with low gas concentration dilutes the geothermal fluid in the reservoir, thus reducing in time the subsequent CO2 emissions. The steam cap formation is a result of the massive extraction of fluids, which leads to a pressure drop in the reservoir that increases the boiling phenomenon. In this way, the vapor fraction of the reservoir increases and the noncondensable gases migrate from the liquid phase to the vapor phase in the shallow region, resulting in a reduction of CO2 concentration in the brine and the steam located in the deeper region of the reservoir. Thus, if the steam is mined from a deep reservoir, the plant’s CO2 emissions decrease with time. Finally, the formation of shallow steam caps may lead to an increase of surface activities such as the formation of new fumaroles, which in the end releases CO2 into the atmosphere. The CO2 emission factor of a geothermal power plant is defined as the mass quantity of CO2 released in the atmosphere per kWh produced. Bertani and Thain [2] estimated that the average global CO2 emission factor is 122 g/kWh; this value is lower in Iceland and in California, where the averages are 34 g/kWh and 107 g/kWh, respectively. In Italy, the emission factor is generally rather high, ranging from 100 to 950 g/kWh, and in Turkey, it can reach 1300 g/kWh, for example, in the Gediz graben.
The difference in these emission factors is based on the chemical composition of the host rocks of the geothermal system. Italy and Turkey have carbonate-bearing rocks while Iceland’s geothermal reservoirs are instead mainly composed of igneous rocks that contain less carbonate and thus release less CO2. The emission factors of geothermal power plants in Italy or Turkey can be higher than the emission factor of a coalfired power plant, which can range between 750 and 1050 g/ kWh; see Ref. [3]. Thus, the main objective of NCG reinjection is the reduction of the environmental footprint of the geothermal power plant. Furthermore, it has been observed that NCG reinjection may have positive effects on the reservoir; see Ref. [4, 5]. NCG reinjection sustains the reservoir pressure, enhances well productivity, and reduces pH, thus inhibiting silica scaling. Moreover, credit institutes and banks are more willing to finance projects with low GHG emissions; thus, reinjection makes financing the construction of the power plant easier. The energy conversion technologies available for geothermal power production are mainly back pressure plants, condensing plants, two-phase flashing binary plants, and single-phase pumped binary plants. The energy conversion technology strongly influences the method employed to reinject the NGC into the reservoir. The choice of energy conversion technology depends on several factors, such as the resource temperature, pressure, flow rate, and chemical composition; see Table 3.1. The backpressure, condensing, and two-phase binary processes use steam, which rises from the production well because the pressure in the reservoir is higher than the pressure at the production well. The single-phase binary uses liquid only, and often submersible pumps are required to extract the geothermal fluid. In a back pressure power plant, the steam and the NCG expand in the turbine after the flash in a separator, and the exhaust gases are released in the ambient at atmospheric pressure. This kind of plant is rarely employed (only 1% of the global geothermal power capacity; see Ref. [6]).
TABLE 3.1 Energy conversion technology depending on the resource temperature. Plant type
200° C
Back pressure
✕
Flash condensing
✕
Two-phase binary
✕
✕
Single-phase (pumped) binary
✕
✕
✕
CO2 emissions from geothermal power plants Chapter 3
In a flash condensing plant, the expanded steam and NCG are condensed in a cooling system to increase the expansion ratio through the turbine; roughly 84% of the global installed capacity is based on this method [6]. The two-phase and single-phase binary power plants use the geothermal fluid to heat the working fluid, which is then expanded in a turbine. In the two-phase power plant, the working fluid is heated by geothermal fluid both in the steam and liquid phases while in the single-phase power plant, only geothermal water in the liquid phase is present. Binary plants are 15% of the total installed capacity [6]. The NCGs exit the heat exchangers at a pressure close to the inlet pressure of the heat exchangers, and the temperature is generally low (60–80°C); the NCGs from back pressure and condensing power plants exit the process at about 1 bar and 100°C. Hence, the NCGs from binary power plants are better suited for reinjection compared to NCGs from back pressure and condensing power plant; see Ref. [7]. In the case of single-phase binary power plants, the NCG remains dissolved in the brine and is often totally reinjected in the reservoir, thanks to the submersible pumps. However, these pumps are usually a critical component due to the high-temperature operating conditions. The higher the NCG content in geothermal water, the higher the design pressure to keep the NCG dissolved in the liquid phase. Hence, the pumps may consume a substantial fraction of the generated power. When the NCG content is high, an alternative solution to the use of submersible pumps should be found. At present, reinjection has not been practiced widely in geothermal power plants. Three main technological methods can be recognized, which are described in detail in Section 3.2: (1) Injection of gas-charged water, dissolving the NCG by means of an absorption column. (2) Injection of compressed NCG and mixing with geothermal fluid in the depth in the reinjection well. (3) Injection of the NCG together with geothermal brine by means of submersible pumps. The most suitable method for NCG reinjection is dictated by the energy conversion technology of the power plant and by the NCG quantity to be reinjected. After the description of the state-of-the-art reinjection methods, NCG reinjection will be analyzed mainly in the case of two-phase and single-phase binary ORC power plants (see Chapter 8 for a detailed description of these conversion technologies). Thus, reinjection methods 2 and 3 are detailed and compared in Section 3.3. In Section 3.4, the feasibility of the reinjection process is assessed as a function of the CO2 concentration and the limit on the reinjection temperature of the geothermal brine.
3.2
45
NCG reinjection successful cases
At present, NCG reinjection has been applied in few cases. Two effective experiences can be found at the Hellisheiði (Iceland) geothermal power plant in the framework of the CarbFix project, and at Umurlu geothermal reservoir (Turkey), where binary technology is employed. In the past, NCG reinjection was carried out at the Coso flash geothermal plant (California, United States) and the Puna geothermal field (Hawaii, United States). Experimental activities to evaluate the possibility of CO2 sequestration in mineral calcite were performed at the Ogachi Hot Dry Rock (HDR) geothermal site and the Hijiori HDR system. Several binary power plants in Germany reinjected NCG mainly by using submersible pumps. The CarbFix project started in 2007 as an industrial and academic research project funded by the European Union (EU). Its overall objective consists of the development of a cost-effective technology for the mineral storage of CO2 in basaltic rock. The pilot reinjection started in 2012, and in 2014 the first industrial-scale injection took place, as reported in Sigfu´sson et al. [8]. In 2017, about 10,000 tonnes of CO2 and 5000 tonnes of H2S were injected in the geothermal reservoir, corresponding respectively to 34% and 68% of the annual emissions of the geothermal power plant. The project developed a technology able to inject CO2 in the young basaltic rock. Here, the CO2 reacts with the basaltic rock to form stable carbonate minerals, and this provides permanent storage of the CO2 and H2S. Gunnarsson et al. [9] describe the process. The NCGs are first dissolved into pure water in a scrubbing tower; the water (36 kg/s, 6 bar, 20°C) is sprayed at the top of the scrubbing tower while the exhaust gases (0.366 m3/s) are injected at the bottom of the tower. The tower is able to trap 56% of the CO2 in the water and 97% of the H2S of the NCG stream; the remaining gases are vented into the atmosphere. The liquid water-CO2 mixture is then pressurized to 9 bar before transport from the capture plant to the injection well. The gas-charged water is injected through a stainless steel pipe to a depth of 750 m, where it is mixed with effluent water (not gas-charged). At the Umurlu geothermal reservoir, the CO2 content in the geothermal water from the production well declined due to the injection of degassed brine in the injection well. The injection of CO2 was developed to restore the NCG content (about 2%) in the extracted geothermal fluid because the CO2 content affects the reservoir pressure significantly (see Ref. [10]). Y€ucetas¸ et al. [10] describe the technical feature of the dual string-dual phase reinjection system: it consists of a pipeline for cold water in the well casing plus an ad-hoc designed compressor. The cold brine and the compressed NCG pass through an injector, then they are pumped together in the injection well and mixed at a depth
46
PART
I Basics of geothermal power plants
of more than 700 m. The reported CO2 flow rate is 2.65 tonne/h, and the CO2 reinjection pressure is 52 bar. From these two principal experiences, two different methods for NCG reinjection can be recognized: (1) Injection of gas-charged water, dissolving the NCG by means of an absorption column. (2) Injection of compressed NCG and mixing with geothermal fluid in the depth in the reinjection well. The reinjection strategy depends on the NCG content of the geothermal fluid and on the technology employed to convert the geothermal energy into electricity. The first method can be applied in geothermal systems with low NCG content and applying a flash steam turbine, such as at the Hellisheiði power plant. However, the absorption column is not a viable solution in the case of geothermal fluid with high NCG content (more than 2%) because part of the NCG is vented into the atmosphere. This can be acceptable in the case of a geothermal fluid with low NCG content (so the total vented NCG is low); however, this practice is not acceptable in the case of high NCG content, and it prevents the achievement of the zero-emission goal. In the case of geothermal fields with high NCG content, such as at the Umurlu geothermal field, the second method was successfully applied. The Umurlu power plant is based on the binary conversion cycle and uses an Organic Rankine Cycle (ORC) to convert geothermal energy into electricity. Gunnarsson et al. [9] estimated that the cost of reinjection at the Hellisheiði geothermal plant is about 25 USD/tonne of gas mixture (CO2 + H2S), comprising capture, transport, injection, and monitoring of the NCG. This cost can be acceptable in the case of low NCG content in the geothermal fluid, but it would be too high in the case of geothermal fluid with a high content of NCG, which is, for example, a common situation in several sites in Tuscany (Italy). Here, the NCG content in the steam may vary from 6% to 10%, with a corresponding flow rate of about 1–2 kg/ s. Considering a cost of 25 USD/tonne, the cost to reinject all the NCG would range between $790,000–$1,580,000 per year. Such a reinjection cost is quite substantial, considering that this power plant may produce 40 GWh/year, meaning that the cost per kWh ranges between 0.0198 and 0.0395 USD/kWh. This cost can be reduced by using the technology proposed at the Umurlu geothermal reservoir. If we consider the total cost of the reinjection plant at about $3 million (comprising the compressor, transport pipeline, injection, and control), a 30-year lifetime of the plant, and a flow rate of 1.5 kg/s, the cost is about 2.1 USD/tonne of gas mixture. This technology seems to be more convenient from the economic point of view. It can also achieve the zero-emission target, and can be more easily applied to the binary ORC power plants if compared to a flash turbine plant. A third effective method employed to reinject the NCG was performed in several single-phase geothermal sites, in
which the NCGs are dissolved in the geothermal brine thanks to the pressure. In this case, the NCGs are kept dissolved in the liquid phase throughout the whole process by keeping the brine pressurized in the heat exchangers. These geothermal power plants are usually referred to as pumped geothermal systems because deep well submersible pumps are employed. Then, the NCGs are reinjected together with the cooled geothermal fluid through the reinjection wells. This method ensures the zero-emission goal and has been performed widely at geothermal sites in Germany. However, this method cannot be used if steam is also present and when the NCG content is high (more than about 2%). In fact, in this case too high a pressure would be required to keep the CO2 dissolved in the liquid brine because the CO2 solubility increases as the pressure increases. The higher the reinjection pressure, the higher the cost, size, and maintenance costs of the pumps. The reinjection of NCG was performed at the Coso geothermal field for 12 years, and at the Puna geothermal field to carry out the silica scale control; see Ref. [11]. At Coso, the effect of NCG reinjection on brine pH reduction was demonstrated: the pH was lowered from 8 to 5, preventing silica scaling. However, the geothermal plant at Coso was based on a dual-flash condensing technology. This led to the passing through of residual oxygen to the production well, resulting in severe corrosion of the equipment. In the end, the NCG reinjection was replaced with conventional H2S abatement and pH-mod systems. At the Puna geothermal field, the low NCG content allowed for total gas injection from the start-up, and the resulting pH reduction was effective in the silica scaling mitigation at a saturation index (SI) of more than 2 [11]. However, the total gas injection was not technically implemented in the case of flash power plants because the NCGs are contaminated with oxygen. This issue does not occur in ORC technology, and thus silica scaling inhibition can be carried out without corrosion problems [11]. Experiments for CO2 injection were performed at the Hot Dry Rock (HDR) geothermal system; see Ref. [12, 13]. Here, the CO2 is dissolved in the river water injected to recover heat from the hot dry rock; the purpose is CO2 sequestration in carbonate minerals (calcite), as proposed at the Hellisheiði geothermal field. In the Japanese experiments, the effect of CO2 injection on calcite scaling was investigated, showing that the injection of CO2 increases the SSI of CaCO3, leading to the deposition of calcite.
3.2.1 Numerical simulations of NGC reinjection in the literature Several numerical simulations and preliminary studies for reinjection can be found in the literature. In this chapter, a few studies are cited to exemplify the approach and the goal of these numerical simulations. This literature review
CO2 emissions from geothermal power plants Chapter 3
is not exhaustive, and the scientific research on this topic remains in continuous development. The interest in this topic may also lie outside the frame of the geothermal power plant because the interest in the CO2 reinjection in geothermal fields, or in geological formations, or even in petroleum reservoirs, deals in general with CO2 sequestration and with the environmental concern regarding the CO2 concentration in the atmosphere. Batini et al. [14] analyzed a typical Italian geothermal field. Kaya and Zarrouk [15] simulated the simultaneous NCG and water reinjection by means of three-dimensional (3D) models. Kaya et al. [16] applied a 3D numerical model to simulate the CO2-water mixture reinjection at the Wairakei-Tauhara geothermal field. Manente et al. [17] proposed three different possible layouts for the reinjection plant of condensing power plants, comparing costs and performances. A total reinjection system for NCG is under study at the Castelnuovo pilot plant (Italy), which is based on ORC technology [14]. The reported numerical simulations show that a closed-loop power system is feasible, and a first evaluation of the reinjection depth and reinjection pressure was assessed. In this case, the compressed NCGs are mixed directly with the steam condensate at a certain depth in the injection well without a previous dissolution of NCG in pure water, as done at the Umurlu geothermal plant. However, the uncertainty on these results is still high; the cost of the injection plant was not estimated; the performance reduction of the power plant, due to compressor own consumption, is still to be defined; and the borehole gas/liquid system should be carefully dimensioned to achieve the correct NCG mixing with the condensate. Possible configurations of flash condensing geothermal power plants with NCG reinjection are proposed in Manente et al. [17], and the costs and performances of three different abatement systems were compared. The proposed layouts comprise an absorption column, similar to the one described by Gunnarsson et al. [9] implemented at the Hellisheiði power plant, but differ for the cooling technology. Solutions with a dry cooling tower, an air condenser (AC), and a hybrid dry cooling tower—AC system were investigated. This is an interesting study for flash technology, but it cannot be applied in the case of binary plants.
3.3 Evaluation of the reinjection process As reported in Section 3.2, three different main strategies may be recognized to carry out NCG reinjection, namely: (i) the injection of gas-charged water, dissolving the NCG by means of an absorption column; (ii) the injection of the compressed NCG, which are mixed with the geothermal fluid in depth in the reinjection well; (iii) the reinjection by
47
means of submersible pumps. The second solution seems to be more suitable than the first in the case of binary power plants, and it is able to achieve the zero-emission goal. The main component of such a technical solution is the compressor. Compressors are available in different types, sizes, and models, each of them suitable for a particular need. The choice of compressor type and compressor design depends mainly on the inlet volumetric flow rate to be compressed on the discharge pressure and the inlet pressure. The NCG stream may contain some percentage of H2S. Particular attention should be paid to the choice of materials for the compressor part in contact with the gases, due to the high corrosion potential of hydrogen sulfide. A comparison of different materials with high corrosion resistance should be carried out in order to assess the best trade-off in terms of performances and costs, depending on the chemical composition of the NCG mixture and of the geothermal fluids.
3.3.1 NCG reinjection process in case of high NCG content in geothermal steam We report here a preliminary dimensioning of the compressor in the case of high NCG content in a steamdominated geothermal field. This kind of geothermal reservoir represents a typical situation of the geothermal field present in Tuscany, Italy. We assume that a steam + NCG stream coming from the artesian well has the following characteristics: l l l l
Mass flow rate: 65 tonne/h. Temperature: 180°C. Pressure: 10 bar(a). NCG content: 8% in weight.
After passing through the binary plant heat exchangers, the condensed steam is separated from the NCG at about 90°C. We assume that the temperature and pressure of the NCG stream are, respectively, 50°C and 9 bar; 50°C is achieved by means of a dedicated cooler to reduce the compression work. In these conditions, the NCG volumetric flow rate is 350 m3/h (ACMH). Considering a reinjection depth of the compressed NCG of 400 m and considering the CO2 solubility in water, we assume that a discharge pressure of the compressor of about 60 bar(a) would be enough to ensure that the NCG mixes with the condensed steam in the reinjection well. In this case, the compressor may be a centrifugal compressor or a reciprocating compressor. Considering a two-stage compressor, it has been computed by means of Aspen Plus that the power required to compress the gas is about 270 kW, with the assumption of an intercooler temperature of 90°C and a compressor isentropic efficiency of 0.8. The flowsheet of the simulation is shown in Fig. 3.1 (Table 3.2).
48
PART
I Basics of geothermal power plants
FIG. 3.1 Aspen Plus flowsheet of the simulation model of the NCG compression system. Streams labeled with NCG1 represent the NCG coming from the heat exchangers.
TABLE 3.2 Description of the streams reported in the Aspen Plus flowsheet shown in Fig. 3.1. Stream label
Phase
Description
NCG1
Two-phase -water and CO2 mixture
The mixture of water steam and CO2 coming from the geothermal power plant enters the separator, where the gaseous phase (stream NCG2) is separated from the liquid phase (condensate, stream COND1)
NCG2 ! NCG3
Gaseous phase
The separated CO2 passes through a cooler to reduce the compression work
NCG3
Two-phase
The cooled two-phase flow passes through a second separator to remove the liquid phase (if present) before the first compression stage
NCG4 ! NCG5
Gaseous phase
First compression stage
NCG5 ! NCG6
Gaseous phase
Intercooling
NCG6 ! NCG7
Gaseous phase
Second compression stage
NCG7
Gaseous phase
Compressor discharge
COND3 ! COND4
Liquid phase
The condensate collected from the separators is mixed and pumped to be eventually reinjected in the reservoir
CO2 emissions from geothermal power plants Chapter 3
3.3.2 NCG reinjection process in case of moderate NCG content in a single-phase geothermal fluid The reinjection solution with the submersible pump can be used to achieve the total reinjection in the presence of a liquid-dominated geothermal reservoir with high NCG content (2% weight). In the case of higher NCG content, the pressure required to keep the CO2 dissolved in the geothermal fluid (liquid brine) would be too high (more than 50 bar) in the whole path from production to the reinjection well. This leads to very expensive submersible pumps, high own consumptions, and high design pressure of the equipment (heat exchangers, pipelines, etc.). Instead of using such pumps, it would be possible to pressurize the geothermal fluid at the wellhead at a lower pressure (7 bar), let the NCGs escape from the liquid brine to the steam phase, collect them, send them to the compressor, and reinject the NCG in depth, where they mix with the liquid brine reinjected by means of reinjection pumps. Such an alternative configuration may lead to cost and own consumption reduction depending on the Injectivity Index (I.I.) of the brine reinjection well, which is defined as the injection rate divided by the injection pressure (tonne/h/bar). If this latter is high, and thus the required pressure of the reinjection pump is low, the solution with the compressor presents some advantages if compared to the use of submersible pumps. To better explain this point, we report a numerical example: we assume a liquid brine with the following features l l l
Volumetric flow rate: 400 L/s. Temperature: 135°C. NCG content: 2%.
Table 3.3 reports the required equipment in the two configurations. The reservoir characteristics at the production well are assumed to be the same. The reinjection solution with the NCG compressor likely requires the use of injection pumps, which are not needed in Configuration 1, when the discharge pressure of the submersible pump is enough to ensure the reinjection pressure at the reinjection wells. In Configuration 2, the NCGs are first separated from the brine, and their temperature is reduced to 50°C to reduce the compression work. The volumetric flow rate of NCG in these conditions is about 1600 m3/h (ACMH), and we consider again a compressor discharge pressure of 60 bar. The intercooler temperature of the two-stage compressor is assumed to be 100°C. The use of the compressor implies a saving on the consumption of the submersible pump DẆ SP, due to the reduction of the discharge pressure (assuming that the reservoir pressure, the productivity index of the well, and the installation depth of the submersible pump are the same
49
TABLE 3.3 Comparison of the equipment required in Configuration 1, without compressor, and in Configuration 2, with compressor and reinjection pumps. Reservoir conditions: flow rate 400 L/s, temperature 135°C, NCG content 2% Configuration 1
Submersible pump: discharge pressure 53 bar No compressor No reinjection pumpa
Configuration 2
Submersible pump: discharge pressure 7 bar NCG compressor: discharge pressure 60 bar Reinjection pump: discharge pressure depends on I.I.
a Supposing that the submersible pump discharge pressure is enough to ensure reinjection.
in the two scenarios). However, the consumption of the compressor Ẇ C and the reinjection pumps Ẇ RP should be taken into account. DẆ SP has been computed as the power saved due to the reduction of wellhead pressure from 53 bar (pressure required to keep the CO2 dissolved in the brine, given the CO2 concentration and brine temperature) to 7 bar. This saving is considered constant because it depends on the variation of the submersible pump discharge pressure only and does not depend on the condition in the production well. Table 3.4 reports the difference between the sum of the consumption of the compressor and the reinjection pump (Ẇ C + Ẇ RP) and the own consumption of the submersible pump DẆ SP as a function of the Injectivity Index of the reinjection well. The difference of the own consumption DẆ DẆ ¼ ðẆ C + Ẇ RP Þ DẆ SP
(3.1)
TABLE 3.4 Difference on the own consumption as a function of the Injectivity Index of the reinjection well. I.I. (tonne/ h/bar)
DẆ SP (kW)
Ẇ C + Ẇ RP (kW)
DẆ / Ẇ gross (%)
20
2453
3570
10.2
30
2690
1.8
40
2250
2.4
50
1986
4.9
60
1810
6.6
70
1684
7.8
80
1590
8.7
90
1517
9.4
50
PART
I Basics of geothermal power plants
has been normalized on the gross power output Ẇ gross of the power plant (considering a 14% gross efficiency). It is possible to see that as the I.I. increases, the extra own consumption decreases and turns negative due to the reduction of the required reinjection pressure. Thus, the use of the compressor has a positive effect on the performances of the power plant in case of a high Injectivity Index. As well as the cost reduction of the submersible pump, keeping a low wellhead pressure leads to a further cost reduction thanks to the lower design pressure of the heat exchangers and of the pipelines transporting the geothermal fluid.
3.4
Feasibility of the reinjection process
The feasibility of the reinjection process should be analyzed case by case, depending on the characteristics of the specific geothermal site. Several parameters may affect the feasibility of the reinjection process because the latter is linked to the CO2 solubility in water: the most important parameters are the reservoir pressure, the CO2 concentration, and the geothermal fluid reinjection temperature. If the CO2 concentration is high, the reinjection pressure of the cooled geothermal fluid may be higher than the pressure of the geothermal reservoir due to the behavior of the CO2 solubility as a function of the temperature. We report here a numerical example to explain this point. We assume that the geothermal fluid exits the steam and brine separator at 185°C, and we consider a liquid brine reinjection limit of 90°C to avoid silica scaling.a For the sake of simplicity, the salinity of the brine has not been taken into account in this example. We analyze the reinjection pressure required to dissolve the CO2 in the liquid brine for different CO2 concentrations using the CO2 solubility model proposed by Duan and Sun [18] (Fig. 3.2): l
FIG. 3.2 CO2 solubility in water at different temperatures as a function of pressure. The solubility of CO2 decreases between 20°C and 90°C in all pressure ranges while the solubility curve at 185°C crosses both the solubility curves at lower temperatures.
1% weight CO2 concentration: at 185°C, the CO2 is totally dissolved in the brine at about 40 bar(a) while at 90°C and at 40 bar(a), the concentration of CO2 in the brine can be 2% in weight; see Fig. 3.3 (dotted line). Hence, in this case, the CO2 can be reinjected totally in the reservoir.
a. Silica scaling consists of the precipitation of silicates due to the temperature reduction of the geothermal fluid; the solubility of the silica generally increases as the solution temperature increases. Thus, as the geothermal fluid is cooled in the heat exchanger of the geothermal plant, the solubility of the silica decreases, and the silica precipitate in solid compounds. This process may lead to the clogging of the geothermal fluid piping and of the heat exchanger tubes, reducing the geothermal fluid flow to the power plant and reducing the heat exchangers’ performances. This issue is often avoided by limiting the geothermal fluid reinjection temperature to keep the saturation index of silica higher than 1.
FIG. 3.3 CO2 solubility in water at 40 bar(a), 170 bar(a), and 410 bar(a) as a function of temperature. l
5% weight CO2 concentration: at 185°C, the CO2 is totally dissolved in the brine at about 170 bar(a) while at 90°C and at 170 bar(a), the concentration of CO2 in the brine can be about 4.7% in weight; see Fig. 3.3 (dashed line). Hence, in this case, only part of the CO2 can be reinjected in the reservoir.
CO2 emissions from geothermal power plants Chapter 3
l
9% weight CO2 concentration: at 185°C, the CO2 is totally dissolved in the brine at about 410 bar(a) while at 90°C and at 410 bar(a), the concentration of CO2 in the brine can be about 6.2% in weight; see Fig. 3.3 (solid line). Hence, in this case, only part of the CO2 can be reinjected in the reservoir.
In the second and third cases, the pressure required to totally reinject the CO2 would be higher than the pressure reservoir, and this is not feasible due to a physical limit. If we observe Fig. 3.2, we see that the CO2 concentration limit for the total reinjection is about 4.2% (where the dotted and the solid lines cross), if the brine temperature is reduced from 185° C to 90°C. Moreover, we can observe that the limit on the reinjection temperature of the brine plays a crucial role in the feasibility of the reinjection. If the geothermal brine could be cooled down to 20°C, the total reinjection could be feasible, from a physical point of view, up to a CO2 concentration of 7.9% (cross of the dotted line and the dashed line).
3.5
51
Some technical aspects of the compressor, the main component of the reinjection system, have been shown. A numerical example representative of a typical situation of the Italian geothermal field has been reported. In the case of single-phase binary power plants, the comparison of two possible configurations to perform the total reinjection has been presented, showing the advantages of the reinjection with NCG compressor depending on the Injectivity Index of the reinjection well. The choice of one configuration depends on several parameters, such as the characteristics of the geothermal reservoir and of the production and reinjection well, on the NCG content. Thus, a best practice guideline for the reinjection of NCG shall be developed, taking into account the possible different features of the single case. Finally, the feasibility of the reinjection process on the basis of the CO2 concentration has been analyzed as a function of the reinjection temperature and pressure, showing how these parameters influence the possibility of the total reinjection in the reservoir.
Closing remarks
State-of-the-art NCG reinjection methods have been outlined. Three main successful cases have been presented, namely the reinjection experience at the Hellisheiði power plant, at the Umurlu geothermal field, and the reinjection by means of submersible pumps. Three different reinjection methods can be identified. NCG injection can be carried out by means of an absorption column and the subsequent mixing of water and NCG above the ground, or by means of the NCG compression alone and the mixing with geothermal fluid at great depth in the reinjection well. If compared to the method involving the absorption column, this latter is more suitable to binary power plants because the NCG can be easily kept separated from the geothermal steam and brine; moreover, this method can achieve the zero-emission goal. Besides these two experiences, the total reinjection of CO2 can be achieved by means of deep submersible pumps in the case of single-phase geothermal systems. The numerical simulations reported in Batini et al. [14] and Manente et al. [17] have been analyzed, showing the feasibility of NCG reinjection in the case of high NCG content and some possible reinjection plant layouts in the case of flash power plants, respectively. Other numerical analyses of the NCG reinjection, which have not been analyzed in detail for the sake of brevity, show interest in the possibility of CO2 sequestration. The NCG reinjection can be implemented more straightforwardly in the case of binary power plants if compared to the flash and back pressure power plants because the NCG stream can be kept separated from the working fluid and it is not contaminated by air.
References [1] ESMAP [Energy Sector Management Assistance Program]. Greenhouse gases from geothermal power production. ESMAP Technical Report 009/16, Washington, DC: World Bank; 2012. [2] Bertani R, Thain I. Geothermal power generating plant CO2 emission survey. IGA News 2002;49:1–3. [3] Anon. World Bank guidance manual: greenhouse gas accounting for energy investment operations. Transmission and distribution projects, power generation projects, and energy-efficiency projects (ver. 2.0). Washington, DC: The World Bank; 2015. [4] Stefa´nsson V. Geothermal reinjection experience. Geothermics 1997;46:99–139. [5] Kaya E, Zarrouk SJ, O’Sullivan MJ. Reinjection in geothermal fields: a review of worldwide experience. Renew Sust Energ Rev 2011;15:47–68. [6] GEA [Geothermal Energy Association]. 2015 annual U.S. & global geothermal power production report, http://geo-energy.org/reports/ 2015/2015%20Annual%20US%20%20Global%20Geothermal% 20Power%20Production%20Report%20Draft%20final.pdf; 2015. [7] Verkı´s Consulting Engineers. Geothermal power plants–CO2 emissions. Verkı´s report; 2015. 14298-001-1. [8] Sigfu´sson B, Arnarson M, Snæbj€ornsdo´ttir SO, Karlsdo´ttir MR, Arado´ttir ES, Gunnarsson I. Reducing emissions of carbon dioxide and hydrogen sulphide at Hellisheiði power plant in 2014-2017 and the role of CarbFix in achieving the 2040 Iceland climate goals. Energy Procedia 2018;146:135–45. [9] Gunnarsson I, Arado´ttir ES, Oelkers EH, Clark DE, Arnarson M, Sigfu´sson B, Snæbj€ornsdo´ttir SO, Matter JM, Stute M, Ju´lı´usson BM, Gı´slason SR. The rapid and cost-effective capture and subsurface mineral storage of carbon and sulfur at the CarbFix2 site. Int J Greenhouse Gas Control 2018;79:117–26. _ Ergic¸ay N, Akın S. Carbon dioxide injection field pilot in [10] Y€ucetas¸ I, Umurlu geothermal field, Turkey. GRC Trans 2018;42.
52
PART
I Basics of geothermal power plants
[11] DiPippo R. Geothermal power generation: developments and innovation. Woodhead Publishing; 2016, ISBN:978-0-08-100337-4. [12] Kaieda H, Ueda A, Kubota K, Wakahama H, Mito S, Sugiyama K, Ozawa A, Kuroda Y, Sato H, Yajima T, Kato K, Ito H, Ohsumi T, Kaji Y, Tokumaru T. Field experiments for studying on CO2 sequestration in solid minerals at the Ogachi HDR geothermal site, Japan. In: Thirty-fourth workshop on geothermal reservoir engineering, Stanford University, Stanford, California, February 9-11; 2009. [13] Yanagisawa N. Ca and CO2 transport and scaling in the Hijiori HDR system, Japan. In: Proceedings world geothermal congress 2010, Bali, Indonesia, 25-29 April; 2010. [14] Batini F, Lisi S, Guglielmetti L, Bellini F, Trinciarelli V, Pucci M. Well engineering and simulation for non-condensable gases total
[15] [16]
[17]
[18]
reinjection systems. In: European geothermal congress 2016, Strasbourg, France, 19–24 September; 2016. Kaya E, Zarrouk SJ. Reinjection of greenhouse gases into geothermal reservoirs. Int J Greenhouse Gas Control 2017;67:111–29. Kaya E, Callos V, Mannington W. CO2-water mixture reinjection into two-phase liquid dominated geothermal reservoirs. Renew Energy 2018;126:652–67. Manente G, Lazzaretto A, Bardi A, Paci M. Geothermal power plant layouts with water absorption and reinjection of H2S and CO2 in fields with a high content of non-condensable gases. Geothermics 2019;78:70–84. Duan Z, Sun R. An improved model calculating CO2 solubility in pure water and aqueous NaCl solutions from 273 to 533 K and from 0 to 2000 bar. Chem Geol 2003;193:257–71.
Chapter 4
Life cycle assessment of geothermal power plants Lorenzo Tostia,b, Maria Laura Parisia,b,c, and Riccardo Basosia,b,c a
Center for Colloid and Surface Science, University of Firenze, Sesto Fiorentino, Italy b R2ES Lab, Department of Biotechnology, Chemistry and
Pharmacy, University of Siena, Siena, Italy c National Research Council—Institute for the Chemistry of OrganoMetallic Compounds, Sesto Fiorentino, Italy
4.1
Introduction
Geothermal power plants employ geothermal fluids extracted from convective hydrothermal reservoirs to produce energy (electricity, heat, or both). The conversion can occur through the direct use of steam or the indirect use of hot water. Generally, low-temperature reservoirs are in a single liquid phase, whereas high-temperature reservoirs can be liquid or vapor dominated. Geothermal fluids show different compositions and gas concentrations depending on the geological formation of the reservoir, geographical location, fluid temperature, and depth. Naturally occurring geothermal systems are known as hydrothermal and are characterized by the local availability of a resource fluid. Enhanced geothermal systems (EGSs) are geothermal systems where a low natural permeability reservoir does not permit industrial use. Thus, EGSs use stimulation techniques to develop suitable porosity/fracture patterns and increase fluid circulation to reach industrial operating conditions. Geothermal development in Europe can be traced back more than a century, but the market is still at an early stage. Recently, a growing interest in geothermal power production has been observed after a decade of only limited development in the capacity of the deep geothermal sector for both electricity and heat supply [1]. This revival momentum is connected with the recognition of the crucial role that geothermal energy could play in the European market and the world as well as future energy scenarios based on the replacement of fossil fuels with renewable sources. During the UN Climate Change Conference (COP21) in Paris in December 2015, this led to the launch of the Geothermal Global Alliance [2]. In this context, the European Strategic Energy Technology Plan (SET) has set ambitious implementation actions for the development of deep geothermal energy to reach the goals of placing Europe at the forefront of the energy decarbonization scene [3]. Nowadays, geothermal energy is considered one of the
most promising renewable energy sources for electricity as well as heating and cooling production that can contribute to sustainable development and a transition toward a lowcarbon economy [4–7]. Nevertheless, some barriers to its full development still exist. Among these, environmental concerns represent one of the main issues. Geothermal energy should be managed in order to make it a reliable, environmentally sustainable, and renewable energy source. However, we should remember that clean energy does not exist, and all the human-made activities somehow determine an impact on nature. This includes the construction and operation of every type of energy process, even based on renewable sources, so that also includes a geothermal plant. The ecological burden of all infrastructure projects should be carefully considered, and environmental regulations and assessment methods are important in particular for the development of geothermal systems as geothermal energy, although renewable, is not clean. Policy and regulation are of paramount importance, even for the development of renewable energies. It is necessary to understand the future impact to be able to come up with regulations or policies that would encourage those production technologies that show the lowest impact on the environment. The life cycle assessment (LCA) has been established as a methodology to quantify and account for potential environmental impacts. Originally focused on accounting for current impacts in existing products and processes, LCA has increasingly become a tool useful to assess the future impacts of a more consequential nature [8, 9]. LCA can, therefore, assist in formulating policies and making environmental-oriented decisions as well as play the role of a supporting tool when formulating environmental regulations. LCA has been widely used to assess the environmental impacts of power generation systems and, in particular,
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00017-2 Copyright © 2021 Elsevier Inc. All rights reserved.
53
54
PART
I Basics of Geothermal Power Plants
renewable energy production [10–16]. In 2003, the European Commission concluded that LCA provides the best framework to assess the potential environmental impacts of products, processes, and services as it covers several environmental aspects, thus approaching the system more comprehensively and holistically [17]. Notwithstanding the application of LCA for many years as the most suitable method to circumspectly detect the full range of ecological burdens with energy-producing systems, the field of geothermal energy exploitation still needs to be developed in terms of completeness and consistency of LCA studies and primary data unveiling. So far, only a few studies have focused on determining the complete life cycle environmental profile of currently operating geothermoelectric installations [18, 19]. The aim of this chapter is to provide a panorama of the environmental impacts of geothermal power plants in relation to their geomorphological characteristics and the variety of technologies used to exploit the resource.
4.2
LCA methodology
The LCA is a methodology that allows us to quantify and compare the potential environmental impacts associated with products and services. LCA is regulated by the ISO 14040–14044 series. The best practices on how to conduct an LCA study are reported in the ILCD handbook [20]. The mentioned LCA ISO standards are quite general because LCA methodology can be applied to a large variety of products and processes, thus allowing such ISO standards to be open to interpretation. The LCA methodology is divided into four phases, as shown and summarized in Fig. 4.1. The first phase is the goal and scope definition. This phase defines the functional unit, system boundaries, level of specificity of data through the definition of foreground and background data, exclusion of life cycle stages or inputs, and the selection of impact indicators and characterization factors. The functional unit (FU) is a quantitative measurement of the function of a system to be used as a reference unit to which all input and output flows are scaled. It allows for consistent comparisons among different geothermal systems and with respect to other electricity/energy-generating technologies that can provide the same function. The second phase is known as the Life Cycle Inventory (LCI) and consists of collecting all relevant information regarding mass and energy flows throughout the defined system. This phase is often the most time-consuming step of an LCA and also the most critical one for obtaining good results. The next phase is the Life Cycle Impact Assessment (LCIA), in which all the input and output flows collected during the LCI are translated into potential environmental impacts by multiplying their cumulative mass value by a
characterization factor (CF). The magnitude of potential impacts can be different depending on the impact assessment method selected for the LCIA step. Some impact assessment methods distinguish between midpoint or endpoint indicators depending on the point at which the impact is calculated throughout its cause-effect chain. Midpoint indicators focus on single environmental issues, for example, climate change or acidification. Endpoint indicators show the environmental impact on three higher aggregation levels, namely the (i) effect on human health, (ii) biodiversity, and (iii) resource scarcity. Moving from midpoint to endpoint indicators simplifies the interpretation of the LCIA results. However, with each further aggregation step, uncertainty in the results increases. The LCA methodology presents many advantages when analyzing complex systems. Its holistic approach allows avoiding the shifting of impacts from one phase of the life cycle to another. LCA can account simultaneously for different impact categories considering many aspects of the environment and human health. Furthermore, the identification of energy and environmental hot spots of a complex system is facilitated by the application of LCA methodology, making it a very useful tool to unravel the complexity that often lies behind energy-producing systems. Finally, it allows us to have a global vision of the product throughout its life cycle, including also impacts that are generally ignored or neglected (e.g., those related to final disposal). Even with the mentioned proven beneficial aspects of LCA, its application to energy systems is not simple. The methods employed for the assessment of environmental performance differ significantly in the selection of system boundaries, coproducts and waste definitions, criteria adopted for the allocation of environmental burdens, and the selection of impact assessment methods. This is a quite general observation for energy systems. In the specific case of the geothermal sector, the matter becomes even more complicated due to the various geological characteristics of geothermal sites. This aspect adds variance on top of the different technologies available to extract and convert deep geothermal energy into electricity and heat. Therefore, some guidelines on how to conduct an LCA of geothermal energy systems need to be established in the near future to obtain more homogeneous results, which will, in turn, allow a fair benefits/disadvantages comparison among different renewables. Some dedicated guidelines have already been established for other specific renewables (see, i.e., photovoltaic energy production).
4.3 Impacts of geothermal energy exploitation As with all anthropogenic activities, geothermal energy exploitation has different effects on the surrounding
Life cycle assessment of geothermal power plants Chapter 4
55
FIG. 4.1 Phases of life cycle assessment methodology.
environment, including humans. In general, these effects are divided into (see Fig. 4.2): l
l
l
l
l
Acoustic effects (i.e., noise pollution during drilling, construction, and operations). Thermal effects (i.e., thermal pollution, the release of vapor into the air, heating, and cooling of the soil in the extraction or reinjection of fluids). Visual/surface effects (i.e., land use, biodiversity disturbance). Physical effects (i.e., induced seismicity and landslides, microseismicity, soil subsidence, geological risk, depletion of groundwater resources, natural radioactivity). Chemical effects (i.e., emissions into the atmosphere, discharge of liquid and solid substances).
The LCA methodology deals only with some of the abovementioned impacts because it is a methodology based on mass and thermodynamic balance. The potential impacts that can be somehow measured by an LCA are those essentially related to the utilization/depletion of natural resources and direct and indirect emissions. Natural resource utilization does not necessarily relate only to the exploitation of the geothermal fluid, but also to the use of raw materials (i.e., iron, copper, water, etc.) that are embedded in products and the energy used during power plant life cycle stages. More recently, land and water use have also been included in the most common impact assessment methods. These impact categories are still subject to improvements because a unanimous scientific
consensus on how to calculate the impacts on these categories has not been reached yet. Direct emissions are mainly represented by the emitted noncondensable gases (NGSs) contained in geothermal fluid either as a free gas or in a dissolved form. NCGs occur naturally in geothermal fluids at different concentrations. Typical NCG concentrations vary from 0.2% to 25% of the steam, with CO2 being generally the most abundant gas followed by hydrogen sulphide (H2S), hydrogen (H2), nitrogen (N2), methane (CH4), ammonia (NH3), argon (Ar), and radon (Rn). Other sources of direct emissions are fugitive emissions of working fluids in the case of organic Rankine cycle (ORC) plants, and emissions from the combustion of diesel fuel to provide energy for the construction plant (including the drilling platform, pumping of muds, etc.) and for the operation of the plant and end-of-life activities (dismantling, well closure). Indirect emissions associated with materials manufacturing and energy production are required during the whole LCA of a geothermal plant. Gathering information related to these emissions is considered a crucial step when performing an LCA of geothermal plants.
4.4
Results and discussion
This section, even though it cannot be assumed as a complete and detailed analysis of the literature, gives a perspective of the main phases of an LCA in terms of the current state of the art in each phase, the main existing drawbacks, and possible future recommendations. Most of the
56
PART
I Basics of Geothermal Power Plants
FIG. 4.2 Graphical representation of the main environmental effects related to the life cycle of a geothermal power plant.
results shown here come from the currently ongoing European Coordination and Support Action project GEOENVI launched in 2018 [21] in the frame of the Horizon 2020 Work Package. In short, part of the GEOENVI project consists of collecting and categorizing the LCAs of geothermal projects/installations based on several factors such as the geographical location, specific goal and scope, types of studies, environmental aspects, technical criteria, functional units, system boundaries, and detailed LCIA methodology (see https://www.geoenvi.eu/ for further information).
4.4.1
Goal and scope definition
The majority of LCA studies on geothermal systems that are currently available in the literature follow an attributional approach [19, 22, 23]. The attributional approach consists of assessing the environmental impacts of energy production from an existing geothermal plant supplying a utility’s network. In some cases, the results of LCA analysis are compared to different geothermal configurations and technologies [22, 24–26] or different renewable energy production technologies [27–30]. The goal and scope of the available LCA studies on the geothermal sector focus on evaluating the impacts and drawing an environmental profile of currently operating power plants. Normally, the assessment is based on existing available data (i.e., environmental permissions, data from environmental monitoring agencies, etc.) or assumptions based on the literature.
The most adopted functional units are the kWh of electricity delivered to the grid (kWhe) or the kWh of heat delivered (kWhth) to a final user. The selection between the two functional units depends on the main function of the power plant. Often, the energy production is cogenerative in shape involving both electrical and heat products and an allocation procedure is therefore needed. However, no consensus has been reached yet on the method that should be mandatorily used to allocate impacts between electricity and heat (energy, exergy, or economic-based methods). Consequently, often no clear information is given on the allocation procedure adopted and whether the system requires it. Concerning the technology and electrical installed capacity, half of the recent studies available in the literature focus on projects with an installed capacity between 10 and 100 MWe and the other half on a small project with an installed capacity lower than 10 MWe. When we look at thermal capacity, most of the studies reported an installed thermal capacity lower than 10 MWth. LCA studies are mainly focused on small (e.g., frequently pilot) plants, where the primary technology adopted is a binary technology.
4.4.2
System boundaries
The life cycle phases that are normally included in an LCA study are (i) the commissioning phase, and (ii) the use phase (Fig. 4.3). The commissioning phase includes the
Life cycle assessment of geothermal power plants Chapter 4
57
FIG. 4.3 Typical boundaries of a geothermal system. The background system includes data on the production of energy and materials consumption, whereas the foreground system includes specific data on energy and materials consumed during commissioning, use, and decommissioning as well as direct emissions from the use phase.
construction work for the wells, wellheads, collection pipelines, power plant building, and all the necessary plant machinery/equipment items. The use phase, on the other hand, is generally divided into two subphases, the operational and maintenance phases. The main processes included in the use phase are those involved in the geothermal fluid exploitation, stimulation, equipment replacement, scaling prevention, drilling of additional wells, and direct emissions to air. Only a few works [27, 31, 32] take into account the endof-life activities, which include all procedures for correct closure of the wells and treatment of wastes produced during all previous phases. However, the end-of-life phase is often disregarded or poorly described at inventory and system description levels. Other studies focus exclusively on a single stage such as the operational stage [22, 33, 34]. In the case of the papers by Bravi and Basosi [33] and Parisi et al. [22], this methodological choice was based on the fact that the environmental profiles of the flash geothermal plants in object were determined essentially by the emissions to air during the operational phase while the other phases of the life cycle were considered irrelevant. Furthermore, the same authors aimed to create a statistical procedure that would allow reducing the variability of emission data measured and published by the national environmental agency.
4.4.3
Life Cycle Inventory
The outcome of an LCA is strictly dependent on the quality of data of the inventory upon which the model is built. Literature review works [23, 35] have underlined the common incompleteness and inaccuracy of LCIs used to perform the LCA of geothermal energy production. While it is necessary to have environmental impacts from renewable energy sources identified, the downside is often found in the LCA execution because primary data is often unavailable [36]. Therefore, in the majority of studies, the LCA practitioner does not make use of primary data but instead relies on available information in the literature and the assumptions or databases containing secondary data, which are
often not representative enough and probably inappropriate for specific geothermal power plants [34, 37]. Recently, particular attention has been dedicated to the evaluation of the environmental performances of EGS plants [32, 38–41]. However, at present, hydrothermal systems dominate current electricity generation in the geothermal sector, and the exploitation of this type of reservoir is predicted to be dominant in the near future, too [42]. The only current available sources of data for flash installations are at the moment those provided by Karlsdo´ttir et al. [43]. In this context, the GEOENVI project [21] has among its objectives to produce a harmonized LCI of the different geothermal installations, which are considered representative of diverse geographical and technological settings.
4.4.4
LCA of geothermal energy production
Based on our recent work conducted within the European project GEOENVI [21], the most adopted impact assessment methods are the CML [44, 45] and the ILCD [20]. Fig. 4.4 shows the most common impact categories included in the LCA of geothermal systems (i.e., on a total of 33 studies). Among impact categories, global warming potential (GWP) is by far the most discussed and the category for which most of the data can be found.
4.4.4.1 Global warming potential Fig. 4.5 reports GWP values (i.e., g CO2 eq./kWh) from different technologies adopted in the geothermal energy production sector as well as data from review works, which include different technologies. As reported already by Parisi et al. [22], Paulillo et al. [52], and Tomasini-Montenegro et al. [23], the majority of direct GHG emissions from dry steam and flash power plants are due to the operational activity. This is because the NCGs that are entrained with the geothermal fluid are released during the condensation step before the condensate is reinjected. The noncondensable fraction is typically composed of more than 90% of CO2, whereas the remaining fraction is shared between other gases such as hydrogen
58
PART
I Basics of Geothermal Power Plants
FIG. 4.4 Histogram of the frequency of inclusion of different impact categories in LCA studies of geothermal plants. Impact categories are not selected from a specific impact method.
sulphide, hydrogen, and methane. Because binary plants perform in a virtually closed loop, the GHG emissions are significantly lower when compared to flash and dry steam technologies and only assigned to the construction phase and pumping activities mainly. On the other hand, binary plants need dry tower systems that occupy more land and might result in a higher impact on land use when compared to dry steam and flash technologies. It must be highlighted that even among dry steam and flash technologies, the geological location and the types of rock present in the reservoir play important roles in determining the concentration of NCGs in the geothermal fluid and consequently, the GHG emissions. The authors report a significant difference in CO2 emissions (i.e., between around 40 and 800 g of CO2 eq./kWh) from flash power plants [22, 23, 33, 35]. These differences can be associated with the presence of carbonate-rich rocks in the reservoir, which increase the CO2 dissolved in the geothermal fluid and consequently lead to a higher CO2 emission during the operational phase. In this context, the debate on whether the geofluid dissolved CO2 should be counted as anthropogenic or as a natural emission is still open. Naturally, geothermal volcanic-based systems release greenhouse gases even without the exploitation of the geothermal resource for energy production. The
crucial point is if such anthropogenic activity accelerates this natural process. The literature available on the topic reports contrasting conclusions [53, 54]. Methane emissions generally occur at low concentrations (i.e., between 0.6 and 0.8 g/kWh) [35] compared to CO2. Often, they do not represent the focus of LCA studies and are disregarded. However, methane has a GHG potential 36 times higher than CO2 [55] and therefore should always be considered in an LCA study. Parisi and Basosi [56] highlighted the importance of considering the contribution of CH4 emissions to the GHG category in the Italian geothermal context. Besides direct emissions from the operational phase (when present), the construction phase also has a significant contribution to GHG. In particular, the production of steel and cement, which are employed during well construction and casing, together with the production and use of diesel for drilling wells, represents the processes with the highest impact on GHG but also other categories such as acidification potential and photochemical ozone formation. For instance, Paulillo et al. [19] reported a contribution of the construction phase process of more than 85% of the total impact in most of the impact categories (i.e., acidification, climate change, particulate matter, photochemical ozone
Life cycle assessment of geothermal power plants Chapter 4
59
FIG. 4.5 Emissions of CO2 eq./kWh from different geothermal energy production technologies by different authors. Blue squares (light gray in print version) report MIN values whereas red squares (dark gray in print version) are MAX values. In case no MIN and MAX values were available, a single value (green squares) (gray in print version) was provided instead.
formation, and the depletion of minerals, fossil fuels, and renewables). When binary systems are assessed, this phase is most likely to become the main contributor because direct emissions are essentially absent.
4.4.4.2 Acidification and eutrophication The direct emissions from the operational phase can contribute in part to acidification and eutrophication (i.e., mainly associated with NH3 emission). Among sulfurbearing gaseous emissions, H2S is by far the most dominant. Although not directly associated with the acidification effect, it is the subject of local environmental concern because of its odor and potential toxicity. However, when dissolved in water aerosol, H2S reacts with oxygen to form amore oxidized sulfur-bearing compounds such as SO2, thus indicating its secondary acidification potential. Fig. 4.6 shows the characterized impact on the acidification of different geothermal technologies. The Italian experience in this context has demonstrated that the abatement of mercury and hydrogen sulphide systems (AMIS) is very effective in the reduction of H2S (and Hg) [22]. In addition to this, in some peculiar fields where the NH3 emissions are most abundant, also a treatment system based on scrubbing with SO2 needs to be employed to reduce ammonia emissions.
Waste treatment processes, especially from drilling activities, might have a significant contribution to the eutrophication freshwater impact category. However, this type of process is often modeled using already existing processes from databases that might not be adequately representative of the geothermal sector. In some cases, for instance, the waste treatment and management processes included in databases account for part of the waste (i.e., muds, excavated soil, etc.) to be spread on soil as a fertilizing agent with a consequential high impact due to leaching of phosphorus (and other metals) into the soil. However, in most situations dealing with geothermal activities, using drilling wastes as a soil amendment is not allowed. Therefore, it should not be modeled by the general waste treatment process, but by a specific one that should be used instead.
4.4.4.3 (Eco)Toxicity If present in geothermal gases, some harmful elements such as mercury (Hg), arsenic (As), or antimony (Sb) may be stripped by gases emitted at plants, included in aerosol particles (drift) emitted from cooling towers in power production plants, and then be deposited on soil and washed out by rain with potential consequences on human and ecotoxicity impact categories. However, most of the inorganic
60
PART
I Basics of Geothermal Power Plants
FIG. 4.6 Emissions of SO2 eq./kWh from different geothermal energy production technologies by different authors. Blue squares (light gray in print version) report MIN values, whereas red squares (dark gray in print version) are MAX values. In case no MIN and MAX values were available, a single value (green squares) (gray in print version) was provided instead.
substances are currently missing in existing inventory databases, and methods are lacking in characterizing the potential exposure and toxicity impacts related to emissions of inorganic substances (i.e., H2S) except for some metals in their cationic form (e.g., mercury II). The results on the toxic category obtained from an LCA cannot model peculiar regional situations. Therefore, LCA practitioners should be very careful when communicating results to the public and, more importantly, they should never associate a potential toxic impact as calculated in LCA to a real environmental and human risk. In fact, the developers of the UseTox model [58], which is the commonly used model for impact assessment, clearly state, “It should be stressed that the characterization factors are useful for a first-tier assessment. In case a substance appears to dominantly contribute to the impact scores for toxicity, it is recommended to verify the reliability of the chemical-specific input data for this substance and to improve the data whenever possible.”
4.5
Closing remarks
Geothermal energy can represent a significant contribution to the reduction of energy production from fossil fuels. In contrast to other renewables, the geothermal source is able to provide a stable energy output, unaffected by the external environment, and thus is suitable for baseload electricity
production. To ensure the transition from fossil to renewables and reach the prefixed goal within the energy sector, it becomes essential to evaluate and improve the environmental performances of a geothermal project. As highlighted in this chapter, the ecoprofile of a geothermal installation through its full life cycle depends both on geological conditions and the technology used to exploit the geothermal energy. The available LCA studies in the scientific literature show that there is no standard available at the moment to approach geothermal system modeling and to assess the life cycle impacts of geothermal energy production. Indeed, each installation is somehow unique. This is due to the peculiarity of the geothermal reservoir and the variety of technology configurations to produce geothermal energy. Thus, it is hard to make a general assessment because, in most cases, geological and local conditions dominate the technological, environmental, and economic aspects of geothermal plants. This criticality directly reflects on the fact that it is very hard to define a typical and representative LCA case study with a comprehensive life cycle inventory available. Furthermore, in the literature, many technical and environmental aspects are considered sensitive information, which is seldom openly disclosed. When available, data are focused almost exclusively on global warming potential, overlooking other potential impacts.
Life cycle assessment of geothermal power plants Chapter 4
Currently, there is a lack of information available on potential LCA impacts from geothermal power plants that need to be overtaken in order to support geothermal energy development. This is particularly true for flash power plants that, nowadays, still represent the biggest share of energy production from geothermal resources. Binary plants have been shown to perform better in terms of direct emissions to the atmosphere due to the closed loop of the geothermal fluid. However, this technology could present a higher impact on the land use category due to the larger surface required for the dry cooling towers. It is important to expand the set of impact indicators beyond the global warming potential to avoid any potential impact shifting. Including a complete set of indicators in LCA studies of geothermal plants would allow a more conscious decision on the best sustainable plant technology, size, and overall plant assets for any specific geographical location.
[12]
[13]
[14]
[15]
[16]
References [1] European Geothermal Energy Council. EGEC geothermal market report 2018; 2019. p. 20. [2] GGA. “Agenda for sustainable development”, and its 17 sustainable development goals (SDGs), https://www.globalgeothermalalliance. org/-/media/Files/IRENA/GGA/Event/GGA-Florence-DeclarationFINAL.PDF?la¼en&hash¼0FFB185E9EC6CE3DD35F439C20170 C1FFD639BB7; 2017. [Accessed 27 February 2020]. [3] European Commission. SET Plan-Declaration of intent on Strategic Targets in the context of an Initiative for Global Leadership in Deep Geothermal Energy, Brussels, http://www.irena.org/News/ Description.aspx?NType¼A&mnu¼cat&PriMenuID¼16& CatID¼84&News_ID¼438; 2016. [Accessed 27 February 2020]. [4] International Renewable Energy Agency (IRENA). Global energy transformation: a roadmap to 2050 (2019 edition); 2019. [5] Sigfusson B, Uihlein A. European Joint Research Centres— geothermal energy status report; 2015. https://doi.org/10.2790/ 959587. [6] Shortall R, Uihlein A, Carrara S. Geothermal energy technology market report 2018, https://ec.europa.eu/jrc/en/publication/geo thermal-energy-technology-market-report; 2019. [Accessed 27 February 2020]. [7] Shortall R, Uihlein A. Geothermal energy technology development report; 2018. [8] Kl€ opffer W. Life cycle assessment: from the beginning to the current state. Environ Sci Pollut Res 1997;4:223–8. https://doi.org/10.1007/ BF02986351. [9] Guinee JB, Heijungs R, Huppes G, Zamagni A, Masoni P, Buonamici R, Ekvall T, Rydberg T. Life cycle assessment: past, present, and future. Environ Sci Technol 2011;45:90–6. https://doi.org/10.1021/ es101316v. [10] Basosi R, Bonciani R, Frosali D, Manfrida G, Parisi ML, Sansone F. Life cycle analysis of a geothermal power plant: comparison of the environmental performance with other renewable energy systems. Sustainability 2020;12:2786. https://doi.org/10.3390/su12072786. [11] Bravi M, Parisi ML, Basosi R. Life cycle assessment of advanced technologies for photovoltaic panel production LCA of Geothermal Systems View Project Novel Photovoltaics LCA View
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
61
project, https://www.researchgate.net/publication/244485998; 2010. [Accessed 6 May 2020]. Maranghi S, Parisi ML, Basosi R, Sinicropi A. Environmental profile of the manufacturing process of perovskite photovoltaics: harmonization of life cycle assessment studies. Energies 2019;12:3746. https://doi.org/10.3390/en12193746. Parisi ML, Maranghi S, Vesce L, Sinicropi A, Di Carlo A, Basosi R. Prospective life cycle assessment of third-generation photovoltaics at the pre-industrial scale: a long-term scenario approach. Renew Sustain Energy Rev 2020;121:109703. https://doi.org/10.1016/j. rser.2020.109703. Peng J, Lu L, Yang H. Review on life cycle assessment of energy payback and greenhouse gas emission of solar photovoltaic systems. Renew Sustain Energy Rev 2013;19:255–74. https://doi.org/10.1016/ j.rser.2012.11.035. Rossi F, Parisi ML, Maranghi S, Basosi R, Sinicropi A. Environmental analysis of a nano-grid: a life cycle assessment. Sci Total Environ 2020;700:134814. https://doi.org/10.1016/j.scitotenv. 2019.134814. Parisi ML, Maranghi S, Basosi R. The evolution of the dye sensitized solar cells from Gr€atzel prototype to up-scaled solar applications: a life cycle assessment approach. Renew Sustain Energy Rev 2014;39:124–38. https://doi.org/10.1016/j.rser.2014.07.079. European Union. Communication from the Commission to the Council and the European Parliament — Integrated Product Policy, Building on Environmental Life-Cycle Thinking, Amtsblatt Der Eur. Union. C80/39; 2003. p. 6. http://europa.eu.int/com m/environment/ipp/ippsum.pdf. [Accessed 27 February 2020]. Karlsdottir MR, Palsson OP, Palsson H. Factors for primary energy efficiency and CO2 emission of geothermal power production. Proceedings of the World Geothermal Congress, Bali, Indonesia; 2010. Paulillo A, Striolo A, Lettieri P. The environmental impacts and the carbon intensity of geothermal energy: a case study on the Hellisheiði plant. Environ Int 2019;133:105226. https://doi.org/10.1016/j. envint.2019.105226. European Commission – Joint Research Centre – Institute for Environment and Sustainability. International reference life cycle data system (ILCD) handbook: general guide for life cycle assessment – provisions and action steps. EUR 24378 EN, Luxembourg; 2010. https://doi.org/10.2788/94987. Dumas P, Peterschmitt A, Garabetian T, Manfrida G, Loschetter A, Ravier G, Blanc I, Manzella A, Torsello L, Scalari S, Batini F, Ragnarsson A, Pa´ll Ingo´lfsso H, Johannesson G, Laenen B, Bozkurt C, Solak O, Nador A. GEOENVI Project: tackling the environmental concerns for deploying geothermal energy in Europe. In: European Geothermal Congress 2019 Den Haag; 2019. Parisi ML, Ferrara N, Torsello L, Basosi R. Life cycle assessment of atmospheric emission profiles of the Italian geothermal power plants. J Clean Prod 2019;234:881–94. https://doi.org/10.1016/j.jclepro. 2019.06.222. Tomasini-Montenegro C, Santoyo-Castelazo E, Gujba H, Romero RJ, Santoyo E. Life cycle assessment of geothermal power generation technologies: an updated review. Appl Therm Eng 2017;114: 1119–36. https://doi.org/10.1016/j.applthermaleng.2016.10.074. Yu T. A life cycle assessment based comparison of large & small scale geo-thermal electricity production systems. In: Proc. 5th Indones. Int. Geotherm. Conv. Exhib. 2017; 2017. Sullivan JL, Wang MQ. Life cycle greenhouse gas emissions from geothermal electricity production. J Renew Sustain Energy 2013;5:063122. https://doi.org/10.1063/1.4841235.
62
PART
I Basics of Geothermal Power Plants
[26] Sullivan JL, Clark C, Han J, Harto C, Wang M. Cumulative energy, emissions, and water consumption for geothermal electric power production. J Renew Sustain Energy 2013;023127. https://doi.org/ 10.1063/1.4798315. American Institute of Physics. [27] Bauer C, Burgherr P, Hirschberg S, Miotti M, Oshikawa H, Schenler W, et al. Energy from the earth: deep geothermal as a resource for the future?; 2015. TA-SWISS 62. vdf Hochschulverlag. [28] Asdrubali F, Baldinelli G, D’Alessandro F, Scrucca F. Life cycle assessment of electricity production from renewable energies: review and results harmonization. Renew Sustain Energy Rev 2015;42: 1113–22. https://doi.org/10.1016/j.rser.2014.10.082. [29] Chiavetta C, Tinti F, Bonolia A. Comparative life cycle assessment of renewable energy systems for heating and cooling. Procedia Eng 2011;591–7. https://doi.org/10.1016/j.proeng.2011.11.2054. Elsevier Ltd. [30] Amponsah NY, Troldborg M, Kington B, Aalders I, Hough RL. Greenhouse gas emissions from renewable energy sources: a review of lifecycle considerations. Renew Sustain Energy Rev 2014;39: 461–75. https://doi.org/10.1016/j.rser.2014.07.087. [31] Pehnt M. Dynamic life cycle assessment (LCA) of renewable energy technologies. Renew Energy 2006;31:55–71. https://doi.org/10.1016/ j.renene.2005.03.002. [32] Pratiwi A, Ravier G, Genter A. Life-cycle climate-change impact assessment of enhanced geothermal system plants in the Upper Rhine Valley. Geothermics 2018;75:26–39. https://doi.org/10.1016/j. geothermics.2018.03.012. [33] Bravi M, Basosi R. Environmental impact of electricity from selected geothermal power plants in Italy. J Clean Prod 2014;66:301–8. https:// doi.org/10.1016/j.jclepro.2013.11.015. [34] Martı´nez-Corona JI, Gibon T, Hertwich EG, Parra-Saldı´var R. Hybrid life cycle assessment of a geothermal plant: from physical to monetary inventory accounting. J Clean Prod 2017;142:2509–23. https://doi. org/10.1016/j.jclepro.2016.11.024. [35] Bayer P, Rybach L, Blum P, Brauchler R. Review on life cycle environmental effects of geothermal power generation. Renew Sustain Energy Rev 2013;26:446–63. https://doi.org/10.1016/j.rser. 2013.05.039. [36] Hetherington AC, Borrion AL, Griffiths OG, McManus MC. Use of LCA as a development tool within early research: challenges and issues across different sectors. Int J Life Cycle Assess 2014;19:130–43. https://doi.org/10.1007/s11367-013-0627-8. [37] Marchand M, Blanc I, Marquand A, Beylot A, Bezelgues-Courtade S, Traineau H. Life cycle assessment of high temperature geothermal energy systems. Proceedings of the World Geothermal Congress, Melbourne, Australia; 2015. [38] Frick S, Kaltschmitt M, Schr€oder G. Life cycle assessment of geothermal binary power plants using enhanced low-temperature reservoirs. Energy 2010;35:2281–94. https://doi.org/10.1016/j.energy. 2010.02.016. [39] Lacirignola M, Meany BH, Padey P, Blanc I. A simplified model for the estimation of life-cycle greenhouse gas emissions of enhanced geothermal systems. Geotherm Energy 2014;2:8. https://doi.org/ 10.1186/s40517-014-0008-y. [40] Peng J, Lu L, Yang H. Review on life cycle assessment of energy payback and greenhouse gas emission of solar photovoltaic systems. Renew Sustain Energy Rev 2013;19:255–74. https://doi.org/10.1016/ j.rser.2012.11.035. [41] Lacirignola M, Blanc P, Girard R, Perez-Lo´pez P, Blanc I. LCA of emerging technologies: addressing high uncertainty on inputs’ variability
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
when performing global sensitivity analysis. Sci Total Environ 2017;578:268–80. https://doi.org/10.1016/j.scitotenv.2016.10.066. Saouter E, Biganzoli F, Ceriani L, Versteeg D, Crenna E, Zampori L, Sala S, Pant R. Environmental footprint : Update of life cycle impact assessment methods : Ecotoxicity freshwater, human toxicity cancer, and non-cancer. Luxembourg; 2018. https://doi.org/10.2760/611799. JRC114227. ´ P, Pa´lsson H, Maya-Drysdale L. Life cycle Karlsdo´ttir MR, Pa´lsson O inventory of a flash geothermal combined heat and power plant located in Iceland. Int J Life Cycle Assess 2015;20:503–19. https:// doi.org/10.1007/s11367-014-0842-y. Guinee J. Handbook on life cycle assessment – operational guide to the ISO standards. Int J Life Cycle Assess 2001;6:255. https://doi. org/10.1007/BF02978784. CML. CML-IA characterisation factors. Leiden University, Inst. Environ. Sci https://www.universiteitleiden.nl/en/research/researchoutput/science/cml-ia-characterisation-factors; 2016. [Accessed 25 February 2020]. Lacirignola M, Blanc I. Environmental analysis of practical design options for enhanced geothermal systems (EGS) through life-cycle assessment. Renew Energy 2013;50:901–14. https://doi.org/ 10.1016/j.renene.2012.08.005. Martı´n-Gamboa M, Iribarren D, Dufour J. On the environmental suitability of high- and low-enthalpy geothermal systems. Geothermics 2015;53:27–37. https://doi.org/10.1016/j.geothermics.2014.03.012. McCay AT, Feliks MEJ, Roberts JJ. Life cycle assessment of the carbon intensity of deep geothermal heat systems: a case study from Scotland. Sci Total Environ 2019;685:208–19. https://doi.org/ 10.1016/j.scitotenv.2019.05.311. Buonocore E, Vanoli L, Carotenuto A, Ulgiati S. Integrating life cycle assessment and emergy synthesis for the evaluation of a dry steam geothermal power plant in Italy. Energy 2015;86:476–87. https:// doi.org/10.1016/j.energy.2015.04.048. Atilgan B, Azapagic A. Renewable electricity in Turkey: life cycle environmental impacts. Renew Energy 2016;89:649–57. https://doi. org/10.1016/j.renene.2015.11.082. Heberle F, Schifflechner C, Br€uggemann D. Life cycle assessment of organic Rankine cycles for geothermal power generation considering low-GWP working fluids. Geothermics 2016;64:392–400. https://doi. org/10.1016/j.geothermics.2016.06.010. Paulillo A, Cotton L, Law R, Striolo A, Lettieri P. Geothermal energy in the UK: the life-cycle environmental impacts of electricity production from the United Downs Deep Geothermal Power project. J Clean Prod 2020;249. https://doi.org/10.1016/j. jclepro.2019.119410. ´ skarsson F, Perez NM. Application of Fridriksson T, Padro´n E, O diffuse gas flux measurements and soil gas analysis to geothermal exploration and environmental monitoring: example from the Reykjanes geothermal field, SW Iceland. Renew Energy 2016;86:1295–307. https://doi.org/10.1016/j.renene.2015.09.034. Frondini F, Caliro S, Cardellini C, Chiodini G, Morgantini N. Carbon dioxide degassing and thermal energy release in the Monte Amiata volcanic-geothermal area (Italy). Appl Geochemistry 2009;24: 860–75. https://doi.org/10.1016/j.apgeochem.2009.01.010. Intergovernmental Panel on Climate Change. IPCC updates methodology for greenhouse gas inventories; 2019. doi: 1037//0033-2909. I26.1.78. Parisi ML, Basosi R. Geothermal energy production in Italy: an LCA approach for environmental performance optimization. In: Green
Life cycle assessment of geothermal power plants Chapter 4
energy technol. Springer Verlag; 2019. p. 31–43. https://doi.org/ 10.1007/978-3-319-93740-3_3. [57] Karlsdottir MR, Lew JB, Palsson P. Geothermal district heating system in Iceland: a life cycle perspective with focus on primary energy efficiency and CO2 emissions. In: 14th Int. Symp. Dist. Heat. Cool; 2014. https://www.researchgate.net/publication/ 273447593_GEOTHERMAL_DISTRICT_HEATING_SYSTEM_IN_ ICELAND_A_LIFE_CYCLE_PERSPECTIVE_WITH_FOCUS_
63
ON_PRIMARY_ENERGY_EFFICIENCY_AND_CO_2_EMIS SIONS. [Accessed 13 February 2020]. [58] Fantke P, Bijster M, Guignard C, Hauschild MZ, Huijbregts MAJ, Jolliet O, Kounina A, Magaud V, Margni M, McKone TE, Posthuma L, Rosenbaum RK, van de Meent D, van Zelm R. USEtox 2.0 documentation (version 1). USEtox International Center hosted at the Technical University of Denmark; 2017. https://doi.org/10.11581/ DTU:00000011.
Chapter 5
Social acceptance of geothermal power plants Spyridon Karytsasa,b and Olympia Polyzoua a
Geothermal Energy Department, Division of Renewable Energy Sources, Centre for Renewable Energy Sources and Saving (CRES), Pikermi, Greece
b
Department of Economics and Sustainable Development, School of Environment, Geography and Applied Economics, Harokopio University (HUA),
Kallithea, Greece
5.1
Introduction
Under the concerns of climate change and energy dependence that are increasing globally, geothermal power plants can assist toward the goal of sustainable development [1]. Nevertheless, the implementation and operation of geothermal power plants are heavily dependent on their acceptance at the local community level where the facility is to be constructed. Based on academic research, the absence of social acceptance may amplify failure risks while increasing costs and project setbacks, possibly even resulting in the cancelation of the works [2–4]. Various interpretations have been provided in relation to effective social acceptance of geothermal power production projects. As per de Jesus [5], “social acceptability is attained if the project activities do not result in drastic changes from the regular conditions of the area and if the affected sectors can see some advantages issuing from the project.” Likewise, Cataldi [6] states that “social acceptability of a profit-purported project is the condition upon which the technical and economic objectives of the project may be pursued in due time and with the consensus of the local communities; consensus to be gained by acting in consonance with the dynamic conditions of the environment, and in the respect of the people’s health, welfare, and culture.” On top of that, Popovski [7] continued by noting that “social acceptability is one of the most important parts of the process of geothermal energy development in a specific environment. It is not possible to complete a successful project if initially not identifying the elements of the local environment, which can influence its social acceptance; and not designing proper organizational, technical, economic, and other solutions in order to remove the negative opinions.” Against this background, and with the further aim of moving toward acceptance of such plants by local communities, the present study focuses on the assessment and
presentation of themes associated with social acceptance in relation to the development and operation of geothermal power plants. Accordingly, Section 5.2 defines social acceptance and its distinct levels while Section 5.3 presents the factors that affect a project’s community acceptance. Section 5.4 reports the socioeconomic impacts deriving from renewable energy projects, and Section 5.5 deals with the measurement of these impacts. Section 5.6 presents cases of social resistance against geothermal energy developments around the world, whereas Section 5.7 describes practices performed by geothermal operators and developers within the view of achieving community acceptance; this section also outlines the role of public authorities in relation to community acceptance practices. The chapter concludes with Section 5.8, where final remarks are outlined.
5.2 Social acceptance of renewable energy technologies Social acceptance is one of the most important elements during the planning, development, and operation of a renewable energy project, particularly of geothermal energy. Social acceptance is a crucial element of the time and effort needed to implement energy projects [8]. Often, it is impossible to proceed with this kind of development if the project developer does not take into account the local parameters (socioeconomic, cultural), defining at the same time the benefits of the plant to local communities to remove social opposition [6, 7]. Besides, the importance of social acceptance is recognized by declarations such as that from European Commission, which said that “The current trend, in which nearly every energy technology is disputed and its use or deployment delayed, raises serious problems for investors and puts energy system changes at risk” [9]. Social acceptance of renewable energy projects
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00004-4 Copyright © 2021 Elsevier Inc. All rights reserved.
65
66
PART
I Basics of geothermal power plants
Sociopolitical acceptance
Social acceptance Market acceptance
Community acceptance
€ FIG. 5.1 The triangle of social acceptance. (Modified from Wustenhagen € R, Wolsink M, Burer MJ. Social acceptance of renewable energy innovation: an introduction to the concept. Energy Policy 2007;35:2683–91.)
was described in 2007 by W€ ustenhagen et al. [10] as a combination of three interdependent levels in the so-called “triangle of social acceptance,” namely sociopolitical acceptance, community acceptance, and market acceptance (Fig. 5.1). The first level, sociopolitical acceptance, is acceptance on a broader and general scale. According to W€ustenhagen et al. [10], this level deals with the acceptance of technologies and policies by (a) the public, (b) key stakeholders, and (c) policymakers. Sociopolitical acceptance includes the public and key stakeholders’ general opinion as well as effective policies and frameworks [10]. Usually, this level of acceptance is very high when referring to renewable energy, as has been stated in social surveys in many countries around the world [11, 12]. However, despite the positive attitude in general, perceptions toward specific projects among local communities can lead to opposition [11]. This brings us to the second level of acceptance, described as community acceptance [10]. Practically, community acceptance refers to acceptance from local societies, including both residents and local authorities, concerning the site selection of a new project. It should be noted that in many cases, local communities are opposed to the project developers and not against the project itself [13]. Wolsink [14] shows that in many cases, the acceptance level of the public is high prior to the installation of the project, drops significantly during the construction phase of the project, and increases again after the completion of the work. This level of acceptance—community acceptance—is the most relevant type when referring to geothermal power plant development and operation. The relevant stakeholders involved in the planning, development, and operation phases of such a project include (a) project developers/operators, (b) local authorities, (c) local and national nonprofit and nongovernmental organizations (e.g., for the environment, wildlife, and society), (d) other interest groups,
and (e) local communities (e.g., residents, business owners, organizations, and landowners). The third level of acceptance is market acceptance. It refers to the process of adoption of innovation or acceptance of technology by both consumers and investors in the market. This dimension also includes the intrafirm acceptance of an innovation, referring to the framework and strategy that should be developed by each firm to achieve public acceptance, taking into account environmental and sustainability issues [10]. In the context of geothermal energy applications, an example of market acceptance is the adoption of ground source heat pumps by homeowners.
5.2.1 Studies on the social acceptance of geothermal energy The research concerning the social acceptance of geothermal energy is still limited but has been increasing worldwide. Particularly, it should be noted that public perceptions of the exploitation and utilization of geothermal energy vary significantly, with perspectives advancing over the years differing across locations [15]. When referring to the acceptance of geothermal energy on a community level, qualitative and quantitative studies have been performed for specific cases in Greece [13], Australia [16, 17], Japan [18], Canada [19], and Italy [15]. Furthermore, with respect to market acceptance of geothermal technologies, work has been carried out for various countries, mainly European, such as Greece [20–22], the United Kingdom [23], and Finland [24]. Likewise, surveys dealing with the public’s awareness level of geothermal energy have been conducted in a variety of places, including Turkey [25], Greece [26], and a collection of European countries [27].
5.3 Factors affecting community acceptance of renewable energy projects In recognition of the significance of community acceptance of renewable energy projects, studies have provided a variety of theoretical frameworks that describe perspectives toward energy transitions. The “NIMBY” (Not in My Backyard) concept was one of the first proposals, suggesting that even though individuals generally support renewable energy source (RES) initiatives, it is possible that they may object to particular project proposals in their locality on the basis of personal interests and motives [28]. Thus, the public is not willing to accept any kind of risk in favor of society [12]. However, relevant research has exceeded the “NIMBY” argument, which presents the individuals opposing the project as selfish, poorly informed, and unaware of the public interest [29]. Alternatively, researchers have offered a variety of potential reasons for lack of community acceptance toward
Social acceptance of geothermal power plants Chapter 5
Personal factors
Socio-psychological factors
• Age
• Political ideology
• Gender
• Emotional connection to location
• Income
• Procedural justice
• Awareness/ experiences
• Local patterns
• Environmental interests
• Trust level
67
Community acceptance of renewable energy projects
Contextual factors
Spatial factors
• Magnitude of development • Technology type • Ownership type • Distributional justice • Perceived benefits/ costs • Level of public engagement
• Regional/local background • Proximity to energy installations • Spatial protectionism
FIG. 5.2 Categorization of factors affecting community acceptance of renewable energy projects. (Modified from Stephanides P, Chalvatzis KJ, Li X, Lettice F, Guan D, Ioannidis A, Zafirakis D, Papapostolou C. The social perspective on island energy transitions: evidence from the Aegean archipelago. Appl Energy 2019;255:113725.)
energy projects. W€ ustenhagen et al. [10] determined three factors that influence social acceptance: procedural justice (a fair decision-making process with the participation of all relevant stakeholders), distributional justice (sharing fairly costs and benefits), and trust (investors and actors from outside the community should inspire trust in local communities). Furthermore, according to Friedl and Reichl [8], the most important factors are considered to be safety issues, noise, environmental pollution, landscape degradation, depreciation of property values, and perceived procedural inequality. The factors have been classified as “personal, sociopsychological, contextual, and spatial” [30], as outlined in Fig. 5.2.
5.4 Socioeconomic impacts of renewable energy projects Current research has focused mainly on the environmental benefits (including global and local pollutants) of RES, whereas socioeconomic impacts have not gained equivalent attention [31]. Nevertheless, socioeconomic benefits are becoming essential for the introduction of renewable energy projects, as policymakers see opportunities for increased revenue, economic development, and job positions in low-growth economies [32]. The socioeconomic impacts of RES developments can be classified into four categories [33]: public perception (social benefits, lifestyle effects, changes of property values, aesthetics, etc.); employment
(job opportunities, diversity of the type of jobs, poverty mitigation, etc.); safety and health (public safety, work safety, etc.); and improvement of local infrastructure (infrastructure development, local empowerment, etc.). RES may provide a plethora of socioeconomic benefits both at the national and local scale. According to the available literature, the key benefits that RES projects may bring at the national scale are [34,35]: l l l l l
Environmental protection. Energy security protection. Diversification of risks. Trade balance improvement. Export opportunities.
In addition, the main socioeconomic benefits at the local level include [34–38]: l
l
l
l l
l
Employment creation (direct, indirect, and induced job positions). Improvement of income and new sources of revenue for local communities (e.g., increase in the tax base, income for landowners, and land-based activities). Cheap energy (reliable and low-cost energy generation can trigger economic development). Increased self-reliance. Customers having a choice concerning their energy source. Supply of power to regions lacking electricity grid access.
68
PART
I Basics of geothermal power plants
FEEDBACK
Input
Activites
Output
Outcome
Goal Alignment
MINUS What would still happen IMPACT FIG. 5.3 Impact value chain. (Modified from Clark C, Rosenzweig W, Long D, Olsen S. Double bottom line project report: assessing social impact in double bottom line ventures. Center for Responsible Business, Working Paper Series, University of California, Berkeley, Paper 13, 2004.)
l
l l
Technological advancements/development of local infrastructure. Enhancement of local population skills and education. Better health and life quality.
These benefits can lead to increased social cohesion and stability (leading to a reduction of rural depopulation, regional development, and rural diversification) resulting from the establishment of an activity that generates employment and income as well as an overall increase in living standards in terms of the environment, health, and education [34]. In addition, it should be mentioned that the social impacts produced within community-based RES projects may lead to additional beneficial sustainability outcomes such as awareness and acceptance of renewable and sustainable technologies, uptake of low carbon technologies, and sustainable/environmental behaviors [39]. Oppositely, the potential negative socioeconomic impacts of RES projects may comprise visual impacts, noise, unpleasant smells, landscape impacts, potential displacement (mandating resettlement when expropriating land), and heritage and culture impacts [35,40]. The impacts created could have a diverse effect on different demographic groups; for instance, they could have a larger— either favorable or unfavorable—effect on women or youth. In this sense, Nelson and Kuriakose [41] emphasize the socioeconomic effects that renewable energy projects can create for women, their families, and their communities’ livelihoods and employment prospects.
5.5 Measuring socioeconomic impacts of renewable energy projects 5.5.1
Defining social impacts
Several attempts have been made to describe “social impact” on common ground, even if it is evident that consensus on the definition of the term has not yet been achieved [42]. Impacts consist of both expected and unexpected long-term and short-term effects [43], whereas the
importance of each social impact may differ depending on the location, the project, and the involved stakeholders. The absence of agreement acts as an obstacle to the academic debate on social impact, along with the use of methods for evaluating it [44,45]. Karytsas et al. [42] delivered a selection of definitions of “social impact” along with definitions of other associated terms. Clark et al. [46] attempted to fine tune all the different perspectives to provide common ground for this issue. This resulted in the “impact value chain” (Fig. 5.3), which distinguishes outputs, outcomes, and impacts, and clarifies the differences between these concepts. Input (resources devoted to the activity), output (direct and measurable outcomes of the activity), outcome (alterations caused to people by the activity), and impact (outcomes minus an estimate of what would have nevertheless happened) are the essential elements of the value chain.
5.5.2 Significance of measuring social impacts Why is social impact evaluation so valuable? As the New Economics Foundation [47] has stated, “What gets measured, gets valued.” Measurements are beneficial only if they are appropriate and effective. Therefore, as mentioned by Muir and Bennett [48] it should be clear “what different stakeholder groups want from measurement and how the outcomes will be used.” It should be recognized, in terms of energy-related projects, that identifying and quantifying nonenergy effects may benefit performance via program design and marketing through considering not only energy-related effects but also stakeholders’ nonenergy priorities [49]. Evaluating social impacts ought to be of great importance to companies, organizations, and institutions, as it can assist them to [45,50–52]: (a) inform and engage stakeholders, thus establishing confidence and mutually beneficial results, (b) apply as a marketing tool and advertising material, (c) strengthen transparency and reputation by delivering measurable results, (d) gain or sustain company or project support, (e) retain a license to operate,
Social acceptance of geothermal power plants Chapter 5
(f) promote the development of a more desirable policy and funding framework, (g) promote product and service development, (h) detect problems early, thus preventing and reducing costs as opposed to unplanned approaches, and (i) improve conformity with international standards and guidelines.
5.5.3 Stages and methods of social impact measurement The evaluation of social impacts can be a complicated process that calls for extensive preparation and requires carrying out procedures and actions with a view of tracking progress against specific goals [53,54]. In the interest of measuring social impact, a methodology—employing different tools or techniques to gather information—is required to coordinate the examination of all dimensions of impact evaluation [55]; nevertheless, up until now, there has been no consistent methodology for estimating social impact [56]. Through the incorporation of the Impact Measurement Roadmap [53] and the GECES Subgroup on Social Impact Measurement [57] methodologies, impact evaluation can be defined as a five-step procedure composed of: (a) target identification, (b) stakeholder identification, (c) establishing appropriate evaluation methods, (d) measurement, validation, and assessment, and (e) reporting, learning, and developing. Developing and implementing effective indicators and benchmarks is an integral part of evaluating social impact, as they demonstrate whether advancements have been made on particular results or objectives, indicating zero, favorable, or unfavorable change over a period of time. This offers a universal language and framework for effective communication among all associated stakeholders [48,50]. The level of the analysis conducted should be determined, namely if it is to be at a microscale (individual or program), mesoscale (organizational or community), or macroscale (societal or sector). This choice should support the selection of proper indicators and relevant data [48]. A further element that should be considered while selecting the indicators is that impact can only be fully measured and interpreted by understanding the local context [54]. Applicable social indicators can fall within the groups of social well-being, energy independence and availability, commerce, competitiveness, resource sustainability, and social acceptance when referring specifically to energy systems [58,59]. How indicators are applied and evaluated composes the processes or techniques employed [48]. There are various approaches for evaluating social impacts. Each approach has its strengths and weaknesses, offering a distinct point of view, while it can be implemented independently to a set of indicators or can be integrated into entire change
69
evaluation frameworks. Companies, organizations, etc., develop different measurement approaches around the world, differing in context, function, and methods [60]. Historically, as stated by Dufour [61], two patterns exist in evaluating social impact, one based on a social accounting and audit (SAA) and the other on social impact assessment (SIA). SAA offers a method to account for the social, environmental, and economic actions of an organization and, where feasible, the implications of these actions [62]. This being said, SIA covers the mechanisms of evaluating, tracking, and controlling expected and unexpected implications—both positive and negative—of proposed initiatives, together with any forms of social change triggered by them [63].
5.6
Cases of controversy
Cases of controversy involving geothermal power projects have been recorded globally; incidents in different parts of the world have indicated that social opposition may substantially postpone, or even prevent, the development and operation of geothermal projects. The diversity of examples shows that the motivations leading to a lack of community acceptance are very complex and location-specific [64]. It can also be noted that populations with higher standards of life quality question more the development of these large-scale projects because the expected individual gains are less valuable to them [64]. According to Reith et al. [65], the issues that may “trigger” these motivations can be classified into (a) environmental, (b) missing involvement, (c) financial, and (d) NIMBY issues. In the following, the cases of the Berlı´n power plant in El Salvador, the Lower Kilauea East Rift Zone in Hawaii, the Milos and Nisyros Islands in Greece, the Mt. Apo and the Tiwi projects in the Philippines, and the projects located in different European countries (Upper Rhine Graben) are presented.
5.6.1
Berlı´n power plant (El Salvador)
The Berlı´n power plant, situated in a high-poverty rural area, was constructed by the public electric utility company (CEL), and in 1999 was passed to the LaGeo company within the framework of the electricity sector’s liberalization. When the power plant began operations, the surrounding population accused the project of having a detrimental impact on their quality of life, then requested the termination of the operation. The residents blamed the power plant for groundwater and air pollution as well as the overall degradation of their health and well-being. The unfavorable opinion was so intense that roads were closed while legal actions against the operation were taken with the assistance of environmental groups [66].
70
PART
I Basics of geothermal power plants
LaGeo, from the beginning of its involvement in 1999, introduced certain initiatives to resolve local resistance. Rather than reacting with aggression and responding to the legal actions, the company worked together with specific local stakeholders to mitigate or resolve any unfavorable impacts while showing that the surrounding communities could take advantage of the geothermal plant and its side products [66].
5.6.2 Lower Kilauea East Rift Zone (Hawaii, United States) The HGP-A well was constructed in 1976 in the Kilauea East Rift Zone on the island of Hawaii (the “Big Island”) by a public-private collaboration. At that moment, the well was identified as one of the universally hottest (358°C). Considerable controversy arose over the specific geothermal project. The pilot HGP-A power plant was not regarded favorably by the surrounding population in view of air and groundwater pollution, noise, and the unpleasant image of the plant itself due to inadequate maintenance. Future research was opposed, sometimes strongly, by people voicing concerns on different issues, namely disturbance of the surrounding communities’ lifestyles, impacts on Hawaiian cultural and religious traditions (disturbance of Goddess Pele worship; disturbance of specific Native Hawaiian practices), possible environmental risks and degradation, and potential unfavorable health and safety effects [67]. Following 10 years of operation, the Puna Geothermal Venture (PGV) managed to develop and operate a 35MWe power plant, after many setbacks and significantly higher costs than had been expected by the initial investors; at that point, most Hawaiian citizens had embraced it as part of their energy supply. Nevertheless, there had been ongoing attention over health and environmental concerns among people residing close to the facility, leading to examination by the US Environmental Protection Agency and monitoring of health issues of citizens connected to geothermal emissions. The relationship between PGV and the surrounding communities appeared to be improved through better communication between the company and its neighbors [67].
5.6.3
Milos Island (Greece)
Geothermal exploration in Milos Island was launched in 1973, and geothermal resources of high (325°C) and low (25–90°C) temperatures were discovered. Following this, the Public Power Corporation (PPC) started building a pilot power plant with the initial drillings done in 1982 near the island’s largest city, Adamantas, where the growth of tourism had just started. Residents’ negative experience with established industrial plants (most of the island had already been impacted by mining activity) resulted in the
local society’s strong opposition to the pilot power plant [68,69]. Moreover, mining companies operating on the island objected to geothermal development, being convinced that it would cause problems with their work [70]. In addition, shortcomings and miscalculations in the course of the pilot power plant’s development and operation stages resulted in pollution of the air, surface water, and seawater. This condition produced further objections, adding to the existing unfavorable opinion of the local community [68]. The consequence of these implications was the surrounding community’s response through protests (1987–89), contributing to the shutdown of the pilot plant. To this day, after nearly 40 years, the citizens’ adverse feelings have not settled. People are still against the exploration of high-temperature geothermal resources, as they think the state will ignore the local community’s perspective. Furthermore, they reject the involvement of PPC Renewable S.A. when referring to any geothermal energy development. In contrast, the municipality has decided to utilize low-temperature resources, an option that is acceptable by most local residents [68,69].
5.6.4
Mt. Apo project (Philippines)
Most consider the case of the Mt. Apo geothermal project to be one of the most divisive Philippine development projects. In 1983, the PNOC Energy Development Corporation (PNOC-EDC) conducted its initial geological studies in the area. A permit for exploration of the area was obtained in 1987; the exploratory wells revealed exploitable geothermal resources. In these early stages, the progress of the work did not create any tensions between the company and the locals. However, some subsequent events, steered by environmental protection initiatives, led to PNOCEDC’s accusation concerning various economic, environmental, social, legal, and cultural violations. In this context, PNOC-EDC responded by calling meetings in order to inform the relevant stakeholders about the project. During these meetings, the company confirmed that those who opposed the project originated from outside the area of the project, even from distant areas. The opposition groups formed a network that would enable them to gather support from nearby areas while also gaining mass media support [71]. The company presented an updated environmental impact plan in 1991, thus obtaining an Environmental Clearance Certificate in 1992. The opposition groups took the case to the Supreme Court two times, which ultimately dismissed their request to block the project, thereby enabling PNOC-EDC to restart exploration work in the Mt. Apo location. As a result, PNOC-EDC, which initially intended to launch project development in 1989, was only able to restart its work in 1992, that is, after a nearly 4-year interruption [71].
Social acceptance of geothermal power plants Chapter 5
5.6.5
Nisyros Island (Greece)
Geothermal exploration on Nisyros Island began in 1973, defining high (400°C) and low (25–90°C) resources. PPC suggested the development of a power plant in 1982. Therefore, two drillings were made in 1983 in the range of the volcano’s caldera, which was the island’s major tourist area of interest. A significant part of the local society objected to the development of a power plant, taking into consideration environmental protection in tandem with the risk of seismic and/or volcanic activity by disrupting the volcano’s equilibrium. Together with the poor example set by the Milos Island events, the experience of the island’s test drilling drove the locals to demonstrate their aversion toward exploiting the geothermal resources of the island via a referendum carried out in 1997 [68,69]. Local community reactions against the construction of a power production unit exist to this day; however, the local community is in favor of the exploitation of lowtemperature geothermal resources. Local people feel they are not notified and their perspectives do not matter when talking about the extraction of geothermal energy. They do not approve PPC Renewable S.A. being responsible for the power plant’s development while the island’s small size and its environmental peculiarity (Natura 2000 area) are other factors that add up to their unfavorable position. Public reactions are followed by activities such as awareness of relevant stakeholders, communication activities focusing on claims against geothermal power plant development, and numerous other actions such as meetings and rallies [69].
5.6.6
Tiwi power plant project (Philippines)
The Tiwi field, situated in the Albay province, is the first geothermal site in the Philippines to be exploited commercially on a large scale. In 1962, the Commission on Volcanology performed geothermal research, finding a potential of 500–800 MW that could generate electricity for about 35 years. The next step, in 1971, was the collaboration between the National Power Corporation (NPC) and Philippine Geothermal Inc. (PGI), regarding the field’s exploration and development. The project was initiated in 1972 before regulatory standards on the environmental and social dimensions were introduced. Consequently, as a result of the lack of a relevant regulatory framework, the first years of the project were surrounded by difficulties. The unfavorable views of locals were triggered by environmental pollution, the NPC’s excessive use of governmental power and the attainment of private land for public benefit, the lack of provided benefits, the loss of hot springs and other tourist attractions, and concerns regarding the company’s hiring processes. Therefore, during the first 10 years of operation, community protests, including
71
vehicle stoning and road barricades, and disagreements with the regulating organizations and local government affected the NPC [72]. Starting in 1992, there was a shift when NPC was obliged to pay real estate taxes to the local government. Similarly, based on following legal and regulatory provisions, the local government started to collect a national wealth tax from the company while the local community and government received a monetary share, based on the electricity sales of the power plant. These revenues were given back to the community through electricity subsidization, and the financing of developments concerning energy, economy, the environment, and health. Furthermore, the company has promoted activities on the environmental issues perceived by the communities, validating the involvement of the surrounding communities in the company’s businesses relevant to local economic, health, welfare, and social aspects [72].
5.6.7
Upper Rhine Graben (Europe)
In 1988, the exploitation of deep geothermal resources for power production started in the Upper Rhine Graben, with the EGS research project in Soultz-sous-Forȇts. Since 2000, geothermal development had been prevalent in the area, with exploration licenses spanning the whole region. However, objections from the public were stated in regard to the unfavorable impacts of deep geothermal exploitation, following the seismic events that occurred between 2006 and 2007. The EGS project was halted after a report showing that Basel would encounter many small earthquakes during the whole lifespan of the power plant’s operation. In addition, there was large public resistance to the Landau power plant while there was an intensification of objections toward geothermal exploitation in the Rheinland-Pfalz region. In particular, local pressure groups hampered the advancement of geothermal resources in Germany after the Landau incidents in 2009. Therefore, the resources in the Upper Rhine Graben are not adequately exploited [73,74]. The surrounding communities are highly aware of deep geothermal energy; nevertheless, project planners have to bring together all related parties as soon as possible through a communication and dialogue approach to achieve broad approval concerning the proposed projects. During this phase, ongoing issues should be detected and resolved while positive and negative impacts should be identified and examined in parallel. In this context, a social acceptance plan, including public meetings and discussions, was adopted by the Trebur project near Groß-Gerau to ensure that the development would have the support of the local communities; the first results showed that the project enjoys the acceptance of the surrounding community [73,74].
72
PART
I Basics of geothermal power plants
5.7 Social acceptance practices performed by geothermal operators and developers When studying the social acceptance techniques employed until now, it is revealed that certain variations occur between periods of time and country categories. Talking about rising and developing economies, the initial records on geothermal social acceptance practices indicate that attention has been administered primarily on creating added benefits for the surrounding communities and the reduction of any kind of unwelcome negative effects. In these instances, the role associated with the local stakeholders has been basically to deliver advice for designing community development services and/or corporate social responsibility (CSR) activities, as introduced for instance in various cases in the Philippines [72,75,76], Indonesia [77,78], and El Salvador [66]. Having said that, Kenya has seemingly focused more on public engagement; this has been accomplished in line with the Environmental Management and Coordination Act (EMCA) created in early 2000 [79]. In this sense, the appropriate documents for Kenya illustrate the creation and execution of advisory actions connected with a variety of local stakeholder panels in which a number of local residents are also included; related examples include Menengai [80], Suswa [81], Olkaria I Units 4–5, Olkaria IV [82], and Eburru [83]. When assessing the quite small number of documents focusing on geothermal power plant social acceptancerelated activities in developed nations, it is shown that the priority has been the planning of engagement actions as well as incorporating distinctive implementation steps and stakeholder categories. One of the primary efforts was performed by Beck [84], offering helpful tips in providing information to the general public of Hawaii. The latest endeavors emphasize equally the information and consultation actions for local stakeholder bodies, as identified within the examples of ARRC/Pawsey Geothermal in Australia [85], Groß-Gerau in Germany [64], and the Upper Rhine Graben [74]. To sum up, based on the available literature on geothermal projects, activities related to social acceptance can be categorized into three pillars: (a) avoidance and reduction of unfavorable impacts, (b) generation of added benefits for surrounding communities, and (c) community engagement practices (Fig. 5.4).
Avoidance and reduction of unfavorable impacts
Community engagement practises
Generation of added benefits for surrounding communities
FIG. 5.4 The three pillars of social acceptance practices.
that can help toward this objective consist of: (a) the formulation of an environmental strategy, concentrating on the steps essential to prevent or minimize any unwelcome impacts [6,86]; (b) proper environmental controlling and designing procedures, and arrangement of activities throughout the project’s development and operation stages [66,87]; (c) built-in processes to guarantee conformity with health, safety, and environmental requirements [66]; (d) the formation of an environmental security fund, intended to be employed in situations of rehabilitation and damage reimbursement caused by the geothermal project’s development and operation [88]; along with (e) the arrangement of different environmental activities, such as reforestation of impacted areas [86]. On top of that, the recognition of cultural locations together with the development of a strategy to safeguard them can reduce the likelihood of disturbing them as a result of the project’s development and operation [81]. Additionally, immediate reimbursement for damages due to the project’s works to private or public property, such as crops, livestock, buildings, roads, and infrastructure, is equally significant. Based on Cataldi [6], in such circumstances, it is crucial for the head of the project to be flexible, to consider compensatory actions, and also to “close” the negotiations quickly, thus being able to preserve proper relations with the surrounding communities.
5.7.1 Avoiding and reducing unfavorable impacts
5.7.2 Generating added benefits for surrounding communities
One of the most significant considerations in the direction of social acceptance should be the avoidance and reduction of unfavorable side effects regarding the natural environment and the people. In line with the reported activities, actions
Generating positive impacts for local communities is possible, either by directly providing funds to local authorities, in most cases being determined by the associated regulatory framework [76,88], or by the implementation of local
Social acceptance of geothermal power plants Chapter 5
development plans. The delivery of funds for these authorities may be either in the form of a portion of the company’s earnings—standing for the exploitation rights of the local energy sources—[76,88] or a share of a levy, right, or fee related to geothermal resource development and exploitation [72]. The accumulated funds could subsidize the cost of electricity within the areas where the geothermal resource is found. As a result of the subsidy, the location can attract even more investments, resulting in more employment opportunities and economic advantages for the locals [72] as well as the initialization of development projects (infrastructure, services, etc.) for the surrounding communities [76,88]. The economic, social, and cultural advancement of communities near the project can indeed be facilitated by local development initiatives. This helps the project manager to achieve the task of supporting the communities hosting the project and acknowledging their commitment to the nation’s security and sustainability [66,81,86]. These activities can even become part of a firm’s CSR program. By doing so, the organization may enhance its reputation and stakeholder ties [77]. This eliminates the uncertainties and complications surrounding geothermal developments [66]; it can thereby obtain a “license to operate,” which can result in a number of long-term monetary and nonmonetary gains [89]. In terms of planning activities to satisfy the needs of surrounding communities, it is suggested that the company should (a) examine and report local economic, social, etc., aspects, (b) address the actions with local authorities, organizations, groups, etc., and (c) track the actions constantly, so that potential initiatives can be strengthened by collected feedback [75,82]. The subsequent activities (Fig. 5.5) can be incorporated in the framework set out above: l
l
l
l
Education advancements: construction of education facilities, development of educational services, supply schools with equipment and material (e.g., books), support local students with scholarships, and delivery of meals to students in cases where it is necessary [81,82,90]. Enhancing health and hygiene: supporting citizen access to health care by supplying medication and health services, increasing access to hospitals, and providing food to vulnerable citizen groups [81,86,89]. Environmental protection: environmental education, environmental conservation, and involvement in emergency disaster management activities (e.g., community relief in the event of floods or drought) [81,82,89]. Enhancing the local economy and entrepreneurship: education courses for developing local skill sets and knowledge on business administration and managerial issues, skills relevant to their jobs, etc. (potentially focused on particular groups such as women and young people), providing local people employment
73
Improving education Promoting culture and sports
Improving health and sanitation
Improving infrastructure
Local environment protection Strengthening the local economy
FIG. 5.5 Generating benefits for surrounding communities.
l
l
opportunities related to the project (based on the type of skills needed within the project carried out), choosing to buy local resources and utilities, providing incentives for local businesses, local production enhancement through technology transfer, offering grants for research valuable for the surrounding community (e.g., agricultural studies), and fostering diversification of the local economy in rural communities via the creation of ecotourism and aquaculture facilities capable of exploiting geothermal potential [76–78,80,83,86,89,90]. Infrastructure development: creation or enhancement of roads, multipurpose buildings, power and water supply networks, and transport services [81,89,90]. Delivering released steam or hot water to be used in municipal buildings, community centers, and other public buildings at a low price or totally free of charge [6]. Fostering cultural and athletic initiatives: planning and funding athletic and cultural activities; building sports facilities and assisting in the preservation of buildings/ parks, etc., with a view to support cultural heritage and recreation activities; and granting research funding or publication of research on crucial aspects of the involved area’s growth potential, culture, traditions, and customs [6,72,78,81,86,89].
5.7.3
Public engagement
5.7.3.1 Defining public engagement Public engagement, a term commonly used in a range of fields, has been described by the National Coordinating Centre for Public Engagement (NCCPE) [91] as “a
74
PART
I Basics of geothermal power plants
Communication
Collaboration
Participation
FIG. 5.6 Levels of community engagement activities.
two-way process, involving interaction and listening, with the goal of generating mutual benefit.” In 2005, Rowe and Frewer [92] proposed three distinct descriptions to distinguish activities previously mentioned as public participation between project developers and operators and the public, based on information flows. These descriptions include: (a) public communication, (b) public consultation, and (c) public participation (Fig. 5.6). Public communication refers to the information that is provided to the public participants (usually local authorities and societies) by the project developers and operators; it is a one-way process. Public consultation refers to the information that is given to the project developers and operators by the public, without interaction between the two sides. Public participation refers to information that is exchanged between the two sides through a dialogue process [92]. Many benefits are derived from public engagement, especially for power plant projects. Public engagement may solve public participants’ and/or project developers’ concerns at an early stage of project development by providing a communication channel among all relevant stakeholders. By building trust and understanding between the different stakeholders, efforts and costs can be reduced during the early stage of project implementation. Although there are many benefits to engaging with the public, challenges still exist. Public engagement may require time, financial cost, and sophisticated planning when high levels of opposition are expected. Furthermore, there are different stakeholders that may be affected by the project; hence, the use of several communication channels is required [93].
5.7.3.2 Review of community engagement practices Engagement practices that involve surrounding communities are of utmost significance for gaining public acceptance of a geothermal power plant, as they strengthen confidence between the business and the local community, minimize negative reactions/disputes, and raise the level of support toward the company regarding project execution. Engaging with the surrounding communities can facilitate the avoidance and reduction of undesirable effects and the provision of positive impacts, thereby enhancing procedural and distributional justice issues between the community and the company. To achieve these goals, it is crucial to adopt an extensive action plan. In the sense of coordination and cooperation
with the surrounding communities, the subsequent activities have been carried out in terms of engagement, based on a review of past geothermal project action plans: l
l
l
l
Execution of a socioeconomic study of the location under consideration during the project development’s initial stages. The study should cover topics such as municipal borders, land use and types of property, economy, natural resources, utilities, public services, transportation, cultural and historical sites, energy resources, stakeholder groups and their perspectives regarding geothermal energy, and benefits enjoyed by local communities [64,74]. Depending on the study results, the public engagement mechanism should be adjusted to the specific conditions [64]. Formation of a group of local actors with the representation of government authorities, members of surrounding communities, environmental conservation associations, representatives from the agriculture and business sectors, etc. Provide the audience with details about the activities of the company and forthcoming plans and discussion with the intention of achieving shared trust. Through such a network, a platform can be established where local communities’ environmental and social priorities can be addressed in a timely manner toward the company accountable for the project, with the aim of addressing each disputable topic and contributing to a common agreement that will support the project’s approval [80,83] [94]. This method enables local expertise, perspectives, and different needs to be incorporated, along with extensive information sharing between all parties [64]. Dialogue covering a substantial portion of the surrounding communities. Provide comprehensive information about geothermal energy, the project being established, and the advantages and disadvantages that surround it. Individuals should be given the option to explore the project’s pros and cons, pose questions, and voice their concerns to the representatives of the project [64,74,85]. Application of awareness programs aimed toward all stakeholders, namely local governmental authorities, government agencies, local people, NGOs, local groups (e.g., consumers), private companies, etc. Awareness practices ought to be carried out during the entire project preparation and execution process. The scope of the awareness activities may include the geothermal
Social acceptance of geothermal power plants Chapter 5
l
resource, project definition, potential environmental implications, initiatives, and opportunities for surrounding communities [64,95,96]. Methods that can be applied to inform the diverse types of audiences comprise project site tours, seminars, websites, brochures, media releases, an education center, a contact office, social media, creation of a demonstration site, involvement in activities, coordination of scientific meetings, and collaboration with groups with relevant interests [73,80,84,97].
l
l
75
team composed of local administration representatives, communities, organizations, etc. [88]. Reckless practices ought to be prevented, particularly at the start of a geothermal power project, as they can possibly result in the creation of a preliminary unfavorable opinion from the surrounding communities; if so, the restoration of a favorable reputation may necessitate large investments in time and effort [87]. Each one of the promises made while engaging with nearby communities must certainly be carried out in reality [88].
5.7.3.3 Guidelines of engagement practices The engagement actions recorded above ought to be ruled by certain guidelines to be able to support their effective application. Via the study of carried-out action plans concerning geothermal project development, the next guidelines have now been recognized: l
l
l
l
l
l
l
l
l
To be able to establish transparency and straightforward interaction, engagement practices should always be planned outdoors, near the surrounding community, rather than being carried out behind “closed doors” (conference rooms, office spaces, or hotels), cutting off the local community [98]. Truthful information originating from legitimate and unbiased sources should certainly be delivered to the locals in an easy-to-understand manner and tailored to the local traditions [96]. The diversity of the local people ought to be acknowledged, bearing in mind sociodemographic aspects, practical knowledge, influence, beliefs, and motivations [95]. To be able to develop an effective interconnection, centered on sincerity and reliability, all affected stakeholders should always be equally treated [88]. Each one of the topics in regard to the project, even the unfavorable ones, should be dealt with publicly; to provide transparent, efficient, and precise communication by all involved sides, a mutually agreed terminology can be introduced [73]. Throughout the allocation of beneficial and unfavorable impacts, all parties involved ought to be taken into consideration, among them those being underrepresented or perhaps not represented at all [86]. A particular individual, assigned as the spokesperson of the project, should reach out in a suitable manner to all associated parties [73]. Community stakeholders can interpret the involvement of high-positioned representatives, coming from the company’s management, within the discussion as genuineness and acceptance of accountability [86]. Organizations’ desire to operate “clear” activities should certainly be stated by tracking the project’s works by a
5.7.4
The role of public authorities
The significance of the position of public authorities on a state, regional, and/or local level must be highlighted in terms of achieving social acceptance, in parallel with the efforts of geothermal developers and operators. Public authorities should assist the process, particularly via the formulation and enforcement of effective legislative and regulatory frameworks and policies, in order to preserve and support the surrounding communities’ interests. Regulations and policies may include, for example, the allocation of a defined percentage of profits for the benefit of the region and the introduction of socioeconomic impact assessments [1]. In addition, public authorities can facilitate the process of social acceptance by engaging in the development of the necessary social infrastructure and the coordination of awareness-raising initiatives.
5.8
Closing remarks
The present work provides an overview of the aspects related to the social acceptance of geothermal power plants. Community acceptance—meaning acceptance by the local communities surrounding the project site—is the most substantial level of acceptance (in accordance with the definition provided by W€ustenhagen et al. [10]) when referring to geothermal power production projects, which are location-specific. Social acceptance should be considered a prerequisite for the smooth and unobstructed development and operation of a geothermal project; lack of acceptance could result in increased costs and efforts, setbacks, and even the cancelation of the project. Moreover, the implementation and operation of such a project can provide a variety of socioeconomic impacts; these may include environmental protection, energy security and trade balance improvement on a national scale, income and employment generation, improvement of living standards, and social cohesion on a local level. In this context, the importance of measuring the socioeconomic impacts generated by geothermal projects should be highlighted. The evaluation and assessment of these impacts
76
PART
I Basics of geothermal power plants
can assist companies in presenting to the relevant stakeholders the benefits created for the surrounding communities, thus reinforcing awareness and trust, strengthening accountability and reliability, sustaining a “license to operate,” and backing the development of a better policy and funding environment. In many cases, people perceive unfavorably the development of a geothermal power plant neighboring their community; such incidents of local resistance have been recorded worldwide throughout the years, including the Berlı´n project (El Salvador), the Lower Kilauea East Rift Zone (Hawaii), the Milos and Nisyros Islands (Greece), the Mt. Apo and Tiwi projects (Philippines), and the Upper Rhine Graben case. The range of cases indicates a variety of conditions and determinants that may lead to a lack of community acceptance. In pursuit of social acceptance, development and operation companies implement a collection of different strategies and practices, which can be summarized as (a) engaging with the local communities, (b) avoiding and reducing unfavorable impacts, and (c) generating added benefits for surrounding communities. It is worth mentioning that based on the engagement activities recorded so far, the focus has been given first of all to communication and consultation activities, whereas active participation of the local communities has not yet been a common practice. In this context, it is vital that public authorities enhance the social acceptance procedure through the formulation of regulations and policies, the development of social infrastructure, and awareness activities. As a closing remark, the importance of public engagement activities should be highlighted. Toward the achievement of social acceptance, local communities should become actively involved through the communication/consultation/collaboration process. Nevertheless, it should be noted that public engagement does not ensure unconditional acceptance by the surrounding communities; however, it is a significant step toward the prevention of local resistance.
References [1] Karytsas S, Polyzou O, Mendrinos D, Karytsas C. Towards social acceptance of geothermal energy power plants. In: Proceedings of the European Geothermal Congress 2019, Den Haag, The Netherlands, June 11–14; 2019. [2] Jobert A, Laborgne P, Mimler S. Local acceptance of wind energy: factors of success identified in French and German case studies. Energy Policy 2007;35(5):2751–60. [3] Batel S, Devine-Wright P, Tangeland T. Social acceptance of low carbon energy and associated infrastructures: a critical discussion. Energy Policy 2013;58:1–5. [4] Enevoldsen P, Sovacool BK. Examining the social acceptance of wind energy: practical guidelines for onshore wind project development in France. Renew Sust Energy Rev 2016;53:178–84.
[5] de Jesus AC. Socio-economic impacts of geothermal development. In: Proceedings of the World Geothermal Congress 1995, Florence, Italy, Pre-congress course on environmental aspects of geothermal development. IGA/CNR-International School of Geothermics; 1995. [6] Cataldi R. Social acceptance of geothermal projects: problems and costs. In: Proceedings of the European Summer School on geothermal energy applications 2001, Oradea, Romania; 2001. p. 343–51. [7] Popovski K. Political and public acceptance of geothermal energy. The United Nations University, IGC2003-Short Course, Reykjavik, Iceland; 2003. p. 31–41. [8] Friedl C, Reichl J. Realizing energy infrastructure projects—a qualitative empirical analysis of local practices to address social acceptance. Energy Policy 2016;89:184–93. [9] European Commission. Energy roadmap 2050, Communication, 885 final; 2011. [10] W€ustenhagen R, Wolsink M, B€urer MJ. Social acceptance of renewable energy innovation: an introduction to the concept. Energy Policy 2007;35:2683–91. [11] Walker G. Renewable energy and the public. Land Use Policy 1995;12(1):49–59. [12] Krohn S, Damborg S. On public attitudes towards wind power. Renew Energy 1999;16(1–4):954–60. [13] Polyzou O, Stamataki S. Geothermal energy and local societies—a NIMBY syndrome contradiction. In: Proceedings of the World Geothermal Congress 2010, Bali, Indonesia, April 25–29; 2010. [14] Wolsink M. Planning of renewables schemes: deliberative and fair decision-making on landscape issues instead of reproachful accusations of non-cooperation. Energy Policy 2007;35:2692–704. [15] Pellizzone A, Allansdottir A, De Franco R, Muttoni G, Manzella A. Geothermal energy and the public: a case study on deliberative citizens’ engagement in central Italy. Energy Policy 2017;101: 561–70. [16] Dowd AM, Boughen N, Ashworth P, Carr-Cornish S. Geothermal technology in Australia: investigating social acceptance. Energy Policy 2011;39(10):6301–7. [17] Carr-Cornish S, Romanach L. Differences in public perceptions of geothermal energy technology in Australia. Energies 2014; 7(3):1555–75. [18] Kubota H, Hondo H, Hienuki S, Kaieda H. Determining barriers to developing geothermal power generation in Japan: societal acceptance by stakeholders involved in hot springs. Energy Policy 2013;61:1079–87. [19] Malo M, Moutenet JP, Bedard K, Raymond J. Public awareness and opinion on deep geothermal energy in the context of shale gas exploration in the province of Quebec, Canada. In: Proceedings of the World Geothermal Congress 2015, Melbourne, Australia; 2015. p. 19–25. [20] Karytsas S, Theodoropoulou H. Public awareness and willingness to adopt ground source heat pumps for domestic heating and cooling. Renew Sust Energy Rev 2014;34:49–57. [21] Karytsas S. An empirical analysis on awareness and intention adoption of residential ground source heat pump systems in Greece. Energy Policy 2018;123:167–79. [22] Karytsas S, Vardopoulos I, Theodoropoulou E. Factors affecting sustainable market acceptance of residential microgeneration technologies. A two time period comparative analysis. Energies 2019; 12(17):3298.
Social acceptance of geothermal power plants Chapter 5
[23] Liu S, Shukla A, Zhang Y. Investigations on the integration and acceptability of GSHP in the UK dwellings. Build Environ 2014;82:442–9. [24] Jung N, Moula ME, Fang T, Hamdy M, Lahdelma R. Social acceptance of renewable energy technologies for buildings in the Helsinki Metropolitan Area of Finland. Renew Energy 2016;99:813–24. [25] Irmak E, Ayaz MS, Gok SG, Sahin AB. A survey on public awareness towards renewable energy in Turkey. In: 3rd international conference on renewable energy research and applications, Milwakuee, USA, October 19–22; 2014. p. 932–7. [26] Karytsas S, Theodoropoulou H. Socioeconomic and demographic factors that influence publics’ awareness on the different forms of renewable energy sources. Renew Energy 2014;71:480–5. [27] Kępi nska B, Kasztelewicz A. Public perception of geothermal energy in selected European countries. In: Proceedings World Geothermal Congress 2015, Melbourne, Australia, April 19–25; 2015. [28] Devine-Wright P. Reconsidering public attitudes and public acceptance of renewable energy technologies: a critical review. Published by the School of Environment and Development, University of Manchester, UK. Available from: http://www.sed.manchester.ac.uk/ research/beyond_nimbyism/; 2007. [29] Wolsink M. Invalid theory impedes our understanding: a critique on the persistence of the language of NIMBY. Trans Inst Br Geogr 2006;31(1):85–91. [30] Stephanides P, Chalvatzis KJ, Li X, Lettice F, Guan D, Ioannidis A, Zafirakis D, Papapostolou C. The social perspective on island energy transitions: evidence from the Aegean archipelago. Appl Energy 2019;255:113725. [31] Del Rı´o P, Burguillo M. Assessing the impact of renewable energy deployment on local sustainability: towards a theoretical framework. Renew Sust Energy Rev 2008;12(5):1325–44. [32] IRENA and CEM. The socio-economic benefits of large-scale solar and wind: an econValue report; 2014. [33] Sheikh NJ, Kocaoglu DF, Lutzenhiser L. Social and political impacts of renewable energy: literature review. Technol Forecast Soc Change 2016;108:102–10. [34] Domac J, Richards K, Risovic S. Socio-economic drivers in implementing bioenergy projects. Biomass Bioenergy 2005;28(2):97–106. [35] Vezmar S, Spajic A, Topic D, Sˇljivac D, Jozsa L. Positive and negative impacts of renewable energy sources. Int J Electr Comput Eng Syst 2014;5(2):47–55. [36] Akella AK, Saini RP, Sharma MP. Social, economical and environmental impacts of renewable energy systems. Renew Energy 2009;34(2):390–6. [37] Jaramillo-Nieves L, Del Rı´o P. Contribution of renewable energy sources to the sustainable development of islands: an overview of the literature and a research agenda. Sustainability 2010;2(3): 783–811. [38] OECD, Linking renewable energy to rural development, Executive Summary, Brief for Policy Makers, (n.d.). [39] Rogers JC, Simmons EA, Convery I, Weatherall A. Social impacts of community renewable energy projects: findings from a woodfuel case study. Energy Policy 2012;42:239–47. [40] K.W. Rapp, Addressing the social impacts of wind power development, Energy Sector Management Assistance Program, International Finance Corporation – World Bank Group, (n.d.). [41] Nelson S, Kuriakose AT. Gender and renewable energy: entry points for women’s livelihoods and employment. Climate Investment Funds; 2017.
77
[42] Karytsas S, Mendrinos D, Karytsas C. Measurement methods of socioeconomic impacts of renewable energy projects. IOP Conf Ser Earth Environ Sci 2020;410:012087. [43] Wainwright S. Measuring impact: a guide to resources. London: NCVO Publications; 2002. [44] Maas KE, Boons F. CSR as a strategic activity: value creation, integration and redistribution. In: Louche C, Idowu SO, Filho WF, editors. Innovative CSR. Sheffield: Greenleaf Publishing; 2010. p. 154–72. [45] Maas KE, Liket K. Social impact measurement: classification of methods. In: Environmental management accounting and supply chain management. The Netherlands: Springer; 2011. p. 171–202. [46] Clark C, Rosenzweig W, Long D, Olsen S. Double bottom line project report: assessing social impact in double bottom line ventures. Working Paper Series, Paper 13, Berkeley: Center for Responsible Business, University of California; 2004. [47] The New Economics Foundation. Measuring social impact: the foundation of social return on investment (SROI). London Business School; 2004. [48] Muir K, Bennett S. The compass: your guide to social impact measurement. Sydney, Australia: The Centre for Social Impact; 2014. [49] Lawrence Berkeley National Laboratory. Evaluating and quantifying the non-energy impacts of energy efficiency. EM&V Webinars; 2016. [50] Golden K, Hewitt A, McBane M. Social entrepreneurship: social impact metrics. MaRS White Paper Series; 2010. [51] Franks D. Social impact assessment of resource projects. In: Mining for development: guide to Australian practice. International Mining for Development Centre; 2012. [52] World Business Council for Sustainable Development (WBCSD), Measuring socio-economic impact: a guide for business, Business solutions for a sustainable world, (n.d.). [53] Epstein MJ, Yuthas K. Measuring and improving social impacts: a guide for nonprofits, companies and impact investors. San Francisco: Berrett-Koehler Publishers, Inc.; 2014. [54] A. Spector, E. Carey, G. Leung, K. McCarthy, K. Moynihan, N. Durakovic, Impact measurement in the clean energy sector, Global Impact Investing Network (GIIN), GIIN Network Insights, (n.d.). [55] Rinaldo H. Getting started in social impact measurement: a guide to choosing how to measure social impact. Norwich: The Guild; 2010. [56] European Union/OECD. Policy brief on social impact measurement for social enterprises: policies for social entrepreneurship. Luxembourg: Publications Office of the European Union; 2015. [57] GECES Sub-group on Impact Measurement. Proposed approaches to social impact measurement in the European Commission legislation and practice relating to EuSEFs and the EaSI. Luxembourg: Publications Office of the European Union; 2014. [58] Carrera DG, Mack A. Sustainability assessment of energy technologies via social indicators: results of a survey among European energy experts. Energy Policy 2010;38(2):1030–9. [59] Dale VH, Efroymson RA, Kline KL, Langholtz MH, Leiby PN, Oladosu GA, Davis MR, Downing ME, Hilliard MR. Indicators for assessing socioeconomic sustainability of bioenergy systems: a short list of practical measures. Ecol Indic 2013;26:87–102. [60] Maas KE. Social impact measurement: towards a guideline for managers. In: EMAN-EU 2008 conference: sustainability and corporate responsibility accounting measuring and managing business benefits; 2008. p. 75–8. [61] Dufour B. State of the art in social impact measurement: methods for work integration social enterprises measuring their impact in a public
78
[62]
[63]
[64]
[65]
[66]
[67] [68]
[69]
[70] [71] [72]
[73]
[74]
[75]
[76]
[77]
[78]
PART
I Basics of geothermal power plants
context. In: 5th EMES international research conference on social enterprise: building a scientific field to foster the social enterprise eco-system, EMES, Jun 2015, Helsinki, Finland; 2015. Available from: https://hal.archives-ouvertes.fr/hal-01458730/document. Gray R. Current developments and trends in social and environmental auditing, reporting and attestation: a review and comment. Int J Audit 2000;4(3):247–68. International Association for Impact Assessment (IAIA). Social impact assessment, http://www.iaia.org/wiki-details.php?ID¼23. [Accessed 12.2017]. Wallquist L, Holenstein M. Engaging the public on geothermal energy. In: Proceedings of the World Geothermal Congress 2015, Melbourne, Australia, April 19–24; 2015. Reith S, K€ olbel T, Schlagermann P, Pellizzone A, Allansdottir A. Public acceptance of geothermal electricity production. GEOELEC: Deliverable 4.4; 2013. Zepeda N, Rodriguez A. Socially responsible geothermal development in El Salvador. In: Proceedings of the World Geothermal Congress 2005, Antalya, Turkey, April 24–29; 2005. Boyd TL, Thomas D, Gill AT. Hawaii and geothermal – what has been happening? Geo-Heat Center Quart Bull 2002;23(3):11–21. Polyzou O. Geothermal energy – sustainable development and local societies. Doctoral dissertation, Athens: National Technical University of Athens; 2007 [In Greek]. Karytsas S, Polyzou O, Karytsas C. Social aspects of geothermal energy in Greece. In: Manzella A, Allansdottir A, Pellizzone A, editors. Geothermal energy and society. Cham: Springer; 2019. p. 123–44. Koutroupis N. Update of geothermal energy development in Greece. Geothermics 1992;21(5–6):881–90. Castro EC. Understanding public opposition to development projects: the Mt. Apo case. GeotherResourCounc Bull 1994;23(8):283–6. Camu M, Santiago R. Social development in the Philippines’ Tiwi geothermal area. In: Proceedings of the World Geothermal Congress 2000, Kyushu – Tohoku, Japan; 2000. Schwellenbach E, van Douwe A. The citizens’ initiative in deep geothermal energy. In: Proceedings of the European Geothermal Congress 2016, Strasbourg, France; 2016. van Douwe A, Stahl LK, Kreuter H. Acceptance of geothermal projects in a critical environment in the Upper Rhine Graben. In: Proceedings of the European Geothermal Congress 2016, Strasbourg, France; 2016. Meidav MZ, Calica MC, Enalpe EE. Model of social responsibility and community service: case history of Unocal/Philippine geothermal INC’s experience with the Mak-Ban and Tiwi fields. In: Proceedings of the World Geothermal Congress 1995, Florence, Italy; 1995. Anaye J, Cala CL. Geothermal energy development as a medium towards total community development: the Philippine example. In: Proceedings of the World Geothermal Congress 2005, Antalya, Turkey; 2005. p. 24–9. Slamet U, Moelyono DG. Maximizing community benefits and minimizing environmental impacts in the Gunung Salak geothermal project, Indonesia. In: Proceedings of the World Geothermal Congress 2000, Kyushu—Tohoku, Japan, May 28–June 10; 2000. Ibrahim RF, McCloskey B, Sutisna E, Pranoto R, Munaf I, Gordon R, Stevensen R. Corporate Social Responsibility implementation in the Darajat geothermal project, Garut, West Java. In: Proceedings of the World Geothermal Congress 2005, Antalya, Turkey; 2005.
[79] Ogola PF. Appraisal drilling of geothermal wells in Olkaria Domes (IV), Kenya: baseline studies and socioeconomic impacts. Geothermal Training Programme, The United Nations University, Reykjavik, Iceland; 2004. p. 267–306. [80] Manyara D, Mading P. Environmental and social considerations in geothermal development: case study Menengai, Kenya: moving towards green and clean economy. GRC Trans 2012;36:1227–32. [81] Chebet SK. Community engagement in Suswa geothermal prospect. GRC Trans 2013;37:779–84. [82] Barasa PJ. Public participation in the implementation of 280MW geothermal power projects at Olkaria in Naivasha sub-county, Naukuru County, Kenya. In: Proceedings of the World Geothermal Congress 2015, Melbourne, Australia; 2015. [83] Barasa PJ, Mathenge RW. Stakeholder engagement through participatory research: a case study of Eburru geothermal wellhead generator in Nakuru County, Kenya. GRC Trans 2015;39:233–8. [84] Beck AG. Dealing with controversial facts: geothermal public information in Hawai’i. GRC Trans 1990;14(1):583–8. [85] Carr-Cornish S, Huddlestone-Holmes C, Ashworth P. The ARRC/ Pawsey geothermal demonstration project: an example of how to engage the community. In: Proceedings of the 2011 Australian Geothermal Energy Conference, Melbourne, Australia; 2011. [86] Wetang’ula GN. Public participation in environmental and socioeconomic considerations for proposed 2.5 MW pilot Eburru geothermal power project, Kenya. In: Proceedings of the World Geothermal Congress 2010, Bali, Indonesia, April 25–29; 2010. [87] ENGINE—Enhanced Geothermal Network of Europe: Increasing policy makers’ awareness and public acceptance, WP5, Deliverable 38, (n.d.). [88] de Jesus AC. Social issues raised and measures adopted in Philippine geothermal projects. In: Proceedings of the World Geothermal Congress 2005, Antalya, Turkey; 2005. [89] Musembi R. Corporate Social Responsibility (CSR) in geothermal development: the case of the Geothermal Development Company (GDC), Kenya. In: Proceedings of the ARGEO-C3, 3rd East African Rift Geothermal Conference, Djibouti; 2010. p. 516–21. [90] Kurgat IK, Omwenga J. Impact of power generation project on the livelihoods of adjacent communities in Kenya: a case study of Menengai geothermal power project. Int J Sci Res Publ 2016;6 (10):610–24. [91] National Co-ordinating Centre for Public Engagement (NCCPE). What is public engagement? https://www.publicengagement.ac.uk/ about-engagement/what-public-engagement. [Accessed 15 January 2020]. [92] Rowe G, Frewer LJ. A typology of public engagement mechanisms. Sci Technol Hum Values 2005;30(2):251–90. [93] Open EI. Public involvement in renewable energy and infrastructure project development, https://openei.org/wiki/RAPID/Best_Practices/ Public_Involvement_in_Renewable_Energy_and_Infrastructure_ Project_Development. [Accessed 16 January 2020]. [94] Thompson R. An assessment of the socio-economic and marine environmental impacts associated with the St. Kitts and Nevis geothermal energy project, Graduate Project, NS. Halifax: Dalhousie University; 2014. [95] Leucht M, K€olbel T, Laborgne P, Khomenko N. The role of societal acceptance in renewable energy innovations breakthrough in the case of deep geothermal technology. In: Proceedings of the World Geothermal Congress 2010, Bali, Indonesia, April 25–29; 2010.
Social acceptance of geothermal power plants Chapter 5
[96] Shoedarto RM, Aries FR, Irawan D, Perdana F, Arisbaya I, Indrawan B. Raising public acceptance of geothermal utilization through direct application in Indonesia. In: Proceedings of the 41st Stanford Workshop on geothermal reservoir engineering, SGP-TR-209; 2016. [97] Carr-Cornish S, Romanach L. Exploring community views toward geothermal energy technology in Australia. Pullenvale, Australia: CSIRO; 2012.
79
[98] Barasa PJ. Integration of environmental management system in monitoring of environmental and social aspects associated with operation of Olkaria II geothermal power plant in Naivasha sub-county, Nakuru County, Kenya. In: Proceedings of the 40th workshop on geothermal reservoir engineering. Stanford, California: Stanford University; 2015.
Chapter 6
Single- and double-flash cycles for geothermal power plants Saeid Mohammadzadeh Bina, Hikari Fujii, and Shunsuke Tsuya Graduate School of Engineering and Resource Science, Akita University, Akita, Japan
Nomenclature c C_ ex _ Ex h m_ P s T W_
specific cost ($/GJ) cost rate ($/s) specific exergy (kJ/ kg) exergy rate (kW) specific enthalpy (kJ/ kg) mass flow rate (kg/s) pressure (bar) specific entropy (kJ/kg/K) temperature (°C) power (kW) investment ($/h)
Z
h w
Greek letters energy efficiency exergy efficiency
Subscripts 0 1,2,3, … des en eq eva gf HEX M O Obj Petro pro s tur wf
6.1
dead state cycle state destructive energy equipment evaporator GeoFluid heat exchanger maintenance operation objective petroleum production isentropic turbine working fluid
Introduction
Besides the advantages of the other renewable resources such as wind or solar energy, geothermal energy wins
the competition because it is the most reliable source of energy. Geothermal is independent of meteorological conditions such as daylight or wind speed, which are necessary for solar and wind energy, and can produce electricity continuously if the reservoir is well assessed in an area. Therefore, regardless of weather and climate conditions, the existence of the reservoir in a region is a sufficient and adequate issue to guarantee constant energy production during the day and even for many years with a negligible power decrement. Geothermal energy can be exploited for indirect applications, especially in low-temperature sources, to use its thermal energy for various purposes such as snow melting, fish farming, etc. [1,2]. The utilization of geothermal energy to generate electricity is also taken into consideration by energy policymakers in the country’s energy mix. In this application type, which is categorized and known as an indirect application of geothermal energy, the heat source can be turned into electricity indirectly using various energy conversion systems such as flash cycles, organic Rankine cycles (ORC), etc. According to the latest report in 2015, the geothermal installed capacity in the world was about 12,635 MWe. This number is estimated to reach around 21,433 MWe in 2020, considering the existing projects worldwide [3]. As was mentioned above, various energy conversion technologies can be chosen and installed on the geothermal reservoir to convert the geofluid energy into electricity. These technologies are selected based on either the reservoir thermodynamic specifications such as temperature and pressure or its chemistry characteristics [4]. However, sometimes other parameters such as the country’s economic or political condition make engineers change their decision and choose an alternative scenario. For example, in Iran, a 5 MW power plant was installed on the Sabalan geothermal field in the northwest of the country while a potential of 55 MW was estimated by previous feasibility studies. This field, which is the only active geothermal field in the country, cannot fully be operated because of economic
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00002-0 Copyright © 2021 Elsevier Inc. All rights reserved.
83
84
PART
II Thermodynamic analysis of geothermal power plants
barriers that cause difficulties in importing the required facilities into the country for a larger power plant. Geothermal power generation currently is based on four common technologies: direct dry steam, flash cycles, binary cycles, and a hybrid cycle that is combined with a Rankine cycle or hybridized with different heat sources such as concentrating solar power. Fig. 6.1 illustrates the percentage of produced energy based on the common power plant classification [3]. As was discussed above, the geofluid heat content plays a vital role in defining the type of energy conversion technology to be installed at a power plant for each reservoir. Although numerous innovative and conceptual cycles have recently been introduced to be used in geothermal power plants, flash cycles are still the most common methods to extract power in geothermal resources. These cycles are appropriate options to generate power from moderate to high temperature hydrothermal-type resources. They are almost similar to the dry steam plants while the steam to the turbine is indirectly obtained through a flashing process (separation). In this chapter, single- and double-flash cycles, which utilize geothermal fluid, are first introduced. In order to analyze the potential of power generation, these cycles were modeled using the thermodynamic equations with the help of EES (Engineering Equation Solver) software. Afterward, an optimization study and parametric analysis were carried out to obtain the optimal flash pressure as a crucial parameter in flash cycles and consequently to maximize the net power output as an objective function. During the optimization process, the steam quality at the turbine outlet, which is another critical parameter in the flash cycles, is restricted and limited to be not less than 85%, which is important to avoid erosion problems in the turbine.
Therefore, the main aim of this part is to maximize the cycle performance, obtaining the best flash pressure and turbine back pressure while obeying the steam quality in the turbine’s outlet at the same time. The result of the optimization process was presented using three-dimensional (3D) graphs comprising net power variation versus flash pressures and turbine outlet quality as they have interrelated effects with each other. Second, the exergy study was conducted for both cycles calculating the destruction rate in each set of equipment. This graph, which is known as the Grassman graph, is drawn side by side for the cycles to show the differences of the cycles. It assists researchers in observing the effect of the secondary flash stage on the geothermal power plants. Moreover, the specific exergy costing method was applied for cycles to calculate the production cost and the total cost of the power plants. In order to compare the cycles, the thermodynamic efficiencies (energy and exergy) and economic performances of each cycle were evaluated at their optimum operating condition. At last, the environmental benefits of using these cycles were compared with nonrenewable resource power plants for generating the equivalent energy.
6.2
System description
Single- and double-flash cycles that use geofluid as a heat source were proposed and designed to be modeled in this research. Fig. 6.2 shows the schematic of the assumed cycles in this study. The number of each state refers to the numbers that have been used for modeling in the software. In the single flash (Fig. 6.1A), the exploited geothermal fluid from the wells (state 1) with high temperature and pressure goes directly to the flashing chamber, where its pressure drops and it is separated into steam and water 182 (26) (brine). This step is necessary because only steam should 181 (2) enter the turbine, as brine causes scaling and erosion problems in the turbine’s blades. The separated steam from the geofluid (state 2) drives the turbine and generates electricity. In this step, the high-pressure and high-temperature 1790 (286) Back Pressure steam experiences a real expansion process passing the Binary turbine (state 3). This cycle finishes easily when both the separated brine from the flashing process (state 4) as Double Flash 5079 (286) well as the condensed exhaust steam from the turbine (state 2544 (68) Dry Steam 5) are reinjected into the reservoir. In the condenser, the Single Flash output of the turbine cools by transferring the heat to the Triple Flash ambient. 2863 (63) The double-flash cycle (Fig. 6.1B) that can be considered as a developed version of the single-flash cycle, works based on the same concept with a minor difference. In this cycle, the resulting brine from the first flash FIG. 6.1 Installed capacity in MWe and number of units for each energy chamber (state 4), which is known as high-pressure conversion technology. flashing (HP), is sent for further flashing in the secondary
Single- and double-flash cycles for geothermal power plants Chapter 6
FIG. 6.2 Process diagram of (A) single- and (B) double-flash cycles for geothermal power plants [4].
85
2 Generator Separator
Turbine Condenser
1
3 4 5
(A)
2
Generator Turbine
3
6 7
4
Turbine
1 8 5
9
(B)
or low-pressure flashing (LP) chamber. This benefits the heat content of the geofluid as much as possible before reinjection to the reservoir and gets more steam (state 6) and directly influences the net power increment. This extra steam is mixed with the outlet steam of the turbine (state 3) and directly goes to the low-pressure turbine (state 4). Then, similarly, the expanded steam after cooling at the condenser (state 9) and secondary flash brine (state 5) is collected and reinjected into the reservoir through a reinjection well. The temperature-entropy (T-s) of both the single- and the double-flash system is drawn based on the above-mentioned thermodynamic processes and combined together in Fig. 6.3. In this figure, the colored area presents the processes of the single-flash system and the dashed lines with numbers define the double-flash system.
6.3 6.3.1
Analysis Energy analysis
Fluid enters each component of the power plant with some known thermodynamic parameters such as temperature and pressure. This section aims to predict other unknown parameters such as enthalpy or entropy, which are required to fully model the cycle by applying energy balance and governing equations for each component. According to the first law of thermodynamics, the steady-state form of energy rate balance for each component of the system can be written as follows: X X _ Þin _ Þout ðmh (6.1) Q_ W_ ¼ ðmh According to the thermodynamic basics, in the phase change process in the separator, the state moves along the
86
PART
II Thermodynamic analysis of geothermal power plants
400
1
350
4
Temperature (°C)
300
2
250
Single flash
6
5
200 150
9
100
3
Double flash
7
8
50 0 0
1
2
3
4
5
6
7
8
9
Entropy (kJ/kg) FIG. 6.3 Temperature-entropy (T-s) diagram for the single- and double-flash systems.
isotherm lines in the pressure-enthalpy diagram. Therefore, during this process, the temperature and pressure of the stream remain constant while its enthalpy and quality change into saturated steam and water. The governing equations for the separator are as follows: T2 ¼ T4
(6.2)
P2 ¼ P 4
(6.3)
where the subscript numbers refer to those numbers shown in Fig. 6.2 for both cycles. Moreover, it is assumed that the steam entering the turbine and the brine injecting into the reservoir are purely saturated steam (x2 ¼ x6 ¼ 1) and saturated water (x4 ¼ x5 ¼ 0), respectively. Therefore, other parameters such as enthalpy and entropy can be determined for these states using flashing pressure (P2 and P6 in double-flash cycle) and their quality (x). It should be noted that the flashing pressures at this level are just assumed as a tentative value, and later, the optimum pressures will be obtained for a maximum net power output balancing the steam mass flow rate and its enthalpy. The mass flow rate of steam (m_ 2 ) and brine (m_ 5 ) as well as a steam fraction (x1) can be calculated by using the calculated quality and the total mass flow rate of the geofluid. x1 ¼
m_ 2 h1 h5 ¼ m_ 1 h2 h5
m5 ¼ m1 m2
(6.4) (6.5)
And similarly, the same equations can be used for the secondary flash to obtain it’s steam (m6) and brine (m5) mass flow rates. The ideal enthalpy of state 3 (h3s) can be determined using the turbine outlet pressure (P3) and its entropy equal to x2, assuming an ideal isentropic expansion process in the
turbine. Afterwards and given that the turbine process is not isentropic in reality and entropy increases during the expansion process, therefore, the actual enthalpy can be calculated by defining an isentropic efficiency (iso) in a single and double flash using Eqs. (6.6), (6.7), respectively. iso ¼
h 2 h3 h2 h3s
(6.6)
iso ¼
h 7 h8 h7 h8s
(6.7)
The isentropic efficiency of turbines in a geothermal power plant can range from 81% to 85% [5]. Finally, the cycles will be fully modeled by modeling the condenser as a simple phase-changing process while the pressure and temperature keep constant. Thus, the power of the turbine, W_ tur (kW) for single- and double-flash cycles, can be calculated as: W_ tur, single ¼ m_ 2 ðh2 h3 Þ W_ tur, double ¼ m_ 2 ðh2 h3 Þ + m_ 7 ðh7 h8 Þ
(6.8) (6.9)
where h is the specific enthalpy (kJ/kg), and m_ is the mass flow rate (kg/s) of steam passing the turbine. The net electrical power output of the cycle, W_ net (kW), is calculated by subtracting the parasitic power consumed by the pump from W_ tur . Thus, the thermal efficiency is evaluated for the mentioned cycle as follows [5]: en ¼
W_ net W_ tur W_ parasitic ¼ Q_ in m_ 1 hgf hbrine
(6.10)
where W_ net expresses the net power output (actual turbine work deducing other parasitic loads) and Q_ in is the input energy rate from the geofluid.
Single- and double-flash cycles for geothermal power plants Chapter 6
6.3.2
Exergy analysis
Exergy is the maximum work when a stream transforms from its initial condition to the environmental condition, and it is defined by P0 and T0 [6, 7]. By neglecting chemical and potential exergy and assuming only thermal interaction with the environment, specific flow exergy can be expressed as follows [8, 9]: ex ¼ ðh h0 Þ T0 ðs s0 Þ
(6.11)
where h and s are the specific enthalpy and specific entropy of the geothermal fluid at the specific state, and the 0 subscripts refer to its properties at the dead state. The exergy flow rate can be calculated using the mass flow rate as follows: _ ¼ m_ ðexÞ Ex
(6.12)
In general, the exergy efficiency of a geothermal plant is the net electrical power output divided by the total input exergy flow rate of the geofluid [9, 10]. Thus, the exergy efficiency can be calculated as follows: ’plant ¼
6.3.3
W_ net _ brine _ Exgf Ex
(6.13)
Exergoeconomic calculation
Exergoeconomic analysis is done by applying the following steps: (1) Applying the exergy cost for each stream and equipment in the systems that is equal to the stream’s exergy rate multiplied by its cost. (2) Calculation of the levelized cost rate for each component based on its purchased equipment cost (PEC). (3) Calculation of the exergy cost rate of each stream. (4) Applying the exergy cost rate balance for the whole power plant and calculating the total and production cost rate. Step 1: In order to calculate the specific cost of each stream, a cost balance for every component should be stated, as shown in Eq. (6.14), which is based on the concept of inlet and outlet exergy from the component. This equation implies that the sum of the inlet exergy cost rate (C_ in ) and the capital cost rate (Z_ eq ) at each component is equal to the outlet exergy cost rate (C_ out ) and the power produced by that component (C_ w ). X X (6.14) C_ in + Z_ eq ¼ C_ out + C_ w Step 2: To calculate the capital cost rate for each component (Z_ eq ), first, the capital cost of each component (Zeq) should
87
be calculated as a function of its technical specifications presented in the literature (Table 6.1). And then the sum of the operating and maintenance cost rate can be considered as follows [13]:
Zeq ∅ CRF Z_ eq + Z_ O&M ¼ h 3600
(6.15)
where ∅ and h denote the maintenance factor and operating hours, respectively. The Capital Recovery Factor (CRF), which converts a present value into a stream of equal annual payments over a specific time and at a specified discount (or interest) rate, can be estimated as a function of i (interest rate) and n (interest rate) [14]: CRF ¼
ið1 + iÞn ð1 + iÞn 1
(6.16)
The required parameters such as the interest rate in Eq. (6.16) for economic modeling which are dependent on a country’s economic condition, are listed in Table 6.2. Step 3: The inlet cost rate (C_ in ) for the stream i can be calculated by multiplying the specific cost of the ith stream (ci) with the exergy rate of the same stream i [17]. _ ¼ c ðm_ exÞ C_ ¼ c Ex
(6.17)
In general, if the number of streams that exit the component is greater than 1, the appropriate auxiliary equations are required to complete the calculation for the auxiliary equations. All cost balance equations and auxiliary equations applied to the system are summarized in Table 6.3. Furthermore, the specific cost of the incoming stream of water can be regarded as zero with no significant error. These linear equations can be solved to obtain the cost flow rates and the unit exergetic costs associated with each stream of the plant. According to the literature, the unit cost of the geofluid (cgf) as the inlet fuel in this system is assumed to be equal to 1.3 $/GJ [18]. Step 4: Finally, a cost rate balance was set for the overall system by applying economic analysis as a function of the fuel (geofluid) cost rate and the sum of operating and maintenance costs of the components, as follows: X (6.18) Z_ eq + Z_ O&M C_ total ¼ C_ fuel + eq
where C_ total and C_ fuel are the total and fuel cost rate of the power plant, respectively. C_ total ¼ c_ pro W_ net _ gf C_ fuel ¼ cgf Ex where c_ pro is the production cost rate of the plant for each kWh, and cgf is the geofluid cost as the system’s input fuel.
88
PART
II Thermodynamic analysis of geothermal power plants
TABLE 6.1 Capital cost rate functions for each component. Component
Capital cost (Z)
Ref.
Separator
sep sep sep 0 Csep ¼ 576:1 397 B1 + B2 FM FP , sep Csep 0 sep sep sep logCsep ¼ K1 + K2 log Vsep + K3 (logVsep)2
[11]
tur 0 Ctur ¼ 576:1 397 FBM Ctur
[12]
8 9 Psep + 1 Dsep > > > > > > + 0:00315
> > > > > : ;
Turbine
2 log W_ tur + K3tur log W_ tur HEX HEX 0 CHEX ¼ 576:1 + B2HEX FM FP , HEX CHEX 397 B1 0 logCtur
Condenser
¼ K1tur
+ K2tur
[11]
logC0HEX ¼ KHEX + KHEX log AHEX + KHEX (logAHEX)2 1 2 3 logFP,
HEX + CHEX log PHEX + CHEX (logPHEX)2 HEX ¼ C1 2 3
Constants Bsep 1 Bsep 2 Ksep 1 Ksep 2 Ksep 3 Fsep M
2.25
Ktur 1
1.82
Ktur 2
1.4974
Ktur 3
0.4485
Ftur BM
0.1074
BHEX 1
1.0
BHEX 2
TABLE 6.2 Assumed economic parameters for the system. Economic parameters
Value, unit
Ref.
Annual operational hours, h
7000 h
[12]
Lifetime of the systems
20 years
[15]
Annual interest rate, i
14%
[16]
Maintenance factor, ∅
1.06
[16]
6.3.4
Validation
In order to validate our developed models, the results of the single- and double-flash models were compared with the previous studies. The input data from Ref. [5] were used in our models, and the results of these two studies were compared to conclude the accuracy of the models. Table 6.4 lists the relevant outputs in both studies. It can be seen that there is an acceptable agreement between the models in these studies.
6.4 6.4.1
Optimization Single flash optimization
Thermoeconomic balance equations are applied for a proposed single-flash cycle that utilizes the geofluid as a heat source. Then,
2.7051
FHEX M
1.0
1.4398
KHEX 1
4.3274
0.1776
KHEX 2
0.3030
3.5
KHEX 3
0.1634
1.63
CHEX 1
0
1.66
CHEX ¼ CHEX 2 2
0
the cycle was optimized to find the best separator pressure resulting the highest energy output, while obeys the steam quality limitation in the turbine outlet (which is higher than 0.86). As is clear, with increasing the flash pressure, it helps to have a higher specific turbine power due to the increment of the turbine inlet enthalpy. On the other hand, a higher flash pressure causes a lower amount of separated steam during the flashing process and consequently results in lower net power output. Therefore, there is an optimum separator pressure that balances between mass flow rate and the enthalpy of steam as it has a reverse effect on these parameters. Fig. 6.4 presents the variation of system parameters with different flash pressures. In addition, Fig. 6.5 illustrates the relationship between separator pressure (P2), net power in kW (Ẇ net), and steam quality at the turbine’s outlet (x3) as another crucial design parameter in the flash systems. The vertical and horizontal axes correspond to the steam quality and separator quality, respectively, and the counters represent the net power outputs. Additionally, the horizontal dotted red line in the figure shows the minimum steam quality, which is considered a limitation during the optimization process. The optimization was conducted using a direct search method in EES software. This method finds the proper parameters inside the considered boundary, which results in the desired value of the objective function. The objective function and also the subjected constraints are given below:
Single- and double-flash cycles for geothermal power plants Chapter 6
89
TABLE 6.3 Exergoeconomic balance equations for each component in the flash geothermal power system. Cycle
Component
Exergoeconomic balance
Axillary equation
Single flash
Geothermal resource
_ 1 C_ 1 ¼ cgf Ex
cgf ¼ 1.3
Separator
C_ 1 + Z_ sep ¼ C_ 2 + C_ 4
c2 ¼ c4
Turbine
c2 ¼ c4, cw ¼ variable
Condenser
C_ 2 + Z_ tur ¼ C_ 3 + C_ W C_ 7 C_ 6 + Z_ cond ¼ C_ 3 + C_ 5
Separator HP
C_ 1 + Z_ sep ¼ C_ 2 + C_ 4
c1 ¼ 1.3
Turbine HP
C_ 2 + Z_ tur ¼ C_ 3 + C_ W
c2 ¼ c5
Separator LP
C_ 4 + Z_ sep ¼ C_ 5 + C_ 6
c2 ¼ c5, cw ¼ variable
Turbine LP
C_ 7 + Z_ tur ¼ C_ 8 + C_ W C_ 11 C_ 10 + Z_ cond ¼ C_ 9 + C_ 8
c6, c7 ¼ 0
Double flash
Condenser
TABLE 6.4 Comparison of various parameters calculated in this study for single- and double-flash cycles with the corresponding results presented in Ref. [5] (the values inside parenthesis). Single flash
Double flash HP
LP
Steam quality at separation
16.4% (16.4%)
14.1% (14.1%)
12.4% (11.7%)
Mass flow rate after separation
82.06 kg/s (82.05 kg/s)
70.7 kg/s (70.6 kg/s)
53.49 kg/s (50 kg/s)
Total net power produced by plant
31,400 kW (31,105 kW)
59,363 kW (49,781 kW)
Overall thermal efficiency
6.22% (7.32%)
11.1% (9.96%)
Overall exergy law efficiency
30.75% (32.73%)
44.59% (43.35%)
Obj : Maximize W_ net ðP2 Þ 1 < P2 ðbarÞ P1
A minimum of 1 bar (ambient pressure) was considered for flashing pressure (P2) in order to prevent air sucking into the flash chamber.
6.4.2
Double-flash optimization
Similarly, the same method in Section 6.4.1 was applied for the double-flash system to find the optimum condition. The
c6, c7 ¼ 0
first and second flash pressures of the double-flash cycle were selected in order to maximize the net power output. This optimization problem can be formulated as: 8 < Obj : Maximize W_ net ðP2 , P6 Þ 1 < P2 ðbarÞ P1 : 1 < P6 ðbarÞ P2 Fig. 6.6 shows the variation of net power output with different flash pressures in a 3D graph. As can be seen, there is an optimum point with specific values of primary flash pressure (P2) and secondary flashing pressure (P6), which results in the highest net power output. It is worth mentioning that this optimum point of the pressures was found while the quality of steam in the turbines (x3 and x8) was kept higher than the standard and desired value (85%).
6.5
Experimental data
The results from well testing and exploration drilling in the Sabalan geothermal field (Table 6.5) were used as input values for the modeled cycles. Because there are nine wells, the total average enthalpy and total mass flow rate were calculated using Eq. (6.18) equal to 1027 kJ/kg and 285 kg/s, respectively. X X ðm_ i hi Þ ðm_ i hi Þ ¼ X (6.19) hin ¼ m_ total m_ i
6.6
Results and discussions
Tables 6.6 and 6.7 illustrate the optimum thermoeconomic parameter values at different stages of the cycles after the optimization process. The listed data in these tables are calculated using the balance equations given in Table 6.2. The state numbers in these tables refer to Fig. 6.2. The reference conditions for exergy analysis are 18°C and atmospheric
90
PART
II Thermodynamic analysis of geothermal power plants
Mass flow rate
Enthalpy
Net power output 2780 75
28k
65 60
26k 55 25k 50
Mass flow rate (kg/s)
Net power (W)
27k
2740
2720
Enthalpy of steam (j/kg)
2760
70
2700 45
24k
40
2680
23k 35 2
4
6
8
10
P2 (bar)
FIG. 6.4 The variation of steam mass flow rate, turbine inlet enthalpy, and net power output with different flash pressures (P2).
FIG. 6.5 The relationship between x3, separator pressure (P2), and net power.
pressure of 1 bar. The average exergetic cost of geothermal water as a fuel input in the overall system is calculated to be 1.3 $/GJ. Moreover, the unit cost of power produced (cwt) for both cycles is estimated as 0.8 cents/kWh and 0.9 cents/kWh, respectively. The results of exergoeconomic analysis as well as the exergy destruction rates at each component for the systems are given in Tables 6.8 and 6.9. The purchase costs for each component are correlated based on the introduced equations
in Table 6.4. Additionally, the economic parameters in Table 6.4 as well as the equations in Section 6.3.2 were used in order to estimate the total and production costs. According to the exergy destruction rates in Table 6.8 for different stages of the single-flash power plant, 22% of the total exergy is returned to the reservoir through the reinjection well. These values for the separator, condenser, and turbine are equal to 22%, 18%, and 13% of the total input exergy, respectively. Lastly, the remaining 41% is
Single- and double-flash cycles for geothermal power plants Chapter 6
91
W _ tot
40410 40249 40087 39925 39764 39602 39481 39319 39158 38996 38834
6]
P [2
P[
]
FIG. 6.6 3D plot of the effect of flash pressures (x, y-axis) on the total power output (z-axis).
the fraction of the exergy of the input geofluid that the power plant converts to electricity. Similarly, Table 6.9 shows the exergy destruction at different stages for the double-flash system. The sum of exergy destruction in both separators shows a 4% improvement compared to the single flash. Additionally, secondary flashing and the use of first flash brine energy caused a considerable improvement of the exergy destruction (9%) in the reinjection. In fact, the double-flash system wastes less exergy compared to the single system, and this phenomenon shows the advantage of these systems. Furthermore, exergy destruction in the turbine and condenser were calculated as 7% and 15%, respectively. With approximately a 10% increment, the remaining 51% is converted to electricity. These values are used to draw the exergy diagram for both systems (Fig. 6.7). This method is more helpful in observing the effect of the double-flash process on the exergy destruction of each component in a system. Additionally, according to the economic calculation of the systems in Tables 6.8 and 6.9, the levelized values of the total cost and the production cost rate for single flash based on a plant lifetime of 20 years were calculated to be 3.061 M$/year and 4.104 $/GJ, respectively. These values for double flash were calculated as 4.148 M$/year and 4.39 $/GJ. Even though
double flash produces higher output, the production cost in single flash results in lower value due to the high purchased equipment cost of the double-flash systems. The total purchase equipment costs of the components are calculated to be $3,272,000 and $9,971,000 for the single- and double-flash systems, respectively. Furthermore, the levelized operation and maintenance cost rates (OMC) for the single- and double-flash systems are estimated at 51.32 $/h and 100.87 $/h (based on a 20-year lifetime of the plant), respectively. The corresponding exergetic cost rates for the singleand double-flash systems are estimated to be 0.0921 $/s.
6.6.1
Environmental benefits
In this section, the environmental benefits of the two different cycles are presented. The cycles are compared with fossil fuel power plants based on the reduction of required fuel and also the emission of greenhouse gases. Because the geothermal power plants do not need fuel, their benefits can be examined by calculating the annual saved petroleum and annual reduction of CO2, as follows [20]: W_ net (6.20) Mpetro ¼ apetro hannual 1000
92
PART
II Thermodynamic analysis of geothermal power plants
TABLE 6.5 Real data of geofluid available from the Sabalan geothermal wells. Well name
Temperature (°C)
Mass flow rate (kg/s)
Specific enthalpy (kJ/kg)
NWS-1
242.96
29
1051.46
a
NWS-3
149.0
0
627.86
NWS-4
230.2
27
991.76
NWS-5D
240.42
56
1039.31
NWS-6D
237.85
46
1027.12
NWS-7D
236.75
46
1021.9
NWS-9D
230.53
37
994.5
NWS-10D
242.8
44
1050.69
a
NWS-11
174.54
0
739.14
Total
237.9
285
1027
a
These wells are dry without geofluid production [19].
W_ net MCO2 ¼ aCO2 hannual 1000
(6.21)
where apetro gives the average required petroleum for producing 1 kWh of electricity and aCO2 is the amount of CO2 released from fossil fuel power plants for producing L 1 kWh. These values are assumed to be apetro ¼ 0:266 kWh and 0:849 kg aCO2 ¼ kWh [11]. The annual saved fossil fuel (Mpetro) and annual reduced CO2 emissions (MCO2) are presented in Table 6.10 for different objective functions. As can be seen, using the flash geothermal power system instead of fossil fuel power plants can have a great advantage for the environment by reducing 25% fuel consumption and CO2 emissions for the equivalent amount of energy production.
Furthermore, to consider the condition of Iran’s energy production, the available data of energy consumption and corresponding released environmental pollutants for a 1 year electricity supply has been used. According to the latest energy balance sheet of Iran in 2015, the total energy generation from power plants, excluding renewable resources, was calculated as 62,896 MW. Details of various air pollutants from operating fossil fuel power plants in the country are presented in Table 6.11. Moreover, the values for single- and double-flash systems show the calculated amount of environmental pollution reduction caused by using these systems in Iran’s condition. In this study, the analyses of the single- and double-flash geothermal power plants were conducted using the thermodynamic and economic concepts for Sabalan, Iran. The reservoir fluid enthalpy and mass flow rate were calculated to equal to 1027 kJ/kg and 285 kg/s, respectively, based on actual data from all existing wells. The EES software was used to model the plant by thermodynamic and economic equations. To make a comparison, the cycles were optimized in terms of achievable higher net power output. The pressure of flashing was defined as a restriction to obtain the higher performance of the systems. The optimum pressure for separation was determined as 3 bar for the single-flash system. The optimum pressure value for the high-pressure separator and the low-pressure separator was calculated to be 7.63 bar and 1.06 bar, respectively. With these optimum pressures, the overall energy and exergy efficiencies and also net power output for the single flash was estimated as 16.26%, 40.06%, and 28,838 kW, respectively. These values for the double-flash cycle were calculated as 17.73%, 50.86%, and 36,055 kW, respectively. The total available exergy from the production wells, according to the Sabalan geothermal fluid, was a constant for the system and was calculated as 70,847 kW. The economic calculation was done for the cycles’ 20year lifetime considering the constant average exergetic cost for geofluid (input fuel) as 1.3 $/GJ. To achieve an economic analysis for each stream and component, the capital
TABLE 6.6 Energetic and exergoeconomic analyses results of the single-flash states. State
m_ (kg/s)
T (°C)
P (bar)
h (kJ/kg)
E_ x ðkWÞ
C_ ð }24=sÞ
c ($/GJ)
1
285
237.9
32.2
1027
70,847
0.0921
1.3
2
64.24
127.4
3
2717
42,757
0.06839
1.6
3
64.24
40
0.073
2275
9717
0.01554
1.6
4
64.24
40
0.073
167.5
206.9
0.01576
1.6
5
220.8
127.4
3
535.4
15,339
0.02463
1.6
Single- and double-flash cycles for geothermal power plants Chapter 6
93
TABLE 6.7 Energetic and exergoeconomic analyses results of the double-flash states. State
m_ (kg/s)
T (°C)
P (bar)
h (kJ/kg)
E_ x ðkWÞ
C_ ð$=sÞ
c ($/GJ)
1
285
237.9
32.2
1027
70,847
0.0921
1.3
2
43.67
168.5
7.63
2766
36,002
0.05033
1.39
3
43.67
101.2
1.06
2480
21,798
0.03047
1.39
4
74.5
40
1.06
2562
38,544
0.05389
1.39
5
74.5
40
0.0738
2246
11,113
0.00401
1.39
6
74.5
168.5
0.0738
167.5
239.8
0.00033
1.39
7
241.3
101.3
7.63
712.3
30,083
0.04206
1.39
8
241.3
101.3
1.06
712.3
25,601
0.03574
1.39
9
30.84
101.3
1.06
2678
16,746
0.02341
1.39
10
210.5
101.3
1.06
424.4
8856
0.01238
1.39
TABLE 6.8 Purchased equipment cost and cost rates of single-flash cycle components. Component
Exdes (kW)
TABLE 6.9 Purchased equipment cost and cost rates of the double-flash cycle.
Cc (×103 $)
ZOMC ($/h)
Component
E_ xdes (kW)
Cc (×103 $)
ZOMC ($/h)
Separator HP
4762 (8%)
43.960
1.03
Separator LP
4481 (6%)
8.090
0.19
Turbine HP
1718 (2%)
2131
50.4
Turbine LP
3866 (5%)
2049
48.39
Condenser
10,873.2 (15%)
36.161
0.855
Reinjection (brine)
9095 (13%)
Net power (kW)
36,055 (51%)
Total input exergy (kW)
66,086 (100%) 4.39
Separator
12,751 (18%)
140.46
3.32
Turbine
4202 (6%)
1998
47.23
Condenser
9510.1 (13%)
32.43
0.7668
Reinjection (brine)
15,339 (22%)
Net power (kW)
28,838 (41%)
Total input exergy (kW)
70,847 (100%)
Produce cost ($/GJ)
4.104
Total cost (M$/ year)
3.061
Produce cost ($/GJ)
16.26%
Total cost (M$/ year)
4.148
Energy efficiency
17.73%
Exergy efficiency
40.06%
Thermal efficiency Exergy efficiency
50.89%
cost of all equipment was correlated by specific equations from the literature. It was estimated that the levelized values of the total cost and the production cost rate of double flash are higher than single flash by 1.087 M$/year and 0.29 $/GJ, respectively. The main reason is because of higher capital
investment due to two more components (low-pressure turbine and flash chamber) in the double-flash cycle. Consequently, it will require higher operation and maintenance fees. In the double-flash cycle, the capital investment and operation and maintenance costs were calculated as
94
PART
II Thermodynamic analysis of geothermal power plants
FIG. 6.7 Exergy flow diagram of single-flash (right side) and double-flash cycles (left side).
TABLE 6.10 Fuel-saving and CO2 emission reduction for cycles in different optimized systems. No.
Cycle
Single flash
Double flash
Improvment status
1
Fuel saving (L)
53,696.36
67,134.41
25%
2
CO2 reduction (kg)
171,384.23
214,274.87
TABLE 6.11 Different pollutants from power plants for energy production based on Iranian power plant operation in 2015 (tonnes) [21]. NOx
SO2
SO3
CO
SPM
CO2
CH4
N2O
Conventional
651,610
627,934
458.6
144,660
31,105
17,744,913
4243
654
Single flash
29,876.7
28,791.14
21.03
6632.74
1426.18
813,614.6
194.544
29.99
Double flash
37,353.64
35,996.41
26.29
8292.66
1783.10
1,017,229.88
243.23
37.49
6.7 M$ and 49.55 $/h, respectively, which were more than the single-flash cycle. Thus, the production cost was estimated to be higher in the double-flash cycle (4.104 $/GJ vs 4.39 $/GJ for single flash).
6.7
Closing remarks
l
l
The following were the main conclusions from this study: l
l
There are optimal values for separator pressures in cycles to maximize the net power output, and the pressures higher or lower than those pressure values deliver less power output. The optimum pressure for single flash was calculated as 3 bar and for double flash, thee optimum pressures were 7.63 bar and 1.06 bar at the primary and secondary separators, respectively. Among the considered cycles, the double-flash cycle had the maximum generated electrical power, which generates 25% more power compared to the single flash at the optimum state (maximum net produced electrical power).
l
l
l
The highest exergy destruction in the single-flash cycle belongs to reinjection (22%), followed by a separator (18%) and condenser (13%). In the double flash, the exergy destruction of these components is improved to 13%, 14%, and 15%, respectively. It was concluded that reinjection had the highest exergy destruction rate in the single-flash cycle while this stream can be used for either direct applications or running a bottoming cycle (an ORC). According to the input energy and exergy into the cycle and also their electricity production, the results revealed that the double-flash system has higher energy and exergy efficiencies. Among the considered cycles, the minimum unit cost of produced electrical power belongs to the single-flash cycle, which is 9.1% less than the double-flash cycle. Eventhoughthedouble-flashcycleproduceshigherenergy, thetotalcostofthesingle-flashcyclecosts1.087 M$/yearless than the double-flash system due to the higher capital investmentforthiscycle($3,272,000vs$9,971,000).
Single- and double-flash cycles for geothermal power plants Chapter 6
References [1] Mohammadzadeh Bina S, Jalilinasrabady S, Fujii H. Thermoeconomic evaluation of various bottoming ORCs for geothermal power pljant, determination of optimum cycle for Sabalan power plant exhaust. Geothermics 2017;70:181–91. https://doi.org/10.1016/j. geothermics.2017.06.007. [2] Mohammadzadeh Bina S, Jalilinasrabady S, Fujii H, Pambudi NA. Classification of geothermal resources in Indonesia by applying exergy concept. Renew Sust Energ Rev 2018;93:499–506. https:// doi.org/10.1016/j.rser.2018.05.018. [3] Bertani R. Geothermal power generation in the world 2010–2014 update report. In: Proceedings world geothermal congress 2015, Melbourne, Australia; 2015. [4] Mohammadzadeh Bina S, Jalilinasrabady S, Fujii H. Exergoeconomic analysis and optimization of single and double flash cycles for Sabalan geothermal power plant. Geothermics 2018;72:74–82. https://doi.org/10.1016/j.geothermics.2017.10.013. [5] Jalilinasrabady S, Itoi R, Valdimarsson P, Saevarsdottir G, Fujii H. Flash cycle optimization of Sabalan geothermal power plant employing exergy concept. Geothermics 2012;43:75–82. https://doi. org/10.1016/j.geothermics.2012.02.003. [6] Rosen MA, Dincer I. Effect of varying dead-state properties on energy and exergy analyses of thermal systems. Int J Therml Sci 2004;43 (2):121–33. https://doi.org/10.1016/j.ijthermalsci.2003.05.004. [7] Rosen MA, Dincer I. Exergy analysis of waste emissions. Int J Energy Res 1999;23(13):1153–63. https://doi.org/10.1002/(SICI)1099-114X (19991025)23:133.0.CO;2-Y. [8] Preißinger M, Heberle F, Br€uggemann D. Advanced organic rankine cycle for geothermal application. Int J Low-Carbon Technol 2013;8 (suppl 1):i62–8. https://doi.org/10.1093/ijlct/ctt021. [9] Dincer I, Rosen MA. Exergy, environment and sustainable development. Elsevier; 2007. [10] DiPippo R. Second law assessment of binary plants generating power from low-temperature geothermal fluids. Geothermics 2004;33 (5):565–86. https://doi.org/10.1016/j.geothermics.2003.10.003. [11] Zhao Y, Wang J. Exergoeconomic analysis and optimization of a flash-binary geothermal power system. Appl Energy 2016;179:159– 70. https://doi.org/10.1016/j.apenergy.2016.06.108.
95
[12] El-Emam RS, Dincer I. Exergy and exergoeconomic analyses and optimization of geothermal organic Rankine cycle. Appl Therm Eng 2013;59(1–2):435–44. https://doi.org/10.1016/j.applthermaleng. 2013.06.005. [13] Bejan A, Tsatsaronis G, Moran MJ. Thermal design and optimization. Wiley; 1995. [14] Khatib H. Economic evaluation of project in the electricity supply industry. In: Power and energy series, 44. The Institution of Engineering and Technology; 2008. [15] Heberle F, Schifflechner C, Br€uggemann D. Life cycle assessment of organic Rankine cycles for geothermal power generation considering low-GWP working fluids. Geothermics 2016;64:392–400. https://doi. org/10.1016/j.geothermics.2016.06.010. [16] Mosaffa AH, Farshi LG, Infante Ferreira CA, Rosen MA. Exergoeconomic and environmental analyses of CO2/NH3 cascade refrigeration systems equipped with different types of flash tank intercoolers. Energy Convers Manag 2016;117:442–53. https://doi.org/10.1016/j. enconman.2016.03.053. [17] Lazzaretto A, Tsatsaronis G. SPECO: a systematic and general methodology for calculating efficiencies and costs in thermal systems. Energy 2006;31(8–9):1257–89. https://doi.org/10.1016/j.energy. 2005.03.011. [18] Akbari M, Mahmoudi SMS, Yari M, Rosen MA. Energy and exergy analyses of a new combined cycle for producing electricity and desalinated water using geothermal energy. Sustainability 2014;6 (4):1796–820. https://doi.org/10.3390/su6041796. [19] Noorollahi Y, Mohammadzadeh Bina S, Yousefi H. Simulation of power production from dry geothermal well using down-hole heat exchanger in Sabalan field, Northwest Iran. Nat Resour Res 2016;25(2):227–39. https://doi.org/10.1007/s11053-015-9270-3. [20] Shengjun Z, Huaixin W, Tao G. Performance comparison and parametric optimization of subcritical organic Rankine cycle (ORC) and transcritical power cycle system for low-temperature geothermal power generation. Appl Energy 2011;88(8):2740–54. https://doi.org/ 10.1016/j.apenergy.2011.02.034. [21] Iran Energy Efficiency Organization (IEEO - SABA). Energy balance sheet, http://www.saba.org.ir/saba_content/media/image/2015/09/ 7811_orig.pdf; 2013.
Chapter 7
Dry steam power plant: Thermodynamic analysis and system improvement Muhammad Aziza and Firman Bagja Juangsab a
Institute of Industrial Science, The University of Tokyo, Tokyo, Japan b Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung,
Bandung, Indonesia
Nomenclature COND CP CT CV CWP _ Ex G h IRR LCOE LRVP m_ MR NCG PR RP s SSC ST W_ net WV x xNCG
condenser compressor cooling tower control valve cooling water pump exergy rate (kW) generator specific enthalpy (kJ/kg) internal rate of return the lowest levelized cost of energy liquid-ring vacuum pump mass flow rate (kg/s) moisture removal noncondensable gas particulate removal reinjection pump specific entropy (kJ/kg K) specific steam consumption steam turbine net generated power (kW) wellhead valve steam quality fractions of NCG in mixture fluid
Subscripts dt in out s t
dry turbine inlet outlet steam turbine
Greek letters h c
efficiency exergy efficiency
7.1
Introduction
Among several types of geothermal power plants, the dry steam power plant has become the simplest and most matured technology since the first power plant was built in 1904 in Larderello, Italy. At that time, Prince Piero Ginori Conti had developed a small steam engine to generate electricity for five light bulbs utilizing the steam that came from the ground. Afterward, he further increased the capacity to about 40 kW to supply the electricity for the factory. Since then, many types of geothermal power plants have been studied and developed, yet the dry steam cycle is still utilized for many geothermal power plants. Basically, the dry steam cycle power plant utilizes direct steam that is produced in the Earth’s interior. However, it has a minimum requirement of steam characteristics to be converted into electricity, which makes it nonapplicable for all geothermal wells. Currently, about 12% of geothermal power plants adopt the dry steam cycle, covering about 27% of the total installed capacity of geothermal power plants [1]. At present, large dry steam reservoirs are found in two different places: Larderello (Italy) and The Geyser (Northern California, United States). Moreover, there are only two known underground steam resources in the United States, including the previously mentioned The Geysers as well as Yellowstone National Park in Wyoming. However, due to Yellowstone’s rich thermal features as well as strong environmental concerns to protect the functioning of hydrothermal systems in the park, the development of power plants in the Yellowstone area is prohibited [2]. Other resources have been discovered in other areas, including Matsukawa (Japan), Poihipi (New Zealand), and Kamojang, Darajat, and Ulubelu (Indonesia).
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00015-9 Copyright © 2021 Elsevier Inc. All rights reserved.
97
98
PART II Thermodynamic analysis of geothermal power plants
The steam, which can be saturated or superheated vapor, is first separated and purified from the solid particles, then sequentially injected into the turbine/generator unit for electricity generation. The dry steam cycle is able to provide a larger amount of thermal energy because of the higher vapor quality of the working fluid [3]. Compared to other types of geothermal power plants, including flash and binary, a single unit of the dry steam power plant is able to generate electricity with an average capacity of about 45 MW, compared to 15 MW for the flash cycle and 40 MW for the binary cycle [4]. In fact, the unit capacity ranges from 8 to 140 MW [5]. It has been projected that only 5% of geothermal plants with a temperature higher than 200°C are employed as dry steam plants [1, 6]. Due to the geothermal reservoir’s limitations on resources and characteristics, their further characteristic analysis is critical for the development of dry steam geothermal power plants. The environmental impacts of the dry steam geothermal power plant are considered minimal, as the extracted fluid basically only contains steam with no minerals (brine). The content of noncondensable gases (NCGs) is basically very low and the sulfur (if it exists) can be separated to be sold or appropriately disposed. In addition, the geothermal fluid is reinjected back into the ground, resulting in a cycle of fluid. In this chapter, the potential of the dry steam cycle is initially described, followed by an explanation about the typical system configuration in terms of thermodynamics and technoeconomic assessments. Further explanation about the function of the main components is followed with recent developments related to the dry steam cycle in order to enhance the system as well as improve the total energy efficiency.
7.2
180
180
0 C
C
(A)
0
(B) 180
180
0
C
C
(C)
0
(D) 180
Dry steam potential
Direct steam utilization has some constraints of geothermal source properties, such as very high steam quality (>90wt %) [1]. A sufficiently high-temperature steam source is preferable for the dry steam power plant in order to achieve an optimum electrical generation system. Steam source temperature mostly depends on the intrinsic properties of each geothermal source. The reservoir of dry steam generally consists of porous rock featuring fissures or fractures that are filled with steam. The steam possibly originates from the magma-correlated activities, such as the evolution of vapor from the magma chamber to the molten rocks [7]. In addition, the rainwater that percolates through the fractures and encounters high-temperature rocks could also be the source of dry steam [6]. Generally, a deeper well has a higher temperature of the steam source. Therefore, determining the optimum point between steam source temperature and the cost of geothermal well drilling is a critical step in dry steam power plant utilization. Fig. 7.1 shows an example of temperature maps for different depths in Korea, indicating an increase of temperature at a deeper location from the Earth’s
C
(E)
0
(F)
FIG. 7.1 Temperature maps for different depths at (A) 1 km, (B) 2 km, (C) 3 km, (D) 4 km, and (E) 5 km in South Korea, which indicate an increase of temperature at a deeper location from the Earth’s crust. In addition, (F) shows the area with a temperature higher than 150°C at a depth of 5 km [8].
crust [8]. A similar map depicting the correlation between the depth and temperature also could be found in [9] for the United States. Therefore, in order to obtain the required specification of steam for the dry steam cycle, the well depth for a plant may be different from others. In order to maintain the continuity of the dry steam, a dry steam reservoir should fulfill several characteristics [1],
Dry steam power plant Chapter 7
including: (a) the heat source must be relatively close to the surface (less than 5 km) in order to sufficiently heat and evaporate the water, (b) the geological structure above the steam reservoir must be sufficiently permeable to permit the steam to pass through to the surface for a long period of time, (c) the fissures and fractures within the reservoir are adequately interconnected to circulate the fluid inside the reservoir, (d) the reservoir and its surrounding rock should have lateral permeability to avoid reservoir flooding by the groundwater, and (e) there must be a self-sealing mechanism caused by mineral precipitation to create a sufficient impermeability. Furthermore, the well depth should be considered in evaluating the geothermal reservoir for the dry steam cycle power plant. The fluid, which contains steam and gases as well as small solid particles, flows into the well from the reservoir with a relatively constant enthalpy. However, the power generation system is located on the ground, which requires the fluid to be flown up to the ground. There will be a loss in energy (enthalpy) due to an increase of potential energy, friction with the well/piping, and heat loss to the surrounding environment. Typically, the enthalpy decreases up to 50–100 kJ/kg compared to that of the reservoir [10]. More than 100 years since the first plant was installed, the dry steam power plant is still one of the main options for geothermal power generation. According to DiPippo [1], the dry steam power plant is the second largest after the single-flash power plant. This number is currently changing as there are some new geothermal power plants that are being constructed. In the United States, dry steam power plants supply a significant portion of 40% of geothermal electricity production. Table 7.1 shows the capacity of dry steam geothermal power plant in each adopting country, covering currently operating, decommissioned, and planned capacities. Some countries have adopted the dry steam cycle, including Italy, the United States, New Zealand, and Indonesia. Italy is the country recognized as the pioneer in adopting and commercializing the dry steam geothermal power plant. The United States is currently the largest country adopting the dry steam
99
geothermal plant, with a total capacity of about 1.5 GW in operation and another 95 MW under construction. Indonesia is further developing its geothermal resources, with dry steam geothermal plants in operation totaling 460 MW with an additional 170 MW to be constructed. In addition, Table 7.2 shows a detailed list of geothermal power plants that are employing the dry steam cycle. The oldest plant was built in 1904 at Larderello in Tuscany, Italy. The geothermal fluid, which is a mixture of vapor and liquid, was extracted from a well with a depth of about 300 m and a temperature of about 150°C [13]. The steam extracted from the well was utilized to generate electricity through the reciprocating steam engine developed by Prince Piero Ginori Conti [14]. In 1905, the capacity was improved to about 20 kW, and in 1920, there were two units of 20 kW engines that could supply all the required electricity for his boric acid factory. The development continued, and finally in 1916, the total capacity reached about 2.75 MW of electricity. The electricity was distributed to the surrounding areas, including Volterra and Pomarance. Ginori Conti adopted the removal system for mineral-laden liquid entrained from the well. Therefore, the corrosion problem could be avoided. The system consists of a cylindrical tank in which the steam was flown initially and the liquid was trapped. This system is quite similar to the final moisture removal adopted in currently operating dry steam plants [14].
7.3
Conversion technology
7.3.1 System structure The words “dry steam” in the dry steam cycle or power plant refer to the condition of the geothermal fluid at the wellhead as well as at the turbine inlet. The steam, which is the working fluid, usually has a temperature ranging from 160°C to 300°C. Therefore, the steam is basically dry saturated or superheated with pressure above the atmospheric pressure. The dry steam power plant has a typically simple system configuration, with a smaller plant
TABLE 7.1 Capacity of the dry steam geothermal power plant in each adopting country, covering currently operating, decommissioned, and planned capacities Country
Total installed capacity (MW)
Decommissioned capacity (MW)
Operating capacity (MW)
Planned capacity (MW)
Italy
1235
440
795
0
United States
2167
488
1584
95
Indonesia
630
0
460
170
New Zealand
110
0
55
55
Japan
24
0
24
0
100
PART II Thermodynamic analysis of geothermal power plants
TABLE 7.2 Detailed geothermal power plants employing the dry steam cycle [11, 12] Year
Country
Plant name
Unit no.
Installed cap (MW)
Geothermal field
Status
1913
Italy
Larderello 1
Test
0.3
Larderello
DEC
1914
Italy
Test/Lago
1
0.2
Larderello
DEC
1916
Italy
Larderello 1
1
3.5
Larderello
DEC
1916
Italy
Larderello 1
2
3.5
Larderello
DEC
1916
Italy
Larderello 1
3
3.5
Larderello
DEC
1926
Italy
Castelnuovo/Serrazzano
1
0.6
Larderello
DEC
1927
Italy
Castelnuovo
1a
0.8
Larderello
DEC
1927
Italy
Castelnuovo
1b
0.8
Larderello
DEC
1930
Italy
Larderello 1
4
3.5
Larderello
DEC
1939
Italy
Larderello 2
1
10
Larderello
DEC
1939
Italy
Larderello 2
2
10
Larderello
DEC
1939
Italy
Larderello 2
3
10
Larderello
DEC
1939
Italy
Larderello 2
4
10
Larderello
DEC
1939
Italy
Larderello 2
5
10
Larderello
DEC
1939
Italy
Larderello 2
6
10
Larderello
DEC
1940
Italy
Castelnuovo
1c
10
Larderello
DEC
1940
Italy
Castelnuovo
1d
10
Larderello
DEC
1940
Italy
Castelnuovo
1e
10
Larderello
DEC
1940
Italy
Castelnuovo
1f
10
Larderello
DEC
1940
Italy
Sasso Pisano
1a
3.5
Larderello
DEC
1940
Italy
Serrazzano
1a
4.2
Larderello
DEC
1940
Italy
Serrazzano
1b
4.2
Larderello
DEC
1946
Italy
Castelnuovo
1
11
Larderello
DEC
1946
Italy
Castelnuovo
2
11
Larderello
DEC
1950
Italy
Larderello 3
1
24
Larderello
DEC
1950
Italy
Larderello 3
2
26
Larderello
DEC
1950
Italy
Larderello 3
3
24
Larderello
DEC
1950
Italy
Larderello 3
4
24
Larderello
DEC
1957
Italy
Serrazzano
1
12.5
Larderello
DEC
1957
Italy
Serrazzano
2
12.5
Larderello
DEC
1958
Italy
Monterotondo
1
12.5
Larderello
DEC
1958
Italy
Sasso Pisano
1
12.5
Larderello
DEC
1958
Italy
Sasso Pisano
2
3.2
Larderello
DEC
1960
Italy
Lago
1
6.5
Larderello
DEC
1960
Italy
Lago
2
12.5
Larderello
DEC
1960
Italy
Lago
3
14.5
Larderello
DEC
1960
United States
The Geysers
1
11
CA-The Geysers
DEC
1963
United States
The Geysers
2
13
CA-The Geysers
DEC
Dry steam power plant Chapter 7
101
TABLE 7.2 Detailed geothermal power plants employing the dry steam cycle [11, 12]—cont’d Year
Country
Plant name
Unit no.
Installed cap (MW)
Geothermal field
Status
1966
Japan
Matsukawa
1
23.5
Iwate
OPR
1967
United States
The Geysers
3
27
CA-The Geysers
DEC
1968
United States
The Geysers
4
27
CA-The Geysers
DEC
1969
Italy
Gabbro
1
15
Larderello
DEC
1970
United States
McCabe
1
53
CA-The Geysers
OPR
1970
United States
McCabe
2
53
CA-The Geysers
OPR
1972
United States
Ridgeline
1
53
CA-The Geysers
OPR
1972
United States
Ridgeline
2
53
CA-The Geysers
OPR
1973
United States
Fumarole
9
53
CA-The Geysers
DEC
1973
United States
Fumarole
10
106
CA-The Geysers
DEC
1975
United States
Eagle Rock
1
110
CA-The Geysers
OPR
1975
Italy
Serrazzano
3
15
Larderello
DEC
1979
United States
Cobb Creek
1
110
CA-The Geysers
OPR
1979
Italy
Radicondoli
1
15
Travale-Radicondoli
DEC
1979
Italy
Radicondoli
2
15
Travale-Radicondoli
DEC
1979
United States
The Geysers
15
59
CA-The Geysers
DEC
1980
United States
Big Geyser
1
97
CA-The Geysers
OPR
1980
Italy
San Martino
1
9
Larderello
DEC
1980
United States
Sulfur Spring
1
109
CA-The Geysers
OPR
1981
Italy
Lagoni Rossi 3
1
8
Larderello
DEC
1983
Indonesia
Kamojang
1
30
Java-Kamojang
OPR
1983
Italy
La Leccia
1
8
Larderello
DEC
1983
United States
NCPA I
1
55
CA-The Geysers
OPR
1983
United States
NCPA I
2
55
CA-The Geysers
OPR
1983
United States
Socrates
1
113
CA-The Geysers
OPR
1983
United States
Sonoma
1
72
CA-The Geysers
OPR
1984
United States
Calistoga
1
80
CA-The Geysers
OPR
1985
United States
Bottle Rock I
1
55
CA-The Geysers
DEC
1985
United States
Grant
1
113
CA-The Geysers
OPR
1985
United States
Lake View
1
113
CA-The Geysers
OPR
1985
United States
NCPA II
3
55
CA-The Geysers
OPR
1985
United States
Quicksilver
1
113
CA-The Geysers
OPR
1985
Italy
San Martino
2
20
Larderello
DEC
1986
United States
NCPA II
4
55
CA-The Geysers
OPR
1986
Italy
Rancia
1
20
Travale-Radicondoli
OPR
1987
Indonesia
Kamojang
2
55
Java-Kamojang
OPR
1987
Indonesia
Kamojang
3
55
Java-Kamojang
OPR
1987
Italy
Pianacce
1
20
Travale-Radicondoli
OPR Continued
102
PART II Thermodynamic analysis of geothermal power plants
TABLE 7.2 Detailed geothermal power plants employing the dry steam cycle [11, 12]—cont’d Year
Country
Plant name
Unit no.
Installed cap (MW)
Geothermal field
Status
1988
United States
Bear Canyon
1
10
CA-The Geysers
OPR
1988
United States
CCPA
1
65
CA-The Geysers
DEC
1988
United States
CCPA
2
65
CA-The Geysers
DEC
1988
Italy
Rancia 2
1
20
Travale-Radicondoli
OPR
1988
United States
West Ford Flat
1
13.5
CA-The Geysers
OPR
1988
United States
West Ford Flat
2
13.5
CA-The Geysers
OPR
1989
United States
Aidlin
1
10
CA-The Geysers
OPR
1989
United States
Aidlin
2
10
CA-The Geysers
OPR
1989
United States
Bear Canyon
2
10
CA-The Geysers
OPR
1990
United States
Cove Fort
DS2
7
UT-Cove Fort
DEC
1991
Japan
Takenoyu
2
0
Kumamoto
DEC
1991
Italy
Valle Secolo
1
60
Larderello
OPR
1991
Italy
Valle Secolo
2
60
Larderello
OPR
1994
Italy
Cornia 2
1
20
Larderello
OPR
1994
Indonesia
Darajat
1
60
Java-Darajat
OPR
1995
Italy
Farinello
1
60
Larderello
OPR
1996
New Zealand
Poihipi
1
55
Wairakei
OPR
1996
Italy
Le Prata
1
20
Larderello
OPR
1996
Italy
Nuova Sasso
1
20
Larderello
OPR
1997
Italy
Carboli 2
1
20
Larderello
OPR
1997
Italy
Monteverdi 1
1
20
Larderello
OPR
1997
Italy
Monteverdi 2
1
20
Larderello
OPR
1997
Italy
Selva
1
20
Larderello
OPR
1998
Italy
Carboli 1
1
20
Larderello
OPR
1999
Indonesia
Darajat
2
90
Java-Darajat
OPR
2000
Italy
Nuova Castelnuovo
1
15
Larderello
OPR
2000
Italy
Travale 3
1
20
Travale-Radicondoli
OPR
2002
Italy
Nuova Gabbro
1
20
Larderello
OPR
2002
Italy
Nuova Lago
1
10
Larderello
OPR
2002
Italy
Nuova Molinetto
1
20
Larderello
OPR
2002
Italy
Nuova Monterotondo
1
10
Larderello
OPR
2002
Italy
Nuova Radicondoli
1
40
Travale-Radicondoli
OPR
2002
Italy
Nuova Serrazzano
1
60
Larderello
OPR
2002
Italy
Sesta
1
20
Larderello
OPR
2002
Italy
Travale 4
1
40
Travale-Radicondoli
OPR
2005
Italy
Nuova Larderello
1
20
Larderello
OPR
2005
Italy
Nuova San Martino
1
40
Larderello
OPR
2007
United States
Bottle Rock II
1
55
CA-The Geysers
OPR
Dry steam power plant Chapter 7
103
TABLE 7.2 Detailed geothermal power plants employing the dry steam cycle [11, 12]—cont’d Year
Country
Plant name
Unit no.
Installed cap (MW)
Geothermal field
Status
2007
Indonesia
Kamojang
4
60
Java-Kamojang
OPR
2008
Indonesia
Darajat
3
110
Java-Darajat
OPR
2009
Italy
Nuova Lagoni Rossi
1
20
Larderello
OPR
2009
Italy
Sasso 2
1
20
Larderello
OPR
2010
Italy
Chiusdino 1
1
20
Travale-Radicondoli
OPR
2011
Italy
Nuova Radicondoli
2
20
Travale-Radicondoli
OPR
United States
Buckeye
1
30
CA-The Geysers
PLN
Indonesia
Darajat
4
110
Java-Darajat
PLN
New Zealand
Geotherm
1
55
Wairakei
PLN
Indonesia
Kamojang
5
60
Java-Kamojang
PLN
United States
WGP
1
35
CA-The Geysers
PLN
United States
Wildhorse
1
30
CA-The Geysers
PLN
DEC, decommissioned; OPR, in operation; PLN, planned.
apparatus and cost compared to other types of geothermal power plants [15]. Because its dominant working fluid is in the vapor phase, the system only requires a few steam cleaning treatments before the steam is fed into the turbine. In this cycle, as the system is directly dependent on the source (geothermal well), there is almost no operating parameter required to be optimized throughout the system. From the general structure of the plant, the dry steam power plant can be categorized into two categories: noncondensing and condensing power plants. The geothermal power plant employing the dry steam noncondensing cycle (direct noncondensing cycle) is the simplest and economically the cheapest technology to convert geothermal energy to electricity. In this cycle, the steam from the well is just simply fed and passed through the turbine, and the exhausted steam from the turbine is abandoned to the atmosphere. Therefore, there is no condenser after the turbine. Generally, about 15–25 kg of steam is required to generate 1 kWh of electricity. However, the dry steam from almost all the geothermal sources contains NCGs of about 2%–10%. Hence, a removal system for these NCGs is required. There are some options that can be adopted for removal, including a two-stage ejector, a vacuum pump, and a turbocharger. In a plant employing a dry steam condensing system, the vapor exhausted from the turbine is not released directly to the atmosphere, but it goes to the condenser where its temperature is reduced to about 35–45°C. The condensate is then reinjected into the ground through a reinjection well. By condensing the used steam, the utilization of steam
becomes more efficient while the environmental risk due to pollution can be avoided. However, due to the addition of these facilities, the capital and maintenance costs of the plants increase. The simplified system configuration of a dry steam power plant with a condensing system is shown in Fig. 7.2. In the dry steam plant, the working fluid throughout the system includes the water vapor (>90 wt%), a small amount of water liquid, and a small amount of NCGs. In general, the system can be divided into three continuous modules: steam cleaning, power generation, and cooling and condensate systems. In the steam cleaning module, geothermal fluid is extracted from a production well that is equipped with a wellhead valve (WV). A dry steam system typically requires a particulate remover (PR) to remove and separate large solid particulates and a moisture remover (MR) to remove condensate and small solid particulates. The system configuration in this steam cleaning module may vary in each plant, depending on the characteristics of the particulate and moisture content in the geothermal fluid. The electricity production from the dry steam plant can be represented by using the temperature-entropy diagram as a thermodynamic cycle, as shown in Fig. 7.3 [16]. The separation, carried out using PR and MR, results in high vapor quality, which is further expanded (1 ! 2) though the turbine for power generation. Afterward, the heat is rejected in the condenser using the cooling water (2 ! 3). The power generation module typically consists of a steam turbine (ST), a generator (G), and a condenser
104
PART II Thermodynamic analysis of geothermal power plants
FIG. 7.2 Simplified system of the dry steam power plant. COND, condenser; CP, compressor; CT, cooling tower; CV, control valve; CWP, cooling water pump; G, generator; MR, moisture remover; PR, particulate remover; RP, reinjection pump; ST, steam turbine; WV, wellhead valve.
FIG. 7.3 Temperature-entropy diagram of dry steam geothermal power plant.
(COND), as shown in Fig. 7.2. In order to maintain the vacuum pressure in COND, steam ejectors are typically used, although vacuum pumps may also be required. The turbine basically operates in wet condition. Therefore, to correct the isentropic efficiency value, the Baumann rule is adopted [3]. The last module is cooling and condensing, which is responsible for heat release and emission treatment of the exhaust fluids before the fluid is pumped back to the reinjection well (for liquid) or released to the atmosphere (for gases). The latter may be different for each plant, depending on the content of NCGs in the geothermal fluids. In several cases, additional chemical treatment plants may be required to remove the hydrogen sulfide before the exhaust gas is released to the atmosphere. A water-cooled condenser is typically employed in the dry steam power plant while the heat in the cooling water is released in the cooling tower (CT).
7.3.2 Example of system and heat balance: Case study of the Kamojang power plant The Kamojang geothermal power plant is the oldest geothermal power plant in Indonesia; it began with Unit 1 in 1975 that had a capacity of 30 MW. The capacity of the plant was extended by the installation of Units 2 and 3 in 1987 with a total capacity of 120 MW. Furthermore, in 2008, Unit 4 with a capacity of 60 MW was constructed, resulting in a total power plant capacity of 200 MW. Fig. 7.4 shows the diagram of the heat balance of the Kamojang Unit 2 power plant, which employs the dry steam cycle. This unit has a power generation capacity of 55 MW, with the steam temperature at the turbine inlet about 162°C. The steam from the well is separated in a mist eliminator or separator. A direct contact condenser is applied in this power plant due to its economic performance [18]; therefore, the hot steam is directly mixed with the cooling
FIG. 7.4 System diagram and heat balance of the Kamojang Unit 2 power plant [17].
106
PART II Thermodynamic analysis of geothermal power plants
water. In addition, an intercondenser is also installed to condensate the steam produced from the first ejector utilized for drawing out NCGs from the main condenser. Furthermore, an aftercondenser is also employed to condensate the steam through a second-stage ejector for further treatment of NCGs. For the cooling tower, a mechanically induced draft type of tower is employed, and a fan at the discharge side is used to draw air through the tower. In this cooling tower, air and cold water are used simultaneously to drop the temperature of water. The NCGs exhausted from the well together with steam consist of 92.2%–95.4% CO2, 1.5%–4.3% H2S, 0.05%–0.2% CH4 and H2, and 0.01%–0.12% N2, He, Ar, and Ne [17]. As NCG is dominated significantly by CO2, the enthalpy of NCG can be approximated as the enthalpy of CO2 itself.
(W_ net ) to the inlet exergy rate of the geothermal steam _ in ). (Ex u ¼
There are several ways to represent and evaluate the performance of a geothermal power plant, including this dry steam cycle. The fundamental and earliest method to evaluate the performance is by using specific steam consumption (SSC) [19]. The SSC is determined as the amount of steam from the geothermal well that is required to produce 1 kWh of electricity. In other words, it can be described as the flow rate of the geothermal steam (m_ s ) consumed to produce 1 kW of electricity. SSC ¼
m_ s W_ net
(7.1)
where W_ net is the net generated power (kW) and m_ s is the mass flow rate (kg/s) of the geothermal fluids. Although SSC is simple and practical to evaluate system performance, it cannot be used directly to compare performances among plants. This is due to the different conditions, including temperature, pressure, NCG amount, and the salinity of the geothermal well. Moreover, similar to any other power generation system, the amount of energy that can be converted into electricity by the dry steam power plant is first limited by the second law of thermodynamics. Further limitations come from heat losses and pressure drops. The ratio of the generated electricity to the energy of geothermal fluids is then defined as the system efficiency (system), as shown in the following equation. system ¼
W_ net 100 m_ s h
(7.2)
where m_ s and h are the mass flow rate (kg/s) and enthalpy (kJ/kg) of the geothermal fluids, respectively. Furthermore, another way to measure the performance of a geothermal plant is by using the utilization efficiency (u), which is defined as the ratio of the net generated power
(7.3)
_ in is calculated as the change of stream enthalpy of the Ex steam deducted by the irreversibility due to the entropy generation (kW). _ in ¼ m_ s exs ¼ m_ s fðhs h0 Þ T0 ðss s0 Þg Ex
(7.4)
where exs is the specific exergy of the geothermal steam (kJ/kg), h is the specific enthalpy (kJ/kg), and s is the specific entropy (kJ/kgK). In addition, the exergy efficiency (csystem) of the dry steam cycle can be calculated as [3] csystem ¼
7.3.3 System performance
W_ net _ in Ex
system wt 100 exs
(7.5)
where wt is the specific work obtained by the turbine per unit of mass (kJ/kg). From an exergy analysis, the irreversibility in this dry steam plant is dominated by the turbine, condenser, and cooler [20]. According to the thermodynamic analysis conducted by Fallah et al. [20], compared to other types of geothermal plants, the dry steam geothermal plant has the highest energy and exergy efficiencies. This leads to the excellent thermoeconomic performance of this cycle, due to the highest internal rate of return (IRR) and the lowest levelized cost of energy (LCOE) [20]. There are several factors that affect the efficiency of a dry steam power plant. The geothermal fluid that is extracted from the production well passes through a series of processes and equipment on its way to the power generation module. During this time, the geothermal fluid loses energy that is not converted into electricity. In a liquiddominated system, which produces two-phase geothermal fluid, there is a significant energy loss during the separation process of steam and water because only steam is used for power generation. In the case of the dry steam system, vapor-dominated geothermal fluid allows less heat loss during the vapor-water separation, which results in higher system efficiency. Another factor that plays a major role in determining system efficiency is NCG. Depending on the reservoir characteristics, geothermal fluid typically contains NCG to 15 wt% [10]. A direct effect of NCG on energy efficiency occurs in the turbine, where it decreases the specific expansion work [21, 22]. Furthermore, the NCG contents in a geothermal fluid may affect the steam quality, corrosion in the piping and equipment, and the power consumption required to remove them from the condenser. Therefore, the dry steam power plant is typically equipped with a gas extraction system, which consumes a significant amount of auxiliary power and increases the capital cost as well as the operational cost. NCGs have a typical composition of
Dry steam power plant Chapter 7
CO2, H2S, NH3, CH4, and H2, and a fraction of inert gases such as N2, He, Ar, and Ne with variations of gas composition for each plant. Generally, CO2 dominates NCG with a fraction higher than 90% while the H2S content can reach 9%, in addition to the small fraction of other gases. In addition, a proper control and abatement of NCG potentially leads to the minimization of the environmental impacts of a dry steam power plant [23]. Further energy losses that affect the system efficiency occur in each component of the power plant, such as the turbine, generator, pumps, etc.; this will be explained in detail in the next section. The overall system efficiency of the dry steam power plant ranges up to 20%, depending on the reservoir enthalpy, as presented in Fig. 7.5. Because dry steam and single-flash systems are technically very similar, with the only difference in the separator, the data are presented together. It is shown that the system efficiency is a function of reservoir enthalpy with a data fitting for the typical model. A variation of the system efficiency represents the characteristic variation of the reservoir and geothermal fluids, such as NCG content that significantly affects the net power generation of the dry steam power plant.
7.4 Configuration and main components of dry steam systems As mentioned above, the dry steam system is relatively simple, with less apparatus compared to other types of geothermal power plant systems. The dry steam system is employed for the vapor-dominated reservoir; therefore, a complicated separation system and its related supporting systems are not required. In this section, the main
107
components of the dry steam system are described, including their characteristics compared to similar components in a conventional steam power plant.
7.4.1 Demister A demister has the function of removing the excess water vapor before the steam is fed into the turbine. A demister is typically a cylindrical tube with steel gratings installed inside to eliminate the water droplets brought by steam from the geothermal well. Steam enters from the top of the demister and directly hits the cone. Due to the differences in pressure and specific gravity, the condensate water droplets and solid particles contained in the vapor will fall to the bottom of the demister. Then, the clean vapor flows through the outlet after passing the filter and is further supplied to the turbine. The performance parameters of the demister are the ability to eliminate the droplets of water expressed in steam dryness (%) and the condensate output rate expressed in the drain flow rate (t/h).
7.4.2 Steam turbine In the dry steam system, dry (vapor-dominated), saturated, or slightly superheated steam is extracted from the production well. Similar to that in conventional steam power plants, the steam turbine in a dry steam system converts the steam energy into mechanical energy to generate electricity in the generator. The turbine is basically a singlepressure unit with impulse-reaction blading [1]. The technological development in the steam turbine allows minimum energy loss, which results in a typical efficiency of up to 86% [1, 24]. Referring to Fig. 7.3, the generated work by the turbine can be represented as follows,
FIG. 7.5 System efficiency of dry steam and single-flash power plants [10].
108
PART II Thermodynamic analysis of geothermal power plants
W_ t ¼ m_ s wt ¼ m_ s ðh1 h2 Þ
(7.6)
The isentropic turbine efficiency (t) can be calculated as the ratio of the actual work to the isentropic work, that is, t ¼
h 1 h2 , h1 h2s
The mass balance of the turbine is defined in Eq. (7.13), and the steam quality at the turbine outlet is calculated based on Eq. (7.14), m_ Total ¼ m_ s + m_ NCG
(7.7)
x¼
therefore, W_ t ¼ m_ s t ðh1 h2s Þ
(7.8)
The ideal enthalpy at the point of 2s (h2s) can be calculated as, s1 s3 (7.9) h2s ¼ h3 + hg h3 sg s3 Besides the mechanical design of the turbine, the steam turbine efficiency is also determined by the steam conditions at the inlet and outlet of the steam turbine. The Baumann rule proposed a simple equation to show the effect of moisture at the turbine outlet, as shown in the following equation [10], x +x in out (7.10) t ¼ td a 2 where t is the turbine efficiency, td is the typical dry turbine efficiency, and xin and xout are the steam qualities at the turbine inlet and outlet, respectively. The value of td is conservatively set to 85%. The coefficient a is an empirical value called the Baumann factor, which may vary from 0.4 to 2 but is usually assumed as 1 [25, 26]. In a dry steam plant, a condensing steam turbine is basically adopted that operates under an intermediate or relatively high-temperature steam source (generally higher than 150°C) [27]. Moreover, the turbine efficiency can drop and become lower than its predicted theoretical value because of several phenomena, including the moisture presence during its expansion and deviation from its isentropic behavior [10]. To operate and control the turbine, two different valves are employed: the main stop and governor valves. They basically throttle the steam, causing a pressure drop. In the steam turbine, the steam is expanding and its pressure energy is recovered by the turbine blades, generating mechanical energy (rotation) [16]. Therefore, two pressure drops basically occur in the steam turbine, leading to the irreversibility. The working fluids that enter and leave the turbine always consist of steam and NCG. Therefore, the calculation of the enthalpy of the turbine inlet and outlet should include both fractions, as shown in the following Eqs. (7.11), (7.12), ht, in ¼ ð1 xNCG Þhs, in + xNCG hNCG, in ht, out ¼ ð1 xNCG Þhs, out + xNCG hNCG, out
(7.11) (7.12)
h s hf , s hg, s hf , s
(7.13) (7.14)
where hm and hs are the specific enthalpies of the mixture and steam (kJ/kg), respectively, while the suffixes in and out refer to the inlet and outlet of the turbine, respectively. The fractions of NCG in the fluid mixture and steam quality are represented by xNCG and x, respectively.
7.4.3 Condenser Because the steam condensate is not recirculated to a boiler as in a conventional power plant, it is available for cooling tower makeup. In fact, an excess of condensate (typically 10–20 wt% of the steam) is available and is usually injected back into the reservoir. Long-term production can deplete the reservoir, and novel ways are being developed to increase the amount of fluid being returned to the reservoir [6, 7]. The use of air-cooled condensers would allow for a 100% return, but so far they have been uneconomic. Mechanically induced draft cooling towers, either counterflow or crossflow, are mostly used for wet cooling systems, but natural draft towers are used at some plants. A direct-contact type condenser is typically used in a dry steam power plant. In a direct contact condenser, heat transfer occurs with the condensation process of steam in accordance with the laws of conservation of mass and energy. Again, the existence of NCGs in the working fluid will have a significant effect during the condensation; therefore, they should be removed. The gas removal systems for NCGs can be installed at the condenser, after condenser, or both at condenser and after condenser [27]. The condenser can be divided into three continuous stages: main, inter-, and aftercondensers. The inter- and aftercondensers are parts of the ejector vacuum system that typically collect the intermediate condensates from the ejectors while separating the gaseous NCGs. All condensate is finally supplied to the main condenser, where the cooling process is carried out by cold water supplied from the cooling tower basin.
7.4.4 Cooling tower A cooling tower in a dry steam power plant has the function of decreasing the temperature of the condensate (hot water stream) from the condenser by exchanging heat with the ambient air. The cool water stream is then returned back to the condenser for another cycle of heat exchange.
Dry steam power plant Chapter 7
Ambient air enters the cooling water with a certain relative humidity (RH) and leaves the cooling water with increased RH up to nearly 100%. The increase of water content in the air comes from the evaporation process of the condensate, which is simultaneously releasing heat from the condensate, resulting in the temperature decrease. There are two important performance parameters of the cooling tower: range and approach. The range is the temperature difference between the hot water that enters the cooling tower and the cold water that leaves the cooling tower. The approach is the temperature difference between the cold water and the wet-bulb temperature of the ambient air. A cooling tower with a large range has an effective cooling process while the approach represents the remaining potential of the cooling process relative to the wet-bulb temperature, as shown in Fig. 7.6. Hot water temperature
Range
Cold water temperature Approach Wet-bulb temperature FIG. 7.6 Illustration of the range and approach of a cooling tower. FIG. 7.7 Decreasing trend line of geothermal fluid production flow rate from Calistoga/The Geysers (United States) [29].
109
The effectiveness of the cooling water can be calculated from the ratio of the range to the total approach and range, as shown in Eq. (7.15). Effectiveness ¼
7.5
Range Approach + Range
(7.15)
System improvements
Due to the characteristic limitation of the dry steam cycle, in order to improve the possibly generated electricity or generation efficiency, there have been several technologies that have been developed and employed. Below are several technologies that have been developed previously.
7.5.1 Utilization of excess steam In the dry steam power plant, the steam is directly fed into the turbine, which means that the reservoir and well condition greatly determine the power plant performance. Geothermal is considered renewable energy, basically due to its sustainability for a very long period. Therefore, it provides the needs of the present generation without compromising the needs of future generations [28]. In order to keep the balance of fluid/heat production by geothermal utilization, most of the geothermal power plants have a reinjection system that returns the condensed fluids to the reservoir. However, for most geothermal reservoirs, the production rate for power generation is higher than the resupply rate by nature, resulting in a declining geothermal production rate, as shown in Fig. 7.7; this figure shows the
110
PART II Thermodynamic analysis of geothermal power plants
decrease of fluid production of the Calistoga field in the United States [28, 29]. Therefore, in order to secure a production rate for longterm plant operation, many geothermal power plants, including the dry steam system, have an agreement of excess steam production that is not effectively used at the beginning of plant operation [17]. On the other hand, not every reservoir has a rapid decline rate as shown in the Calistoga plant. An observation of other plants such as Kamojang in Indonesia shows an insignificant decreasing trend for 30 years of steam production, leaving the excess steam unused for a long period [17]. One of the performance improvements is by utilizing the excess steam to provide additional power generation. Prananto et al. [17] proposed the utilization of the excess steam of the Kamojang plant, which is optimized by determining the most suitable type of steam jet ejector gas liquid-ring vacuum pump (LRVP) removal system. Under the optimized configuration, the net power generated increases up to 7.6% of the current 200 MW installed capacity. With steam production freely available from the excess steam, this additional power generation is expected to improve the electrification ratio in the surrounding area. Another method to utilize the excess steam is by adopting an additional system or cycle to generate the unused steam. Setyawan et al. [30] proposed and evaluated the integration of a new bottoming binary cycle that utilizes the excess steam in the steam receiving header. From their study, which focuses on the Kamojang power plant, the energetic and exergetic efficiencies can be improved from 16.4% to 19.4% and 18.7% to 28.7%, respectively.
7.5.2 Improvement on each component The simple configuration of the dry steam power plant is also the limitation of system improvement. In order to maintain the simple configuration of the dry steam cycle, it only leaves space for improvement in the performance of each component. Rudiyanto et al. [16] reported a thorough exergy analysis on a dry steam power plant, including optimization regarding the ambient temperature effect. From the exergy analysis, it was suggested that the largest irreversibility occurs in the condenser, where the remaining energy is released to the ambient. The exergy analysis provided insight into the location and magnitude of the irreversibility of each component in the dry steam cycle, giving basic consideration regarding the effect of a component improvement on the overall system performance. Additionally, the ambient temperature is reported as one of the important factors that determines the dry steam power plant performance. Location selection, a preliminary study on the ambient condition. and an environment assessment are critical for dry steam power plant development.
7.6
Closing remarks
As the most mature geothermal technology, the dry steam power plant has been widely utilized as one of the main technology options since the first plant began more than 100 years ago. The dry steam cycle power plant basically utilizes direct steam that is produced in the Earth’s interior with a simple configuration of several primary components, such as a demister, a steam turbine, a condenser, a cooling tower, and pumps. Currently, dry steam power plants have a promising potential, supplying about 27% of the total geothermal power plant installed capacity. In the dry steam plant, the working fluid throughout the system includes water vapor, a small amount of water liquid, and a small amount of NCGs. In general, the system can be divided into three continuous modules: a steam cleaning system, power generation system, and a cooling and condensate system. The system performance can be represented by several parameters such as specific steam consumption, which is determined as the amount of steam from the geothermal well that is required to produce 1 kWh of electricity. The system performance of the dry steam power plant is basically determined by the performance of the main components such as demister, turbine, condenser, and cooling tower. Several efforts have been made to improve the performance of the dry steam power plant. The excess steam of a geothermal production well can be utilized and optimized by determining the most suitable type of steam jet ejector gas LRVP removal system. Under the optimized configuration, the net power generated increases up to 7.6% of the current 200 MW installed capacity. The exergy analysis on dry steam power plants gives an overview of the irreversibility that occurs in each component, providing a basic consideration regarding the effect of a component improvement on the overall system performance.
References [1] Dipippo R. Geothermal power plants: Principles, applications, case studies and environmental impact. vol. 1; 2008. https://doi.org/ 10.1017/CBO9781107415324.004. [2] Bryan TS. The geysers of yellowstone; 2018. [3] Chamorro CR, Mondejar ME, Ramos R, Segovia JJ, Martı´n MC, Villaman˜a´n MA. World geothermal power production status: energy, environmental and economic study of high enthalpy technologies. Energy 2012;42:10–8. https://doi.org/10.1016/j.energy.2011.06.005. [4] Prananto LA, Fauzi Soelaiman TM, Aziz M. Utilization of excess steam through dry steam cycle at Kamojang geothermal power plant. Energy Procedia 2017;142:160–5. https://doi.org/10.1016/j.egypro.2017.12.026. [5] IRENA. Technology brief: geothermal power; 2017. p. 111–3. [6] Kruger P, Otte C. Geothermal energy: resources, production, stimulation. Stanford, CA: Stanford University Press; 1973. [7] Chierici A. Planning of a geothermoelectric power plant: technical and economic principles. In: Proc. United Nations conf. new sources
Dry steam power plant Chapter 7
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
energy sol. energy, wind power geotherm. energy, vol. 3. New York: United Nations; 1963. p. 299–313. Lee Y, Park S, Kim J, Chan H, Koo M. Geothermal resource assessment in Korea. Renew Sust Energ Rev 2010;14:2392–400. https://doi.org/10.1016/j.rser.2010.05.003. Blackwell DD, Negraru PT, Richards MC. Assessment of the enhanced geothermal system resource base of the United States. Nat Resour Res 2007;15:283–308. https://doi.org/10.1007/s11053007-9028-7. Zarrouk SJ, Moon H. Efficiency of geothermal power plants: a worldwide review. Geothermics 2014;51:142–53. https://doi.org/ 10.1016/j.geothermics.2013.11.001. International Geothermal Association. Geothermal power database n.d. https://www.geothermal-energy.org/explore/our-databases/geo thermal-power-database/ ([Accessed 20 February 2020]). OpenEI. List of dry steam plants.Natl Renew Energy Lab n.d. https:// openei.org/wiki/Geothermal_Steam_Power_Plant#tab¼List_of_ Dry_Steam_Plants ([Accessed 19 May 2020]). Burgassi PD. Historical outline of geothermal technology in the Larderello Region to the middle of the 20th century. In: Stories from a heated earth our geotherm heritage; 1999. p. 195–219. DiPippo R. Geothermal power plants: evolution and performance assessments. Geothermics 2015;53:291–307. https://doi.org/ 10.1016/j.geothermics.2014.07.005. Hanbury O, Vasquez VR. Life cycle analysis of geothermal energy for power and transportation: a stochastic approach. Renew Energy 2018;115:371–81. https://doi.org/10.1016/j.renene.2017.08.053. Rudiyanto B, Illah I, Pambudi NA, Cheng CC, Adiprana R, Imran M, et al. Preliminary analysis of dry-steam geothermal power plant by employing exergy assessment: case study in Kamojang geothermal power plant, Indonesia. Case Stud Therm Eng 2017;10:292–301. https://doi.org/10.1016/j.csite.2017.07.006. Prananto LA, Juangsa FB, Iqbal RM, Aziz M, Soelaiman TAF. Dry steam cycle application for excess steam utilization: Kamojang geothermal power plant case study. Renew Energy 2018;117:157–65. https://doi.org/10.1016/j.renene.2017.10.029. Soelaiman TAF. Geothermal energy. In: Rashid MH, editor. Electric renewable energy systems. Boston: Academic Press; 2016. p. 114–39 [Chapter 7] https://doi.org/10.1016/B978-0-12-804448-3.00007-4. DiPippo R. Geothermal power generation: Developments and innovation. Elsevier Inc; 2016. https://doi.org/10.1016/C2014-0-03384-9.
111
[20] Fallah M, Ghiasi RA, Mokarram NH. A comprehensive comparison among different types of geothermal plants from exergy and thermoeconomic points of view. Therm Sci Eng Prog 2018;5:15–24. https:// doi.org/10.1016/j.tsep.2017.10.017. [21] Vorum M, Fitzler E. Comparative analysis of alternative means for removing noncondensable gases from flashed-steam geothermal power plants. Golden, CO: National Renewable Energy Lab; 2000. https://doi.org/10.2172/758765. [22] Khalifa HE, Michaelides E. Effect of non condensable gases on the performance of geothermal steam power systems; 1978. https://doi. org/10.2172/6081081. [23] Tomasini-Montenegro C, Santoyo-Castelazo E, Gujba H, Romero RJ, Santoyo E. Life cycle assessment of geothermal power generation technologies: an updated review. Appl Therm Eng 2017;114:1119– 36. https://doi.org/10.1016/j.applthermaleng.2016.10.074. [24] Pakpahan P, Ridwan RH, Lubis IES. Parasitic load efficiency program Kamojang generation business unit Pt. Indonesia Power, Indonesia; 2010. p. 25–9. [25] Baumann K. Some recent developments in large steam turbine practice. J Inst Electr Eng 1921;59:565–623. https://doi.org/ 10.1049/jiee-1.1921.0040. [26] DiPippo R. Geothermal power plants of the United States: a technical survey of existing and planned installations; 1978. https://doi.org/ 10.2172/6389636. [27] Buonocore E, Vanoli L, Carotenuto A, Ulgiati S. Integrating life cycle assessment and emergy synthesis for the evaluation of a dry steam geothermal power plant in Italy. Energy 2015;86:476–87. https:// doi.org/10.1016/j.energy.2015.04.048. [28] Rybach L. Sustainable use of geothermal resources: renewability aspects. In: Lect. sustain. use oper. policy geotherm. resour. short course prior to int. conf. IGC2003, Iceland, UNU/GTP; 2003. [29] Sanyal SK, Butler SJ, Brown PJ, Goyal K, Box T. An investigation of productivity and pressure decline trends in geothermal steam reservoirs. GRC Trans 2000;24:873–7. [30] Dany Setyawan N, Agung Pambudi N, Utomo F, Huat Saw L, G€urt€urk M, Mohammadzadeh Bina S. Performance improvement of drysteam geothermal power plant by employing bottoming binary system. IOP Conf Ser Earth Environ Sci 2019;249. https://doi.org/10.1088/17551315/249/1/012022.
Chapter 8
Binary geothermal power plant Yusuf Bas¸ o gula, Onur Vahip G€ ulerb, and Ali Kec¸ ebas¸ b a
Department of Mechanical Engineering, Engineering Faculty, Adıyaman University, Adıyaman, Turkey b Department of Energy Systems Engineering,
Technology Faculty, Mu gla Sıtkı Koc¸man University, Mu gla, Turkey
Nomenclature C cp E_ _ Ex h m_ P Q_ s T W_
heat capacity rate (kW/K) specific heat capacity (kJ/kg K) energy rate (kW) exergy rate (kJ/s or kW) specific enthalpy (kJ/kg) mass flow rate (kg/s) pressure (kPa) heat transfer rate (kW) specific entropy (kJ/kgK) temperature (°C or K) power (kW)
Subscripts 0 Cond Conde Cool d ex f F_Pump g Gen i is k max min NCG o Preheat Recup st Turb Vap
dead state condenser condensation cooling destruction exergy fluid feed pump gas generator inlet isentropic location maximum minimum noncondensing gases outlet preheater recuperator steam turbine vaporizer
Greek letters D « h r c
difference () effectiveness (%) energy or first law efficiency (%) density (kg/m3) flow exergy (kJ/kg)
8.1
Introduction
The growing world population and the desire to increase the quality of life for human beings are increasing the demand for energy, which also increases the need for energy production. Today, the need for energy is met with both fossil and renewable energy resources. It is an inevitable fact that the interest in renewable energy will also grow, considering that fossil fuels will be exhausted after a certain period of time. The depletion of fossil resources and the formation of harmful emission gases that are released when used are drawing the attention of the world to renewable energy sources. When renewable energy sources are used for electricity, heating, and cooling production, it is much more advantageous in terms of greenhouse gas emissions than fossil-based energy types. The best-known renewable energy sources today are wind, solar, hydraulic, geothermal, and biomass. However, the renewable electricity produced is not economically competitive against fossil-based energy sources [1]. They are basic base power plants that use fossil energy sources as the primary energy source in most electrical grids. Fortunately, geothermal energy is the only renewable energy source that is not affected by external weather conditions. Thus, geothermal energy can be used as the basic base power plant. In addition, it is thought that geothermal energy has received the least attention in the world and this may be related to the fact that geothermal energy is mostly located all over the world, near the borders of the tectonic Earth surface [2]. Geothermal energy is thermal energy in hot water, steam, gas, or hot dry rocks under pressure accumulated at various depths of the Earth’s crust. In other words, it is derived from heat generated and stored in the mantle and core of the Earth [3]. The heat in the Earth’s crust is estimated to be 5.4 billion EJ. If 0.1% of this energy could be used, it would meet the energy consumption of the world for the next 10,000 years. However, the technology required to use energy at this scale has not yet been developed [4]. In
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00013-5 Copyright © 2021 Elsevier Inc. All rights reserved.
113
114
PART II Thermodynamic analysis of geothermal power plants
regions where geothermal energy is available, reservoirs are divided into three classes: low- (T < 90°C), medium- (90°C 150°C) enthalpy sources, depending on the fluid temperature [5]. The primary use of geothermal energy is electricity generation. Historically, Prince P.G. Conti installed the first device to generate electricity from a geothermal steam well in Larderello, Italy, in 1904, and in 1913 the first commercial geothermal power plant (GPP) was connected to the grid with a 250 kW turbo alternator [6]. The success of these experiences led to the production of small-scale electricity from geothermal sources in Beppu, Japan, in 1919 and from the geysers of California in 1921. Wairakei station in New Zealand, a milestone in the history of technology for electricity production, started operating in 1958 and became the first facility to use wet steam technology (flash steam technology). Significant developments have been achieved in the use of geothermal energy by establishing power plants in the United States in 1960, Mexico in 1961, Japan in 1966, and China in 1970. Previous plants used geothermal reservoirs containing dry steam [7]. In 1977, the first double-flash power plant became operational in Japan, and GPPs using binary cycles were put into operation in 1984 [8]. Thus, electricity production from geothermal energy has spread to the world. GPPs use hydrothermal sources that contain water (hydro) and heat (thermal). Hot water or steam is used to turn a turbine and then generate electricity. GPPs need hydrothermal sources to draw heat from hot water or dry steam wells (about 150–370°C) [9]. In this case, a minimum temperature of 150°C is generally required to generate electricity from geothermal sources. The power plant design is usually a function of the temperature and pressure of the existing geothermal resource. Almost all GPPs are examined in four categories: dry steam, single-flash, double-flash, and binary cycles [10]. With the advanced development of GPPs, according to Bertani [11], the geothermal total installed power of the world reached 12,640 MW. 5079 MW of this consists of a single-flash cycle, 2863 MW of dry steam, 2544 MW of double flash, 1790 MW of binary cycle, 182 MW of triple flash, and 181 MW of back-pressure and hybrid cycles. All cycles, except for the binary cycle, are used at temperatures above 150°C of the geothermal source. However, with the binary GPP, very low temperatures can also be used for electricity generation. In cases where the geothermal fluid exits the production well at a temperature of less than 150°C (typically 110–180°C), it is not possible to convert the liquid fluid to steam. For this reason, binary GPPs are typically used in these applications. They are closest in thermodynamic principle to a conventional fossil or nuclear power plant where the working fluid is subjected to a real closed cycle. In a basic binary GPP, geothermal fluid (brine) is used as a pressurized liquid to provide heat input into an organic Rankine cycle (ORC). The geothermal fluid heats an
organic working fluid (e.g., R32, R134a, n-pentane, isopentane) in the evaporator of a simple Rankine cycle, and then the working fluid evaporates. This steam generates electrical energy as it passes through a turbine that rotates a generator. Consequently, this cycle is called “binary” because both geothermal fluid and organic working fluid are used in the cycle. Nowadays, binary GPPs operating according to the ORC principle have emerged as a promising development to generate electricity from sources at low temperatures. It is extremely important to evaluate binary GPPs, which have attracted a lot of attention in the world, from the thermodynamic point of view at the project design stage before proceeding to production. In addition, performing these analyses in existing power plants directly affects power plant performance and therefore the amount of electricity produced. The design of both efficient and low-cost GPPs is one of the problems that energy engineers face constantly. Especially in developing countries, it has become extremely important to develop more accurate and systematic approaches to improve the design of energy systems, with an increasing need for energy and increasing global demand for energy to reduce the environmental impact of these systems. In many binary GPP projects, thermodynamic analysis has proven to be a meaningful tool [12, 13]. The thermodynamic analysis is concerned with the first law of thermodynamics, which primarily expresses the principle of conservation of energy. Thermodynamic losses occurring within a system are generally not fully identified and evaluated by energy analysis. Exergy analysis also evaluates the location, magnitude, and effects of the irreversibilities that occur in an energy conversion system. Nowadays, most GPPs are designed with energy performance criteria according to energy analysis. Their energy losses are determined by the exergy analysis. In addition, the exergy analysis is done after performing the energy analysis. This real approach can illuminate the healing areas of a system. Over the years, this type of analysis has been extensively addressed and applied to GPPs. Therefore, in this chapter, the methodology of the binary GPP with double pressure on the exergy analysis is discussed in detail.
8.2
Binary GPP
Binary geothermal systems are used in regions where the source temperature is not high. As the geothermal source temperature decreases, the liquid ratio in the two-phase geothermal fluid coming from the source increases. This situation prevents the use of geothermal resources in flash technology. Binary geothermal systems can be used when the geothermal fluid temperature is between 110°C and 180°C [12]. It is not possible to convert liquid fluid to steam, especially in regions below 150°C. In other words, geothermal fluid cannot be sent directly into the turbine. For
Binary geothermal power plant Chapter 8
this reason, another fluid with a lower boiling point temperature than water is used. In ORCs, heat transfer from the geothermal fluid to this second fluid is performed. The second fluid is sent to the turbine. The plants whose working principle is realized in this way are called binary GPPs. A binary GPP consists of two different cycles. The first part is completed by the geothermal fluid coming out of the well as two phases (liquid and steam); it then enters the vaporizer and preheater and is sent back to the ground. The second part is a closed cycle, and it is a system that uses organic refrigerant as fluid. A preheater and vaporizer are also evaporated from the organic working fluid heated by geothermal fluid. Then the organic working fluid passes through the turbine. The working fluid from the turbine is sent to the condenser. In this section, after the condensed organic working fluid becomes completely liquid, it is pressurized with the feed pump and enters the preheater again. In this cycle, the turbine is connected to a generator to generate
FIG. 8.1 A schematic flow diagram of a binary GPP with double pressure.
115
electricity. The most significant difference in the binary cycle from other cycles is that it can produce electricity at lower temperatures. However, to reduce thermodynamic losses caused by geothermal fluid (brine) in evaporators as a heat exchanger, a double pressure cycle is designed. These losses are caused by the heat transport process across a large temperature difference between geothermal fluid (brine) and organic working fluid. That is, they can be reduced by maintaining the close match between the cooling curve of geothermal fluid (brine) and the heatingboiling curve of the organic working fluid. Thus, binary GPPs have been designed where two different ORCs (double pressure) are fed from a single geothermal fluid. The flow chart of such binary GPPs with double pressure is shown in Fig. 8.1. As a case study, the data of the plant drawn schematically in Fig. 8.1 are used. The arterial geothermal fluid obtained from the production wells for the plant is mixed,
116
PART II Thermodynamic analysis of geothermal power plants
and geothermal fluid (brine as the saturated liquid) and steam (as saturated steam) are separated into two separate phases in each well. These production wells have a temperature between 170°C and 165°C, a pressure between 10 and 14 bar, and a flow rate of 1100–1500 tons/h. First, steam is obtained from the separators at 165°C temperature, 8.33 kg/h flow rate, and 1040 kPa pressure. Nearly 30% of this steam consists of NCGs (mostly noncondensing gases with CO2). The remaining 70% is the geothermal fluid vapor. Then, the liquid geothermal fluid from the separators at 165°C, 1040 kPa, and 445 kg/h is prepared for the organic Rankine cycle (ORC). In Fig. 8.1, the binary GPP with double pressure consists of two ORCs that are side by side and different from each other. The first cycle at a flow rate of 160 kg/h is the first level (with high pressure, 1261 kPa) while the cycle at 196 kg/h is called the second level (with low pressure, 687 kPa). First, the geothermal fluid is pressed on the first level vaporizer (Vap_1) and the organic working fluid (npentane) in the first level is heated. Then, the geothermal fluid coming out of Vap_1 at 136°C and 730 kPa passes to the second level vaporizer (Vap_2) and the working fluid (n-pentane) in the ORC at the second level is heated. Due to the low temperature and pressure of the geothermal fluid into Vap_2 at the second level, the steam obtained in the separators is given at a flow rate of 8.33 kg/h, a pressure of 1040 kPa, and a temperature of 165°C. Both liquid brine and steam pass to Vap_2 without mixing. At the outlet of Vap_2, the NCG at a 2.25 kg/h flow rate and the steam at a 5.25 kg/h flow rate are released into the atmosphere at 107°C. On the other hand, geothermal fluid (brine), which is condensate at 107°C and a 0.83 kg/h flow rate, is sent to reinjection. For the first level in Fig. 8.1, n-pentane heated in Vap_1 at 137°C and 1261 kPa is sent to the turbine (Turb_1), turning it. The n-pentane, which comes out of TURB 1 at 82°C and 150 kPa, is sent to the recuperator (Recup). The recuperator is used to reduce the excessive cooling load of the condenser. The n-pentane exiting the recuperator is cooled with air in the condenser units (Cond_1) and made liquid. Then it is recirculated with the feed pump (F_Pump_1). The n-pentane at 1261 kPa is pumped to the recuperator by the feed pump. The heat stored in the recuperator is returned to the n-pentane and its temperature is increased. The n-pentane exiting from the recuperator is sent to the preheater (Preheat_1) and from there to the vaporizer (Vap_1). Thus, the cycle in the first level is continuously recirculated. As shown in Fig. 8.1, the n-pentane of 109°C and 687 kPa from Vap_2 is sent to the turbine (Turb_2). The n-pentane coming out of the turbine passes to the air-cooled condenser (Cond_2) at 69°C. The n-pentane cooled in the condenser is pumped to the preheater (Preheat_2) with a 687 kPa by the feed pump (F_Pump_2). Then, by switching
back to Vap_2, the cycle is completed in the second level. The difference from the first level is that no recuperator (Recup) is used while geothermal fluid vapor is provided in Vap_2 in ORC. The turbines in both ORCs are connected to the generator via the same shaft. Therefore, the rotational speeds of the turbines at rpm should be identical. If the loads in the turbines differ, the system will drop in power or the system will shut down. To achieve this, the geothermal fluid vapor is supplied to the vaporizer at the second level. Today, in accordance with the application purpose and design, it is seen in binary GPPs where recuperators in both ORC are present and/or steam is supplied to both vaporizers.
8.3
Thermodynamic analysis
Exergy analysis is a thermodynamic analysis based on the Second Law of Thermodynamics, making it possible to realistically and significantly evaluate and compare energy systems and state changes. Exergy or second law efficiency found by exergy analysis is compared with maximum performance in the real system. Thus, there are locations, quantities, and causes of thermodynamic losses under exergy analysis. For the analysis of energy conversion systems, energy analysis should be done before performing exergy analysis. In other words, the First Law of Thermodynamics should be applied to the system. Considering the design scheme for the binary GPP shown in Fig. 8.1, the energy analysis is applied to the plant first. In this case, neglecting the potential and kinetic energies as well as mass and energy balances can respectively be given as follows. X X m_ o (8.1) m_ i ¼ X X Q_ net W_ net ¼ m_ 0 h0 m_ i hi (8.2) _ Q_ net , W_ net , and h denote the mass flow rate, net where m, heat rate, work rate, and the specific enthalpy, respectively. The exergy balance used for the plant is given below. X X X T0 _ _ d (8.3) Q k W_ ¼ m_ i ci m_ o co + Ex 1 Tk _ d is the exergy rate associated with exergy where Ex destruction. Subscript zero indicates properties at the restricted dead (ambient) state of P0 and T0. Q_ k is the heat transfer rate crossing the boundary at temperature Tk at location k. In addition, c is the specific flow exergy as given by c ¼ ðh h0 Þ T0 ðs s0 Þ
(8.4)
The unit exergy rate at the state number specified in Fig. 8.1 can be determined as follows:
Binary geothermal power plant Chapter 8
_ ¼ mc _ Ex
117
(8.5)
where s is the entropy. T0 evaluates the thermodynamic performance, and the exergy efficiency of the whole system (ex) can be expressed as ex ¼
_ output Ex _ input Ex
(8.6)
where the subscript output refers to net output or product or desired value, and the subscript input refers to given or used value.
8.3.1
FIG. 8.2 Flow diagram of vaporizer 1 in the first level of the plant.
1 with a high temperature and pressure and transfers the heat
General components of the binary GPP to n-pentane as an organic working fluid and then leaves
The thermodynamic analysis of binary geothermal power plant (GPP) is determined by using mass, energy, and exergy balance equations. The binary GPP consists of components such as the evaporator (Vap_1 and 2, Preheat_1 and 2, Recup), the turbine (Turb_1 and 2), the condenser (Cond_1 and 2), the feed pump (F_Pump_1 and 2), the fan units (Fans_1 and 2), and the generator at the first and second levels (see Fig. 8.1). In the binary GPP, the definitions of all components together with the mass, energy, and exergy balances are given in the following sections.
8.3.1.1 Evaporator The evaporator is a component that provides heating of the working fluid by drawing heat energy from a medium or fluid. In binary GPPs, a component such as a vaporizer, preheater, recuperator, etc., is called the evaporator. All evaporators mentioned here are shell-and-tube heat exchangers. Three different model approaches are generally used to determine the evaporator performance. These are (i) the distributed modeling approach, (ii) the zone modeling approach, and (iii) the single node aggregate modeling approach. Herein, the single node aggregate modeling approach is used. In the modeling approach, the e-NTU method is chosen because it provides a more accurate and faster response than the logarithmic mean temperature difference (LMTD) method in providing thermal model connections between components. It is assumed that the heat exchangers are well insulated and that all heat transfers will be between the brine and the working fluid. 8.3.1.1.1 Vaporizer 1 (Vap_1) The vaporizer is one of the most critical parts of a binary GPP. It enables the organic working fluid and the geothermal fluid in the closed loop to exchange heat without mixing. In a binary GPP, the vaporizer for the first level (Vap_1) is designed as a horizontal shell and tube heat exchanger. The flow diagram of Vap_1 in the first level is shown in Fig. 8.2. The geothermal fluid (brine) enters line
from line 2. The working fluid enters from Preheat_1 to Vap_1 through line 11 and exits from line 12 by increasing the temperature from the geothermal fluid in Vap_1. The change of heat between geothermal fluid (brine) and organic working fluid in Vap_1 is given in Fig. 8.3. The figure shows the temperature changes in the heat exchanger versus the amount of heat. In addition, the flow pattern of Vap_1 is considered an inverse (opposite) flow. Geothermal fluid enters Vap_1 at T1 temperature and then exits at T2 temperature. In Vap_1 as a heat exchanger, the cooling process of the geothermal fluid and the evaporation process of the organic working fluid are required. In the location where the organic working fluid reaches saturation temperature in Vap_1, the difference between the temperatures of the organic working fluid and the geothermal fluid is the least. This position in the heat exchanger, which acts as an evaporator, is defined as the pinch point.
FIG. 8.3 Change of temperature-heat between the geothermal fluid and organic work fluid of vaporizer 1.
118
PART II Thermodynamic analysis of geothermal power plants
The mass, energy, and exergy balance equations for Vap_1 according to the state numbers specified in Fig. 8.1 are given below: Mass balance: m_ 1 ¼ m_ 2
(8.7)
m_ 11 ¼ m_ 12
(8.8)
for the geothermal fluid and the organic working fluid, respectively. Energy balance: The e-NTU method has been taken into account in order to model the change of inlet and outlet temperatures of Vap_1. Thus, l
For heat transfer during the heating process C1 ¼ m_ 1 cp,1
(8.9)
C11 ¼ m_ 1 cp,11
(8.10)
C min ,Vap_1 ¼ min ðC1 , C11 Þ (8.11) Q_ Vap_1,heat ¼ eVap_1 C min , Vap_1 TVap_1, pinch T11 (8.12) DTpinch ¼ TVap_1,pinch T12
(8.13)
where ek ¼ qreal, k/qmax, k is the effectiveness coefficient. l
For heat transfer during the boiling process QVap_1,boil ¼ m_ 1 h12g h12f ¼ C1 T1 TVap_1, pinch (8.14)
l
FIG. 8.4 Flow diagram of vaporizer 2 in the second level of the plant.
is sent to the reinjection pipeline from line 9. NCG can be released into the air from line 10 at the top of Vap_2 with unused steam or sent to another system to produce CO2. In addition, the temperature and pressure of the geothermal fluid coming from line 2 at the outlet of Vap_1 decrease and leave from line 3. Preheater 2 (Preheat_2) is the line of workflow that enters Vap_2 with line 18, and n-pentane rises from Vap_2 (line 19) by increasing its temperature and pressure. The change in temperature among the geothermal fluid (brine), steam, NCG, and organic working fluid in Vap_2 at the second level is given in Fig. 8.5. The geothermal fluid enters Vap_2 from different lines both at the T1 temperature and the T10 temperature as steam. Here, the NCGs in the steam are sent to the atmosphere or CO2 facility at T10 temperature. The condensing steam is given to reinjection at T9 temperature. The organic working fluid passes through the heat exchanger without mixing with the others. The mass, energy, and exergy balance equations for Vap_2 are defined as below.
For total heat transfer Q_ Vap_1 ¼ Q_ Vap_1,heat + Q_ Vap_1,boil E_ Vap_1 ¼ E_ 1 E_ 2 E_ 12 E_ 11 Vap_1 ¼
E_ 12 E_ 11 E_ 1 E_ 2
(8.15) (8.16) (8.17)
Exergy balance: _ d,Vap_1 ¼ Ex _ 1 Ex _ 2 Ex _ 11 _ 12 Ex Ex ex,Vap_1 ¼
_ 11 _ 12 Ex Ex _ 2 _ 1 Ex Ex
(8.18) (8.19)
8.3.1.1.2 Vaporizer 2 (Vap_2) A horizontal shell and tube heat exchanger are in Vap_2 of the GPP’s second level. The flow diagram of Vap_2 is shown in Fig. 8.4. However, its design is quite different from Vap_1 at the first level. Namely, geothermal fluid vapor and NCGs enter with a single pass through line 10 at the bottom of Vap_2. The steam condenses at Vap_2, then
FIG. 8.5 Change of temperature-heat between geothermal fluid (brine), steam, and organic work fluid of vaporizer 2.
Binary geothermal power plant Chapter 8
119
Mass balance: m_ 2 + m_ 10st + m_ 10NCG ¼ m_ 3 + m_ 100st + m_ 100NCG + m_ 9 m_ 18 ¼ m_ 19
(8.20) (8.21)
for the liquid and vapor forms of the geothermal fluid, and the organic workingfluid, respectively. Energy balance: l For heat transfer during the heating process C1 ¼ m_ 1 cp,1
(8.22)
C1 ¼ m_ 10 cp,10
(8.23)
C18 ¼ m_ 18 cp,18
(8.24)
C min ,Vap_2 ¼ min ðC10 , C18 Þ
(8.25)
0
Q_ Vap_2,heat ¼ eVap_ 2 C min ,Vap_2 ðT10 T18 Þ l
DTpinch ¼ T3 T19 l
(8.27) (8.28)
For total heat transfer Q_ Vap_2 ¼ Q_ Vap_2, heat + Q_ Vap_2,boil (8.29) 0 E_ Vap_2 ¼ E_ 2 E_ 3 + E_ 1 E_ 10 + E_ 9 E_ 19 E_ 18 (8.30) E_ 19 E_ 18 Vap_2 ¼ 0 E_ 2 E_ 3 + E_ E_ 10 + E_ 9
8.3.1.1.3
_ 18 _ 19 Ex Ex 0 _ 3 + Ex _ Ex _ 10 + Ex _ 9 _ 2 Ex Ex 1
(8.34)
m_ 17 ¼ m_ 11
(8.35)
for the geothermal fluid and the organic working fluid, respectively. Energy balance: Similar to equilibrium equations achieved for the evaporators, these can be derived. Thus, (8.36) E_ Preheat_1 ¼ E_ 4 E_ 6 E_ 11 E_ 17 Preheat_1 ¼
(8.32) (8.33)
Preheater 1 (Preheat_1)
The flow diagram of the preheater 1 (Preheat_1) in the first level is shown in Fig. 8.6. In the tube part of the heat exchanger, the geothermal fluid passes through line 4 and exits from line 6. On the other hand, n-pentane passes through line 17 from the shell section and passes to Vap_1 with line 11. This component is based on the principle of increasing the temperature of the organic fluid that will enter the vaporizer by passing through a preheater. Because the geothermal fluid coming out of the vaporizer
E_ 11 E_ 17 E_ 4 E_ 6
(8.37)
Exergy balance: _ 4 Ex _ 6 Ex _ 17 _ 11 Ex _ d,Preheat_1 ¼ Ex Ex ex,Preheat_1 ¼
Exergy balance:
ex, Vap_2 ¼
m_ 4 ¼ m_ 6
(8.31)
1
_ 2 Ex _ 3 + Ex _ 0 Ex _ 10 + Ex _ 9 _ d,Vap_2 ¼ Ex Ex 1 _ 19 Ex _ 18 Ex
is still at high temperature. It is the same as the vaporizer as a working principle but smaller in capacity. The mass, energy, and exergy balance equations for Preheater 1 are. Mass balance:
(8.26)
For heat transfer during the boiling process QVap_ 2,boil ¼ m_ 19 h19g h19f ¼ C2 ðT2 T3 Þ
FIG. 8.6 Flow diagram of preheater 1 in the first level of the plant.
_ 17 _ 11 Ex Ex _ 6 _ 4 Ex Ex
(8.38) (8.39)
8.3.1.1.4 Preheater 2 (Preheat_2) The flow diagram for preheater 2 (Preheat_2) in the second level of the GPP is presented in Fig. 8.7. In Preheat_2, the geothermal fluid passes through line 5 and exits from line 7. The n-pentane as the organic working fluid enters from line 22 and passes to Vap_2 with line 18. The mass, energy, and exergy equations for Preheater 2 can be expressed as follows.
FIG. 8.7 Flow diagram of preheater 2 in the second level of the plant.
120
PART II Thermodynamic analysis of geothermal power plants
Mass balance:
Mass balance: m_ 5 ¼ m_ 7
(8.40)
m_ 22 ¼ m_ 18
(8.41)
for the geothermal fluid and the organic working fluid, respectively. Energy balance: E_ Preheat_2 ¼ E_ 5 E_ 7 E_ 18 E_ 22 Preheat_2 ¼
E_ 18 E_ 22 E_ 5 E_ 7
(8.43)
ex,Preheat_2 ¼
8.3.1.1.5
_ 22 _ 18 Ex Ex _ 7 _ 5 Ex Ex
(8.46)
for the organic working fluid. Energy balance: E_ recup ¼ E_ 17 E_ 16 E_ 13 E_ 14 recup ¼
(8.42)
Exergy balance: _ 18 Ex _ d,Preheat_2 ¼ Ex _ 5 Ex _ 7 Ex _ 22 Ex
m_ 13 ¼ m_ 14 ¼ m_ 16 ¼ m_ 17
E_ 13 E_ 14 E_ 17 E_ 16
(8.47) (8.48)
Exergy balance: _ d,recup ¼ Ex _ 17 E_ x16 Ex _ 14 _ 13 Ex Ex ex,recup ¼
(8.44)
_ 14 _ 13 Ex Ex _ 16 _ Ex 17 Ex
(8.49) (8.50)
(8.45)
Recuperator (Recup)
A recuperator is a type of heat exchanger that has separate flow paths for each fluid throughout its passages and heat is transferred through the separating walls. Most recuperators operate as counter-flow heat exchangers. Only the first level of the GPP discussed here uses a recuperator. The flow diagram of the recuperator (Recup) is given in Fig. 8.8. In the Recup, only organic working fluid is passed through s the hell and tube without mixing. The heat of the organic working fluid from line 13 at the exit of turbine 1 (Turb_1) is taken over by the Recup, and the organic working fluid is removed from line 14. Then, the organic working fluid passes to condenser 1 (Cond_1). On the other hand, the organic working fluid, whose pressure is increased in feed pump 1 (F_Pump_1), enters the Recup from line 16. The temperature taken from the organic working fluid between lines 13–14 in the Recup is given back to organic working fluid and its temperature is increased. Thus, along line 17, it is sent to the preheater 1 (Preheat_1). The mass, energy, and exergy balance equations for the recuperator are given below.
8.3.1.2 Turbine The steam turbine is used to convert the thermal energy of high-pressure steam into mechanical energy. The steam turbine is a very safe, efficient, and economical method that generates power. High-pressure steam is sent to the turbine blades. Mechanical work is produced in the turbines by a rotary motion. Electricity is then generated from mechanical energy by a generator connected to the turbine. The traditional approach to generate electricity from heat is to use steam turbines in a Rankine cycle. It is different from the traditional Rankine cycle as it uses an organic working fluid instead of water in the organic Rankine cycle (ORC). ORC fluid is generally chosen as a low boiling point working fluid capable of capturing heat at low temperatures. Therefore, different turbine designs should be used in ORC that can work with low temperatures and low-quality heat. In a binary GPP, two turbines are connected to the same shaft and thus, electricity is produced with two turbines and a single generator. The power of the generator can be calculated with: (8.51) W_ Gen ¼ Gen W_ Turb_1 + W_ Turb_2 where W_ Gen is the generator efficiency in the transmission of rotating shaft power. 8.3.1.2.1
FIG. 8.8 Flow diagram of the recuperator in the first level of the plant.
Turbine 1 (Turb_1)
The flow diagram of the turbine at level 1 is shown in Fig. 8.9. In vaporizer 1 (Vap_1), the working fluid is heated to the boiling point and then evaporated. The superheated organic fluid vapor enters the turbine from line 12 at the first level. It directs a series of blades to rotate a shaft in the turbine and then expand. Thus, by converting the kinetic energy gained by the expansion of the steam process, it causes the production of rotational shaft power. The fluid
Binary geothermal power plant Chapter 8
FIG. 8.9 Flow diagram of turbine 1 in the first level of the plant.
leaves the turbine from line 13 with the pressure and temperature lowered. Turbine power can be neglected in the adiabatic process, but is calculated assuming constant flow when the potential and kinetic energy are stable. Thus, the balance equations of the turbine can be determined as: Mass balance: m_ 5 ¼ m_ 7
FIG. 8.10 Flow diagram of turbine 2 in the second level of the plant.
E_ Turb_2 ¼ E_ 19 E_ 20 W_ Turb_2 Turb_2 ¼
_ 19 Ex20 W_ Turb_2 _ d,Turb_2 ¼ m_ 19 Ex Ex ex, Turb_2 ¼
for the organic work fluid. Energy balance:
(8.53) E_ Turb_1 ¼ E_ 12 E_ 13 W_ Turb_1 Turb_1 ¼
h12 h13 h12 h13,is
(8.54) (8.55)
Exergy balance: _ 12 Ex13 W_ Turb_1 _ d_Turb_1 ¼ m_ 12 Ex Ex ex,Turb_1 ¼
8.3.1.2.2
W_ Turb_1 E_ 12 E_ 13
(8.56) (8.57)
h19 h20 h19 h20,is
(8.60) (8.61)
Exergy balance:
(8.52)
W_ Turb_1 ¼ m_ 12 ðh12 h13 Þ ¼ m_ 12 Turb_1, is ðh12 h13,is Þ
121
W_ Turb_2 E_ 19 E_ 20
(8.62) (8.63)
8.3.1.3 Condenser The condenser is a device that removes excess heat from the system with the help of water or air. Conventional power plants use a water-cooled system to cool the steam turbine exhaust vapor using nearby water wells such as cooling towers or lakes. Therefore, there is a large amount of water consumption. The plants with air-cooled condensers have the advantages of saving water and being environmentally friendly. Moreover, the air-cooled condenser is widely applied in industries, and maintenance costs are low. However, the net power output of the plant is reduced due to a large amount of power input to the air-cooled condenser fan units. These condensers are horizontal air-cooled heat exchangers containing fans in tube bundles. The fan unit transfers air from under the tube bundles to the atmosphere. For this purpose, fans are driven by electric motors.
Turbine 2 (Turb_2)
Turbine 2 (Turb_2), whose flow chart is given in Fig. 8.10, operates on the same principle as Turb_1. Thus, the mass, energy, and exergy balance equations for Turb_2 are given below: Mass balance: m_ 19 ¼ m_ 20
(8.58)
for the organic working fluid. Energy balance: W_ Turb_2 ¼ m_ 19 ðh19 h20 Þ ¼ m_ 19 Turb_2, is ðh19 h20,is Þ (8.59)
8.3.1.3.1
Condenser 1 (Cond_1)
The flow diagram of the air-cooled condenser 1 (Cond_1) in the binary GPP is shown in Fig. 8.11. As shown in the figure, the low-pressure steam coming out of the recuperator (line 14) flows toward the condenser 1 (Cond_1) at the first level 1 and is cooled. The condensed organic working fluid is then sent from line 15 to the feed pump 1 (F_Pump_1). On the other hand, the air is used to condense the organic working fluid and is directed from line 23 to line 24 with the help of the fans. Thus, the organic working fluid is cooled. For Cond_1, the mass, energy, and exergy balance equations are presented as follows.
122
PART II Thermodynamic analysis of geothermal power plants
At the first level of the binary GPP, there are 33 fan units in total, with three rows and 11 in each row. The energy and exergy equations for the total fan units are expressed as follows: m_ 23 ðP24 P23 Þ W_ Fan ¼ r23 Fan _ y,Fan ¼ W_ Fan Ex _ 24 Ex _ 23 Ex eFan ¼
8.3.1.3.2 FIG. 8.11 Flow diagram of condenser 1 in the first level of the plant.
Mass balance: m_ 14 ¼ m_ 15
(8.64)
m_ 23 ¼ m_ 24
(8.65)
for the organic working fluid and the air as cooling fluid, respectively. Energy balance: The e-NTU method has been taken into account in order to model the change of inlet and outlet temperatures of Cond_1. Thus, l
For heat transfer during the cooling process C23 ¼ m_ 23 cp,23
(8.66)
C14 ¼ m_ 14 cp,14
(8.67)
C min ,Cond_1 ¼ min ðC23 , C14 Þ
(8.68)
Q_ Cond_1,cool ¼ eCond_1 C min ,Cond_1 ðT14 T23 Þ l
(8.69)
For heat transfer during the condensating process
_ 23 _ 24 Ex Ex W_ Fan
(8.76) (8.77) (8.78)
Condenser 2 (Cond_2)
Condenser 2 (Cond_2) on the second level of the binary GPP operates on the same principle as Cond_1. However, the organic working fluid coming out of Turb_2 without a recuperator is sent to Cond_2, and the cooling of the organic working fluid is carried out there. The flow diagram of Cond_2 for a binary GPP is shown in Fig. 8.12. The mass, energy, and exergy balances can be calculated by Mass balance: m_ 20 ¼ m_ 21
(8.79)
m_ 25 ¼ m_ 26
(8.80)
for the organic working fluid and the air as cooling fluid, respectively. Energy balance: Similar to equilibrium equations achieved for Cond_2, these can be derived. Thus, (8.81) E_ Cond_2 ¼ E_ 20 E_ 21 E_ 26 E_ 25 Cond_2 ¼
E_ 20 E_ 21 E_ 26 E_ 25
Exergy balance: _ 26 Ex _ 20 Ex _ 21 Ex _ 25 _ d,Cond_2 ¼ Ex Ex
(8.82)
(8.83)
QCond_1,conde ¼ m_ 15 h15g h15f ¼ C23 ðT24 T23 Þ (8.70) l
For total heat transfer Q_ Cond_1 ¼ Q_ Cond_1, cool + Q_ Cond_1, conde E_ Cond_1 ¼ E_ 14 E_ 15 E_ 24 E_ 23 Cond_1 ¼
E_ 14 E_ 15 E_ 24 E_ 23
(8.71) (8.72) (8.73)
Exergy balance: _ 14 Ex _ 15 Ex _ 23 _ 24 Ex _ d,Cond_1 ¼ Ex Ex ex,Cond_1 ¼
_ 15 _ 14 Ex Ex _ 23 _ 24 Ex Ex
(8.74) (8.75) FIG. 8.12 Flow diagram of condenser 2 in the second level of the plant.
Binary geothermal power plant Chapter 8
ex,Cond_2 ¼
_ 21 _ 20 Ex Ex _ 25 _ 26 Ex Ex
123
Mass balance: (8.84)
There are 36 fan units for three rows and 12 in each row in the second level of the GPP. Its energy and exergy equations can be easily found by performing calculations similar to the equilibrium calculations of the fan units in Cond_1.
m_ 15 ¼ m_ 16
(8.85)
for the organic working fluid. Energy balance: W_ F_Pump_1 ¼ m_ 15 ðh16 h15 Þ ¼ m_ 15 F_Pump_1,is ðh16,is h15 Þ (8.86) F_Pump_1 ¼
8.3.1.4 Feed pump The pump is a device used to pressurize liquids. The pumps can be used in a wide range of applications such as wells, cooling towers, the automobile industry, fuel injection, and the medical industry. The pumps regulate the flow and transmission of the working fluid in a system. All pumps operate according to the principle of volume increase or decrease. Due to this pressure difference, both suction and discharge processes take place. In ORCs, a multistage centrifugal feed pump is used to transmit the organic working fluid from condensers to preheaters. The energy consumption of the feed pumps is calculated by assuming insignificant potential and kinetic energy in the steady state and constant flow in adiabatic operation.
8.3.1.4.1 Feed pump 1 (F_Pump_1) The flow diagram of the feed pump is shown in Fig. 8.13. After the organic fluid exits the turbine at a lower pressure and cools in the recuperator and condenser, respectively, it is pumped by feed pump 1 (F_Pump_1) to a higher pressure again. The mass, energy, and exergy balances for the F_Pump_1 can be managed by the following equations.
FIG. 8.13 Flow diagram of feed pump 1 in the first level of the plant.
h16,is h15 h16 h15
Exergy balance: _ 16 Ex _ 15 _ d,F_Pump_1 ¼ W_ F_Pump_1 Ex Ex ex,F_Pump_1 ¼ 8.3.1.4.2
_ 15 _ 16 Ex Ex _ W F_Pump_1
(8.87)
(8.88) (8.89)
Feed pump 2 (F_Pump_2)
The organic working fluid condensed in Cond_2 is pressurized with F_Pump_2 and sent to the preheater to be heated. The flow diagram of F_Pump_2 is shown in Fig. 8.14. According to this flow diagram, the mass, energy, and exergy equations are given below. Mass balance: m_ 21 ¼ m_ 22
(8.90)
for the organic working fluid. Energy balance: W_ F_Pump_2 ¼ m_ 21 ðh22 h21 Þ ¼ m_ 21 F_Pump_2,is ðh22,is h21 Þ (8.91) F_Pump_2 ¼
h22,is h21 h22 h21
(8.92)
FIG. 8.14 Flow diagram of feed pump 2 in the second level of the plant.
124
PART II Thermodynamic analysis of geothermal power plants
Exergy balance: _ d, F_Pump_2 ¼ W_ F_Pump_2 Ex _ 22 Ex _ 21 Ex ex,F_Pump_2 ¼
_ 21 _ 22 Ex Ex _ W F_Pump_2
8.3.1.6 Selection of organic working fluid (8.93) (8.94)
8.3.1.5 Overall system After analyzing each of the components of the binary GPP from a thermodynamic perspective, the entire plant can be summarized by looking at it as a whole. The performance of the cycle is evaluated with the First Law of Thermodynamics using energy efficiency: th ¼
W_ net W_ net ¼ _ _ E input E 1 + E_ 10
(8.95)
The above-mentioned efficiency is used for the cycle, not the plant. If net cycle power is used to meet facility auxiliary power needs such as well pumps, cooling fans, and station lighting, all these parasitic loads must be removed from the net cycle power to obtain net plant power. Because binary cycles tend to have energy efficiency in the range of 8%– 15% [14], any decrease in net power can have a serious impact on plant performance. On the other hand, exergy efficiency is an important indicator for GPPs according to the Second Law of Thermodynamics. It can be defined as the ratio of real net power plant power to the maximum theoretical power that can be obtained from the geothermal fluid in the reservoir state, as stated below: W_ net W_ net ¼ ex ¼ _ 10 _ input Ex _ 1 + Ex Ex
(8.96)
FIG. 8.15 The temperature-entropy diagram of the first level for the plant.
The choice of organic working fluid has major implications for the performance of a binary GPP. Although there are many options for these fluids, there are many restrictions on health, safety, and environmental impact in addition to the thermodynamic properties of fluids. For the critical temperature and pressure values of 374°C and 22 MPa of water, respectively, these values of all candidate organic working fluids are much lower than those of water [15]. Especially for hydrocarbons, supercritical cycles can be considered. These provide a better match between the cooling curve of the geothermal fluid (brine) and the heating-boiling line of the working fluid. This situation reduces the thermodynamic losses in evaporators (e.g., vaporizer, preheater) as a heat exchanger. Moreover, the position of the saturated vapor curve becomes important on the entropy-temperature curve of the fluids, especially in the turbine-related process. It requires excessive heating before the turbine to avoid excessive humidity at the turbine outlet according to the selected working fluid. Some hydrocarbon and organic working fluids with a positive slope of the saturated vapor line can be used. These are clearly observed in Figs. 8.15 and 8.16 in Section 8.4. A binary GPP using n-pentane as the organic working fluid is selected. The critical temperature and pressure of n-pentane as a hydrocarbon are nearly 194°C and 3.2 MPa, respectively.
8.4
Calculation procedure
As mentioned before, the mass, energy, and exergy balance equations of all system components are given in detail to evaluate the thermodynamic performance of a binary GPP
Binary geothermal power plant Chapter 8
125
FIG. 8.16 The temperature-entropy diagram of the second level for the plant.
with double pressure. To solve all these equations, some conditions and assumptions are taken into account. Thus, the analysis becomes simpler. Some assumptions and conditions managed in the thermodynamic analysis are listed below: l
l
l
l
l
l
l
l
l
l
All processes occur within the steady-state and stable volume conditions. Differences in kinetic and potential energy are neglected. Heat loss occurring in components and pressure drops within valves and pipes are not considered. Thermodynamic properties of pure water are used for geothermal fluid (brine). Air is assumed to be an ideal gas with homogenous distribution in air-cooled condensers. Isentropic efficiencies of turbines, pumps, and fans are used. Direct drive balancing between the generator and turbines is ignored by opening injection valves on turbines. X 0.98 and X 0.02, respectively, in the quality of steam (X) to prevent excessive damage to the blades of the turbine and pump are accepted. Geothermal fluid temperature for reinjection does not fall below 80°C. DTpp > 2°C for pinch points (pp) of evaporators defined by the heat exchanger manufacturer is used.
On the binary GPP with double pressure discussed in this study, real data (pressure, temperature, and mass flow rate) recorded by the Supervisory Control and Data Acquisition (SCADA) program of the plant are collected momentarily for the state points shown in Fig. 8.1. However, the average daily or hourly values can be used because it takes a long
time to take measurements for the gas phase at some points, especially in organic working fluid lines. In another aspect, the flow measurements of gaseous working fluids are very costly. With this aim, the temperature, pressure, and mass flow rate collected on April 14, 2013, for each state within the binary plant are listed in Table 8.1. The dead state values for both water as the geothermal fluid and n-pentane as the organic working fluid are assumed to be 25°C and 101.325 kPa (see Table 8.1). It may be useful to demonstrate the thermodynamic process on a temperature-entropy (T-s) diagram using data collected from the plant. The T-s diagram of the closed-loop in the first level of the plant is given in Fig. 8.15. It is seen from Fig. 8.15 that the saturated vapor curve of n-pentane is a more positive slope than that of water. The n-pentane as the organic working fluid spends most of the heat it receives by passing through turbine 1 (Turb_1) at points 12–13 in order to generate energy. It should be noted that point 13 (the outlet of the turbine) is well above the saturated vapor curve. Therefore, at points 13–14, the aim is to decrease energy to be spent in the condenser 1 (Cond_1) by passing through the recuperator (Recup). The n-pentane is liquefied by passing through Cond_1 between points 14–15. It is pressurized by passing it through feed pump 1 (F_Pump_1) between points 15–16. Between points 16–17, the n-pentane passing through the Recup from F_Pump_1 is returned heat previously stored in the Recup. Between points 17–11, heat transfer is made from geothermal fluid to n-pentane passed through preheater 1 (Preheat_1). At point 12, the n-pentane should be supercritical before the turbine. This can be provided by a better match between the cooling curve of the geothermal fluid (brine) and the heating-boiling line of the organic working fluid. Different cases are encountered
TABLE 8.1 Thermodynamic variables recorded and calculated at the state points of the GPP. State no
Fluid type
Temperature, T (°C)
Pressure, P (kPa)
Enthalpy, h (kJ/kg)
Entropy, s (kJ/kgK)
Mass flow rate, ṁ (kg/s)
Exergy rate, E˙x (kW)
1
Brine
164
1040
692.43
1.98
445
52,693
0
Steam
165
1040
696.78
1.99
5.83
699
0
1
NCG
165
1040
629.69
2.64
2.50
380
2
Brine
136
730
573.07
1.70
445
36,010
3
Brine
110
690
459.30
1.41
445
22,589
4
Brine
110
690
459.30
1.41
222.50
11,295
5
Brine
110
690
459.30
1.41
222.50
11,295
6
Brine
89
590
371.51
1.18
222.50
7021
7
Brine
81
570
337.74
1.08
222.50
5616
8
Brine
85
590
354.54
1.13
445
12,447
9
Brine
107
690
448.50
1.38
0.83
40
10
Steam
107
690
448.50
1.38
5.25
253
10’
NCG
107
690
575.29
2.58
2.25
257
11
n-Pentane
105
1261
176.25
0.51
160
4776
12
n-Pentane
137
1261
516.04
1.35
160
20,142
13
n-Pentane
82
150
44.76
1.36
160
7237
14
n-Pentane
60
150
398.48
1.24
160
6193
15
n-Pentane
31
150
11.44
0.04
160
123
16
n-Pentane
37
1261
3.07
0.00
160
520
17
n-Pentane
55
1261
47.00
0.14
160
1141
18
n-Pentane
106
687
179.16
0.52
169
5018
19
n-Pentane
109
687
469.92
1.28
169
16,512
20
n-Pentane
69
119
416.78
1.32
169
5713
21
n-Pentane
33
119
7.00
0.02
169
157
22
n-Pentane
39
687
5.93
0.02
169
431
23
Air
25
101
424.29
3.88
2000
0
24
Air
25.5
106
424.70
3.87
2000
8232
25
Air
25
101
424.29
3.88
2000
0
26
Air
25.5
106
424.70
3.87
2000
8232
1
Note: The dead state values are 25°C and 101.325 kPa for temperature and pressure, respectively.
Binary geothermal power plant Chapter 8
with each change of critical pressure. The lines of geothermal fluid (brine) and air on the T-s diagram are added to the graph as a reference for understanding the process. A graphic similar to the T-s diagram given in Fig. 8.15 is also drawn for the second level of the plant. Thus, the T-s diagram of the cycle in the second level of the plant is presented in Fig. 8.16. The difference from the other graphic is that there is no recuperator in the cycle. In addition, it is the use of both geothermal fluid (brine) and its steam for vaporizer 2 (Vap_2) at points 18–19. Using the data listed in Table 8.1 and all balance equations in Section 8.3, a program that can perform a thermodynamic analysis of the plant can be written on the platform of MATLAB version R2015a, software by MathWorks [16]. There is a need for software programs such as Engineering Equation Solver (EES), REFPROP, CoolProp, etc., that can call the thermodynamic properties of fluids in MATLAB. Thus, the thermodynamic properties of any fluid for two known parameters such as hi ¼ f(Ti; Pi) can be taken from the mentioned software programs. For this purpose, the CoolProp program [17] is selected for the program in MATLAB. For this process, it can be used in thermodynamic tables and cards [18]. It is clear from the thermodynamic analysis presented in Section 8.3 that enthalpy and entropy values should be determined for the fluid at the state number specified in Fig. 8.1. It can be simplified by using a table presented in Table 8.1 to keep track of calculations in MATLAB. For example, enthalpy and entropy for n-pentane at state number 11 are found to be h11 ¼ fn-pentane(105 ° C; 1261 kPa) ¼ 176.25 kJ/kg and s11 ¼ fn-pentane(105 ° C; 1261 kPa) ¼ 0.51 kJ/kgK, respectively, in Table 8.1. Using
127
Eq. (8.5), the unit exergy rate for the fluid at the state number is calculated, as listed in the last column of Table 8.1.
8.5
Results and discussion
In this section, the thermodynamic performance of a binary GPP with double pressure is evaluated. Using the methodology described above in detail, exergy analysis is carried out. In this analysis, first, data such as temperature, pressure, and mass flow are obtained by measuring instruments or devices over the plant. The sensitivity of the instruments or devices that measure these data is an important indicator in the real and accurate evaluation of the analysis results. The uncertainties of all these data can be calculated separately via the method proposed by Kline and McClintock [19]. For the measured data such as temperature, pressure, and volumetric flow rate, their total uncertainties are 0.69% (°C), 1.21% (kPa), and 1.46% (m3/s), respectively. Therefore, the exergy efficiency of the overall plant is obtained as 1.54%. The exergy analysis is applied for the plant whose values are given in Table 8.1. The exergy analysis results of the plant are listed in Table 8.2. As seen in Table 8.2, an exergy input of approximately 53,772 kW from the geothermal fluid has occurred to the system. 15,866 kW of this total exergy input amount is determined as exergy amount spent on reinjection, 16,877 kW is exergy destruction that occurs in the plant component, and the remaining 257 kW is the exergy loss rate of NCGs released or stored in the atmosphere.
TABLE 8.2 The results of exergy and analysis for all components in the binary GPP. Component, k
Fuel exergy rate, E˙x (kW)
Produced exergy rate, E˙x (kW)
Exergy destruction rate, E˙x (kW)
Exergy loss rate, E˙x (kW)
Vaporizer 1
16,682
15,366
1317
–
Vaporizer 2
13,950
11,494
2456
–
Preheater 1
4274
3635
639
–
Preheater 2
5678
4587
1092
–
Condnser 1
6069
4094
1975
–
Condenser 2
5554
3063
2491
–
Recuperator
1041
621
420
–
Turbine 1
12,908
12,066
842
–
Turbine 2
10,802
9000
1802
–
Feed Pump 1
2324
398
1926
–
Feed Pump 2
2192
275
1917
–
Total Plant
53,772
21,029
16,877
15,866
128
PART II Thermodynamic analysis of geothermal power plants
FIG. 8.17 The results of an exergy analysis (Sankey diagram) for the plant.
The exergy destruction amounts of the plant components are clearly shown in Fig. 8.17. The highest exergy destruction in components occurs in condenser 2 (Cond_2) and vaporizer 2 (Vap_2) with 4.6%. They are followed by condenser 1 (Cond_1) with 3.7%, and feed pumps 1 (F_Pump_1) and 2 (F_Pump_2) with 3.6%. The power generation of the plant is 22% of the intake exergy of the geothermal fluid in turbine 1 (Turb_1) as 12 MW and 17% in turbine 2 (Turb_2) as 9 MW. The total electricity generation of the plant is approximately 21 MW. As seen in Fig. 8.17, the total exergy destruction amount of the entire plant is 16,877 kW. The most exergy destruction in the plant occurs at condenser 2 (Cond_2) with 2491 kW. It is followed by vaporizer 2 (Vap_2) with 2456 kW and condenser 1 (Cond_1) with 1975 kW. Thus, the improvement of these components mentioned is a priority. Another remarkable situation should be the improvement of fans and pumps.
FIG. 8.18 Change of exergy efficiency for plant components.
The exergy efficiencies of the power plant components are given in Fig. 8.18. The component with the highest exergy efficiency is turbine 1 (Turb_1) with 93%. It is then followed by vaporizer 1 (Vap_1) with 91% and preheater 1 (Preheat_1) with 85%. The components of the plant operated with the lowest efficiency are the feed pumps (F_Pump_1 and F_Pump_2). As seen in the figure, the components in the first level of the plant operate with higher exergy efficiency than those in the second level. The reason for this is the low geothermal fluid temperature in the lowpressure line. However, adding geothermal fluid steam to the low-pressure line is also not effective. The exergy efficiencies of turbines 1 and 2 (Turb_1 and Turb _2) are calculated as 93% and 83%, respectively. These types of components need to be improved due to the low efficiency of the condensers and pumps on both lines. The exergy efficiency is low in the use of lowtemperature geothermal resources. The exergy efficiency
Binary geothermal power plant Chapter 8
in the plant is calculated as 39.1%. As a result, 60.9% of the exergy amount entering the plant disappears as waste heat (exergy loss and destruction). On the other hand, the power spent in operating pumps, fans, and auxiliary equipment is met by the gross power generated by the plant. Therefore, the total exergy efficiency of the plant is determined as 14.48%. As a result of exergy analysis, it is seen that the most efficient components in the system are the turbines, evaporators, and preheaters. Considering the amount of exergy destruction, condenser 2 (Cond_2), vaporizer 2 (Vap_2), and condenser 1 (Cond_1) are the equipment that has priority in improvement. A high amount of exergy destruction concludes that these components need to be improved first. However, it is unclear in terms of whether it was due to the interaction between components or the technology levels of the component as the cause of exergy destruction. Therefore, it is not possible to determine how the improvement should be made.
8.6
Closing remarks
In this chapter, a binary geothermal power plant (GES) technology with an air-cooled organic Rankine cycle (ORC) was introduced. According to the first and second laws of thermodynamics for a binary GPP with double pressure, the mass, energy, and exergy balance equations were shown. As a case study, the flow characteristics and components of an existing plant were given. Exergy analyses were made according to the Second Law of Thermodynamics using data collected from the plant under real operating conditions. As a result of the exergy analysis applied to the plant, it was calculated that the highest exergy destruction occurs in reinjection as well as the condenser, vaporizer, and pumps. The total exergy input to the system is 53.8 MW, and 75.8% of it caused exergy destruction by the components. 23.7% of them were sent to reinjection, and its 0.5% were released to nature with NCGs. According to the results of the analysis, the total exergy production in the system was 21 MW, and the exergy efficiency of the system was 39.1%. As a result of the exergy analysis of the plant, the highest exergy destruction occurs in the condensers (Cond_1 and Cond_2). When the components with the highest exergy destruction were ranked, it occurred in condenser 2 (Cond_2), vaporizer 1 (Vap_1), condenser 1 (Cond_1), and the feed pumps (F_Pump_1 and F_Pump_2). Their values were calculated as 4.6%, 4.6%, 3.7%, 3.6%, and 3.6% of the total exergy input, respectively. In this order, the components that need improvement are determined.
129
References [1] Pfenninger S, Keirstead J. Comparing concentrating solar and nuclear power as baseload providers using the example of South Africa. Energy 2015;87:303–14. [2] Bleicher A, ve Gross M. Geothermal heat pumps and the vagaries of subterranean geology: energy independence at a household level as a real world experiment. Renew Sustain Energy Rev 2016;64:279–88. [3] Malafeh S, ve Sharp B. Role of royalties in sustanable geothermal energy development. Energy Policy 2015;85:235–42. [4] Flovenz OG. General aspects of geothermal energy. Geo-survey for the World Bank at renewable energy training program, Iceland: University of Iceland; 2012. [5] Lee KC. Classification of geothermal resources by exergy. Geothermics 2001;30:431–42. [6] Chamorro CR, Mondejar ME, Ramos R, Segovia JJ, Martin MC, ve Villamanan ve MA. World geothermal power production status: energy, environmental and economic study of high enthalpy technologies. Energy 2012;42:10–8. [7] Dickson M, ve Fanelli M. What is geothermal energy? Pisa, Italy: Instituto di Geoscienze e Georisorse, CNR; 2004. [8] Bertani R. Geothermal power generation in the world 2005–2010 update report. In: World Geothermal Congress, Bali, Indonesia; 2010. [9] EIA. Geothermal power plants, http://www.eia.gov; 2020. [Accessed 15 January 2020]. [10] Dipippo R. Geothermal power plants: principles, applications, case studies, and environmental impact, 2. Baskı. Oxford, UK: Elsevier Ltd; 2008. [11] Bertani R. Geothermal power generation in the world 2010–2014 update report. Geothermics 2016;60:31–43. [12] Kec¸ebas¸ A, G€okgedik H. Thermodynamic evaluation of a geothermal power plant for advanced exergy analysis. Energy 2015;88:746–55. [13] G€okgedik H, Y€ur€usoy M, Kec¸ebas¸ A. Improvement potential of a real geothermal power plant using advanced exergy analysis. Energy 2016;112:254–63. € [14] Ozkaraca O, Kec¸ebas¸ P, Demircan C, Kec¸ebas¸ A. Thermodynamic optimization of a geothermal-based organic Rankine cycle system using an artificial bee colony algorithm. Energies 2017;10(11):1691. [15] Reynolds WC. Thermodynamic properties in SI: graphs, tables and computational equations for 40 substances. Stanford, CA: Department of Mechanical Engineering, Stanford University; 1979. [16] MathWorks, Matlab, http://www.mathworks.com/products/matlab/. [Accessed 15 October 2019]. [17] Bell IH, Wronski J, Quoilin S, Lemort V. Pure and pseudo-pure fluid thermophysical property evaluation and the open-source thermophysical property library coolprop. Ind Eng Chem Res 2014;53(6):2498–508. [18] C ¸ engel YA, Boles MA. Thermodynamics: an engineering approach. Boston: McGraw-Hill; 2001. [19] Kline SJ, McClintock FA. Describing uncertainties in single-sample experiments. Mech Eng 1953;75:3–8.
Chapter 9
Solar-geothermal power plants Kosmas A. Kavadias, Panagiotis Alexopoulos, and George Charis Laboratory of Soft Energy Applications and Environmental Protection, Mechanical Engineering Department, University of West Attica, Athens, Greece
9.1
Introduction
Energy plays a leading role in human existence, survival, and development, and consequently, the population increase and standards of living inevitably lead to increased energy needs. Conventional energy resources, which currently account for 80% of the global energy demand, are continually declining, leading to uncertainty over the future of energy. This fact is a significant problem in terms of the sustainability of energy resources and can cause a failure to meet the future energy demand over the next hundred years. This situation has led the international community toward alternative energy solutions such as renewable energy. According to the IRENA Global Energy Transformation Roadmap to 2050 report about the future of the energy systems’ decarbonization, the renewable energy sources (RESs) share should range above 70% by 2050 [1]. The same report states that there is a clear correlation between the energy demand, the energy efficiency, and the share of renewable energy while scenarios with high shares of renewable energy are those with higher efficiency and therefore lower overall energy demand. In this direction, geothermal energy can significantly contribute to global energy demand. An essential advantage of geothermal energy is that geothermal power stations are baseload systems capable of providing constant power. Additionally, it is a predictable energy source that is not affected by weather conditions. According to the Electric Power Research Institute, the theoretical heat stored in the rocks down to 3 km depth within the crust is estimated as 4.3 1013 TJ [2, 3], equivalent to the world energy consumed for almost 100,000 years [4]. The amount of heat within 10,000 m of the Earth’s surface is estimated to contain 50,000 times more energy than the sum of the world’s oil and gas resources [5]. The total heat content of the Earth is on the order of 12.6 1021 TJ and of the crust 5.4 1018 TJ [6]. From the high thermal energy of the Earth, only a fraction of this energy can be exploited. So far, the utilization of geothermal energy is limited to areas where geological conditions enable a thermal energy carrier to transfer the heat from the deep hot zones to near the surface and
therefore create geothermal resources. According to [7], the expected geothermal electricity potential ranges from a minimum of 35–70 GWe to a maximum of 140 GWe; but the potential may be orders of magnitude higher, based on the enhanced technology of geothermal systems. According to [4], the most likely value for the technical potential of geothermal resources suitable for electricity generation is 210 GWe. Theoretical estimations indicate that the magnitude of hidden resources can be 5–10 times larger than the estimate of identified resources, whereas conservative estimates set the technical potential for geothermal power production at 200 GWe. Bromley et al. [8] state that the maximum technical potential of geothermal energy is far higher due to the high heat content of the Earth’s crust and the high natural flow of heat to the Earth’s surface. Nevertheless, the future of the geothermal energy growth rate will be mainly influenced by economics, the energy demand, the material constraints, and the social factors [8]. The undiscovered geothermal resources, that is, offshore resources and areas with no prominent surface expression, could further increase the available potential up to 1500 GWe in the future. By considering fracturing and extracting heat from all the deep crustal rocks, the global geothermal technical potential is estimated at 1000–2000 GWe [8]. Geothermal energy is a renewable energy source with a dynamic presence in the global power grid over the last 40 years. The installed capacity of geothermal power plants reached 13.9 GW by the end of 2019 [9] with a mean annual growth rate during the last decade of 4% [2]. A large proportion of installed geothermal capacity is located in island nations or regions (43%), providing not only a valuable source of electricity generation for these regions but also heat for a wide range of applications [2]. The world geothermal heat use reached 563,000 TJ in 2014 [2]. The annual electricity production was 88.4 TWh in 2018, amounting to approximately 1% of global electricity generation [9]. In terms of the reliability of the geothermal energy fields, a disadvantage remains the long-term reduction in resource productivity [4, 5]. The geothermal production fluid’s temperature decline has an important impact on power plant performance because a reduction in heat input
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00010-X Copyright © 2021 Elsevier Inc. All rights reserved.
131
132
PART II Thermodynamic analysis of geothermal power plants
to the power block decreases not only the available thermal energy but also affects the power plant’s efficiency. Moreover, it is well known that high temperatures are required for power generation, which is a major drawback for the medium-low geothermal energy fields. The decline in power generation impacts the economic efficiency of the project because the reduction in energy production leads directly to a decrease in revenues from energy sales. Additionally, cases with power purchase agreement contracts could be subject to economic penalties if power generation falls below a specific level. Those penalties could even lead the plant operator to purchase the power deficit from elsewhere. Table 9.1 presents data from different geothermal power plants around the world, indicating a decrease in power production during their operation [10]. According to the data, the mean annual decrease since the beginning of operation ranges between 0.3% and 2% per year, indicating the risk of investment in geothermal power generation. To address those drawbacks, a combination with other renewable sources, comprising a hybrid energy system,
TABLE 9.1 Decline in power generation in worldwide geothermal power plants [10].
Plant
Start
Installed capacity (MWe)
Running capacity (MWe)
CaliforniaThe Geyser (US)
1971
1529.0
833.0
Cerro Prieto (MX)
1973
150.0
131.0
Oita (Hatchobaru) (JP)
1977
110.0
80.0
Kizildere (TC)
1984
20.4
10.0
Larderello (IT)
1985
542.5
411.7
CaliforniaHeber (US)
1985
52.0
47.0
Travale/ Radicondoli (IT)
1986
160.0
126.6
CaliforniaEast Mesa (US)
1989
37.0
34.2
Nevada (Brady) (US)
1992
26.0
20.0
Miravalles (CR)
1993
144.0
132.5
can overcome disadvantages and operate by complementing each other to achieve better overall performance. Hybrid energy systems combine different sources with satisfying energy demand. Although one might consider that a country’s electricity grid constitutes a hybrid energy system (because load demand is met by different energy sources, e.g., coal, natural gas, etc.), the term was established and used for smaller-scale electricity systems that necessarily include renewable energy sources [11]. Geothermal energy can be substantially combined with all other renewable energy systems to form a hybrid renewable energy plant. Nevertheless, the most interesting combination is with solar energy and, more specifically, with solar thermal power systems that have a direct effect on the operation of the geothermal power plant. In these cases, combined heat production from two different resources optimizes the energy efficiency of a conventional thermal cycle. The use of geothermal and solar energy is strengthened by the fact that many of the areas with high geothermal potential also possesses high solar potential (Fig. 9.1). Solar energy is considered the most abundant and longterm resource of all other renewable energy sources. It is the primary energy source for the wind, biomass, waves, and ocean thermal energy. The Earth’s surface receives about 1.7 1014 kW of solar radiation [12]. Humans have realized the power of the sun since the early stages of our existence. The utilization of the sun’s power begun in the 7th century BC when a magnifying glass was used to concentrate the sun’s rays to produce fire. The first large-scale solar applications were based on the concentration of solar radiation to produce high-temperature thermal energy for melting iron, copper, and other metals. Back in the 18th century, the French chemist Lavoisier built the first large-scale solar application, a solar furnace made of powerful lenses to concentrate solar radiation to produce a heat of 1750°C [13]. The technology of concentrating solar radiation to increase solar energy density has been developed since then, and nowadays, large concentrated solar power plants are installed for indirect electricity generation. These installations are combined with conventional power units to provide heat to the thermal cycle of a steam or gas turbine for producing electricity by substituting the conventional boiler and minimizing the use of fossil fuels. Solar power plants using concentrated technology have been growing rapidly in recent years, and technology options include many different concentrator devices. The concentration ratios achieved (i.e., the ratio of collector aperture area to receiver area) can vary over several orders of magnitude from a few tens up to 10,000 [14]. The thermal fluid in concentrating solar collectors can achieve temperatures that allow high thermodynamic energy conversion in thermal cycles.
Solar-geothermal power plants Chapter 9
133
FIG. 9.1 World geothermal (A) and solar energy potential maps (B). (Soruce: (A) Limberger J, Boxem T, Pluymaekers M, Bruhn D, Manzella A, Calcagno P, et al. Geothermal energy in deep aquifers: a global assessment of the resource base for direct heat utilization. Renew Sustain Energy Rev 2018;82:961–75. doi:10.1016/j.rser.2017.09.084. (B) Global Solar Atlas. https://globalsolaratlas.info/map?c¼11.523088,8.173828,3.)
134
PART II Thermodynamic analysis of geothermal power plants
The idea of hybridizing a geothermal power plant with a concentrated solar power unit in areas that possess a high potential of both resources is a promising solution for further exploitation of geothermal energy [2, 3]. The use of concentrated solar collectors, for example, to mitigate the resource productivity decline, avoids the risk of geothermal field expansion, which may or may not provide geothermal fluid with the desired temperature and flow properties. Moreover, the thermal capacity of the solar field could be dynamically matched with long-term geothermal resource productivity reduction by expanding the field, thus allowing existing geothermal power plant installations to operate closer to their capacity while at the same time extending their operational life. The challenge in designing solar-geothermal power plants lies in the intermittent nature of solar energy versus the constant nature of geothermal energy. If there is sufficient storage for thermal energy from the sun, then the two energy sources could become fully compatible. Hybrid geothermal power plants use the same basics as the standalone cases but combine a different heat source into the process, for example, heat from concentrating solar collectors. This heat is added to geothermal brine, increasing its temperature and the power production or added to the geothermal brine before reinjecting it into the well to maintain the productivity of the geothermal source. Other hybrid configurations include preheating and superheating options for organic Rankine cycles (ORCs) and direct steam generation (DSG) in which the geothermal brine gets vaporized inside the solar field. The Stillwater project in the United States, operated by ENEL Global Renewable Energies, has launched such a hybrid system, combined concentrating
solar thermal and geothermal energy and also including a photovoltaic power plant [5].
9.2 The concentrating solar thermal power plant Concentrating solar thermal power plants use mirrors (reflectors) to concentrate beam radiation to a receiver to produce heat able to drive conventional steam and gas turbine cycles for electricity production. The concentration of solar radiation reduces the required absorption surface of the collector and, therefore, significantly reduces the overall heat loss. Concentrated radiation heats the fluid circulating through the tubular receiver, thus converting solar radiation into thermal energy in the form of the sensible or latent heat of the fluid. The fluid temperatures achieved provide significant amounts of thermal energy to be exploited either as heat or for electricity generation. The global technical potential of concentrating solar power amounts to almost 3,000,000 TWh annually [15]. The concentrating solar power global capacity reached 5.5 GW by the end of 2018 with dominant technology, the parabolic troughs [16]. Spain and the United States remain the market leaders, although China and Morocco led the market to new additions in 2018 [16]. According to SolarPACES (a program of the International Energy Agency [17]), more than 500 MW were added in 2019, surpassing 6 GW of total installed power operating worldwide. That number is expected to reach more than 9 GW by taking into account the projects under construction and development (Fig. 9.2) [18].
FIG. 9.2 Worldwide concentrated solar power plants that are operational, under construction, or in development. (Data from IEA. CSP projects around the world. SolarPaces; 2020. https://www.solarpaces.org/csp-technologies/csp-projects-around-the-world/ (Accessed 18 July 2020).)
Solar-geothermal power plants Chapter 9
Parabolic trough collectors (PTCs) generate hightemperature thermal energy. A PTC system is a linearfocused solar collector consisting of a parabolic mirror that reflects direct solar radiation onto a tubular receiver located in the focal line of the parabola (Fig. 9.3). The heat transfer fluid temperatures ranges between 100°C and 450°C [19]. The collectors rotate around an axis following the daily motion of the sun. The rotation of the collector around its axis is controlled by a tracker that detects and monitors the position of the sun. The arrangement of the PTC solar field depends on the nominal operating power, the heat transfer fluid, and the use (direct or indirect steam production). Solar collectors are arranged in rows, and the whole field is usually rectangular-shaped, almost square. The production unit is located near the center of the solar field for reducing the heat losses. The heat transfer fluid is directed from the solar field to the center block, and back to the solar collectors following the same circular procedure. The collectors are structured of silver-coated glass mirrors. A typical aperture is in the range of 1–2 m width [19]. They consist of glass coated with zinc and melt under controlled pressure and temperature conditions. The result of this process is a thin and smooth glass without optical distortions, with corrosion resistance, and ease of shaping for the construction of large collectors. A protective layer of copper covers the silver-coated surface, and layers of epoxy material are finally added. These materials provide construction hardness, high mechanical resistance to elastic
135
deformations, resistance to welding, high chemical resistance, and insulation. The standard PTC receiver consists of an inner stainlesssteel absorption tube covered by a three-layer liner. The inner layer is either copper, aluminum, or lead. The intermediate layer is a ceramic-metallurgical composition, and the outer layer is ceramic. The absorbance of the receiver is higher than 96%, and the thermal radiation emission coefficient does not exceed 10% at 450°C [20]. The steam in parabolic collector systems is produced either by direct or indirect methods. In the direct steam production method, the steam is produced directly from the solar field. The cold fluid is driven by a pump into the solar field, crosses the receiver lengthwise, changing its pressure and increasing its temperature, and then enters the steam turbine (Fig. 9.4). In indirect steam production (Fig. 9.5), the heat transfer fluid is circulated in the suction pipe, then enters a heat exchanger (in a closed circuit) where steam is generated for the steam turbine of the Rankine cycle. As stated earlier, in the direct steam generation systems, the heat transfer fluid is the water/steam while in the majority of the PTC indirect steam production systems, the heat transfer fluids used are mainly synthetic oils that achieve temperatures up to 400°C. Synthetic oils are characterized by their affordable price, low vapor pressure, good thermal stability, and long lifetime. Except for the synthetic oils, the molten salt is also used with the operating temperature reaching 560°C [19]. The molten salt is a high heat capacity fluid enhancing the integration of the thermal
FIG. 9.3 Single solar parabolic trough collector unit. (Source: HelioTrough – TSK Flagsol. http://www.heliotrough.com/english/history/history.html.)
136
PART II Thermodynamic analysis of geothermal power plants
FIG. 9.4 Direct steam production PTC system.
FIG. 9.5 Indirect steam production PTC system.
energy storage into the plant with a lower investment cost due to the higher difference between the cold and hot tanks [20]. However, the higher temperatures achieved increase the thermal losses of the fluid. It is quite corrosive with a high coagulation temperature compared to synthetic oils, making the use of antifreezing necessary. The pressurized
gases can also be used as heat transfer fluids. They are environmentally friendly and can operate with thermal stability at higher temperatures. The heat transfer fluids must have a sufficiently high evaporation temperature at normal pressures, as this criterion determines the operating temperature of the steam
Solar-geothermal power plants Chapter 9
137
TABLE 9.2 Properties of common heat transfer fluids [20]. Fluid
Temperature range (°C)
Properties
Synthetic oil
12–400
Relatively high operating temperature, flammable
Mineral oil
10 to 300
Relatively cheap, flammable
Water/steam
0 to >500
Require high pressure and thick walls
Silicone oil
40 to 400
Odorless, nontoxic, expensive, flammable
Nitrate Hitec
142–538
High coagulation and operating temperature, corrosive
Nitrate Hitec XL
120–500
Nitrate Hitec Solar Salt
238–593
Ionic liquids
75 to 416
turbine cycle and, subsequently, its efficiency (Table 9.2). It should also have low viscosity, high conductivity, and heat capacity to facilitate the process of heat exchange in the exchanger. Low coagulation temperatures are preferable to avoid the need for protection during periods of low temperatures. It should also be less flammable or explosive as possible and economical.
9.3
Hybrid solar-geothermal plants
The basic idea behind a hybrid installation is to achieve a combined result where the energy produced is superior to the corresponding energy of the individual parts (solar and geothermal). The hybrid unit should be designed in such a way that the two energy sources complement each other.
Good thermal properties, expensive
Ideally, the best result would be obtained if the sources were both always available to achieve a higher degree of concurrence. For the geothermal-solar hybrid systems, the solar source is intermittent in contrast to the geothermal energy, as solar radiation depends on the sun availability (day-night) and the cloud cover that limits the incident solar radiation. Places with high fluctuations in cloud cover create severe practical limitations in the design of hybrid plants, as the production organization depends significantly on the weather forecast. From an economic point of view, solar and geothermal installations have significant differences, with concentrated solar power plants being more expensive than geothermal power plants (Fig. 9.6). Additionally, given their thermodynamic and economic efficiencies, any attempt to hybridize
FIG. 9.6 Comparative diagram for the initial cost and levelized cost of energy (LCOE) of CSP and geothermal power plants. (Data from Li K, Liu C, Jiang S, Chen Y. Review on hybrid geothermal and solar power systems. J Clean Prod 2020;250:119481. doi: 10.1016/j.jclepro.2019.119481.)
138
PART II Thermodynamic analysis of geothermal power plants
solar and geothermal energy would significantly favor geothermal energy. However, because the economic approach also depends on other parameters that vary around the world (labor costs, material costs, financial incentives, government subsidies, etc.), it would be prudent to consider possible hybrid configurations where they could be considered thermodynamically and economically efficient. The challenge in solar-geothermal hybrid plants is the ability to convert geothermal energy into electricity in a more economically viable way. An advantage in this direction is that the concentrated solar field does not need an energy conversion unit because with appropriate configurations, it can directly supply thermal load through heat exchangers in the geothermal power system. In this way, capital saving is achieved in contrast to geothermal units with low thermal energy potential, which is considered relatively unprofitable due to the low efficiency in the conversion of thermal energy into electricity [21]. The design of solar-geothermal units is similar to the integration with binary ORC power plants. The main options are presented in Table 9.3 and Figs. 9.7–9.13. Each of these options can also incorporate thermal energy storage (TES) in the solar power system [23, 26].
The geothermal preheating option (Fig. 9.7) is the least penetrating solution in terms of the additional geothermal equipment required because solar energy is simply added to the incoming geothermal fluid via a heat exchanger. This design was implemented at the hybrid solar-geothermal plant in Stillwater, Nevada, the United States, by the Italian company Enel Green Power [27]. The only difference is that this unit is an organic binary cycle instead of a flash steam unit (Fig. 9.8). With the preheating option, there is no change in the mass flow of the geothermal brine, while the thermal energy produced by the solar field increases the steam production. However, the amount of energy added to the system is limited by the conditions of the incoming brine and especially its temperature. The operating temperatures of the separator, the vaporizer, and the turbine play a decisive role in the maximum amount of thermal energy that can be given to the system from the solar field. In the organic cycles, the limitation of the added thermal energy lies in the conditions and temperature operating range of the organic fluid (isopentane, pentane, butane) [28]. It is worth noting that the relatively small temperature differential required in the heat exchanger limits the temperature range along the solar field. The result is the high mass flow
TABLE 9.3 Possible scenarios of integration of the solar field in a flash steam unit or ORC [23]. Reheating condensated geothermal brine and reinjecting in cold well (Figs. 9.9 and 9.10)
Instant boiling of geothermal brine and reinjection in the thermal cycle (Figs. 9.11 and 9.12)
There is no change to the mass flow rate of the geothermal brine inside the flash tank
Higher specific enthalpy inside the flash separator
There is no change in the dimensions of the flash tank
Higher specific enthalpy inside the vapor-water separator
Increased steam flow rate in the turbine
Reheating the condensed brine minimizes the possibility of scaling issues inside the brine/heat transfer fluid heat exchanger
Improved vapor quality to the turbine
Reheating the condensed brine minimizes the possibility of scaling issues inside the brine/heat transfer fluid heat exchanger
Much of the produced solar thermal energy of the solar field is sent to the power block as steam
Increased steam flow rate to the turbine
The extended temperature difference across the solar field
Ability to produce direct steam within the solar field and inject it into the power cycle
Possibility for scaling issues inside heat exchanger between brine and solar heat transfer fluid
Increased mass flow rate of the geothermal brine inside the flash tank
It is recommended to have a geothermal fluid treatment equipment before it enters the vaporizer
Specified temperature difference across the solar field depending on the inlet conditions of the geothermal brine
The energy of the solar field is not transferred directly to the thermal cycle but to the geothermal fluid that is reinjected into it (only for recirculation models)
Preheating geothermal brine with solar energy (Figs. 9.7 and 9.8) Advantages
Constraints
Solar-geothermal power plants Chapter 9
FIG. 9.7 Preheating option for geothermal brine with solar energy at a flash steam unit [22].
FIG. 9.8 Enel’s Stillwater hybrid facility preheating option for geothermal brine with solar energy in a binary organic cycle [23].
139
140
PART II Thermodynamic analysis of geothermal power plants
FIG. 9.9 Reheating of the condensed geothermal brine and mixing with the production well fluid [23].
FIG. 9.10 Reheating of the condensed geothermal brine and reinjection into the reservoir [24, 25].
FIG. 9.11 Vaporizing of the condensed geothermal brine through the boiler of the solar field and its injection into the steam intake network of the turbine.
FIG. 9.12 Direct steam generation of the condensed geothermal brine through the solar field with the use of an accumulator and partial recirculation of the water.
142
PART II Thermodynamic analysis of geothermal power plants
FIG. 9.13 Superheating of the heat transfer fluid of a binary organic cycle with thermal energy from a solar field.
of the heat transfer fluid of the solar field. The placement of the heat exchanger before the thermal cycle is likely to cause problems with the scaling and deposition of solids that inadvertently flow into the pumped geothermal fluid. Therefore, the presence of sediment control and pretreatment equipment is considered necessary at the entrance of the vaporizer [29]. The reheating scenario of the concentrated geothermal fluid (Fig. 9.9) and its reinjection into the thermal cycle has the same characteristics as the first option but does not incorporate its significant limitations. With the installation of reheating, clean water is pumped from the condenser and reheated until it reaches the inlet conditions of the geothermal fluid. This design allows for a higher temperature range within the solar field and the possibility of integrating thermal storage. Also, the scaling issues are very few, due to the relatively high quality of the thermal cycle condensate water. However, the presence of geothermal fluid recovery equipment at the condenser outlet to retain solid particles is recommended. An alternative configuration of the concentrated geothermal fluid reheating scenario is depicted in Fig. 9.10. The geothermal fluid after reheating is not mixed with the fluid coming from the production well but is injected with high pressure into the injection (cold) well. In this way, the underground reservoir is used as thermal storage,
achieving higher efficiency and longevity while maintaining a high level of thermal load [30]. Also, a doubleflash power cycle is presented to obtain the most of the geothermal source (for higher-enthalpy reservoirs) [24, 25]. There is no change in the size and temperature range of the solar field or in the treatment of sediments. It is worth mentioning that it is possible to assemble the two scenarios (Figs. 9.9 and 9.10) so that lower-power heat sources can be utilized or to optimize existing high-enthalpy geothermal installations whose reservoir temperatures are declining [31]. Two alternatives to the direct reheating scenarios of the condensed geothermal fluid and its reinjection into the thermal cycle are presented in Figs. 9.11 and 9.12. In these scenarios, the geothermal fluid after condensation evaporates at the boiler of the solar field. The first uses a preheater and vaporizer with thermal storage [32] while in the second, the condensed geothermal fluid recovers its thermal energy and eventually evaporates through the solar field. The second scenario is also called direct steam generation within the solar field, according to [22]. The main advantage of the technology is the direct production of steam outside the thermal cycle and its injection into the steam intake network of the turbine. It is worth noting that the DSG system is considered preferable between the two because during the design, it can be dimensioned in such a
Solar-geothermal power plants Chapter 9
way that thermal storage of the heat transfer fluid is not required. This option is not included in the scenario of Fig. 9.11 because the conditions of the heat transfer fluid in the solar field should remain constant (mass flow rate, temperature, and pressure). Besides, for the DSG system, it is recommended to use a separator before injection into the thermal cycle. Before the fluid enters the solar field, boiler, feedwater treatment, and discharge of condensed gas equipment are considered necessary for the operation of the circuit within the pipeline network of solar panels [33]. Finally, Fig. 9.13 shows the corresponding scenario for an organic binary thermal cycle, with the difference that the energy transferred from the heat exchanger to the grid acts as a superheater [34, 35].
9.4
Operational analysis
Three types of geothermal power cycles are detected in the hybrid solar-geothermal configurations: the single-flash steam cycle with one vapor-water separation unit, the double-flash cycle with two vapor-water separation units, and the organic binary cycle with two different heating transfer fluids. The basic equations for the thermodynamic cycles, the solar field, and the hybrid configuration are presented below according to [19, 22, 29, 30, 33, 36–38].
9.4.1
Single-flash geothermal unit
In the single-flash thermal cycle, the saturated geothermal fluid from point 0 on the saturation curve (Fig. 9.14) is vaporized at a constant enthalpy to point 1, where the separator is located. Changes in the kinetic or potential energy of the fluid are not considered, and the whole process is
143
considered adiabatic. The separation of steam water is done under constant pressure (lines 1–2 and 2–3). The quality of the steam and its mass flow rate is determined by the mass and energy balance of the separator as: m_ 1 ¼ m_ 2 + m_ 3
(9.1)
m_ 1 h1 ¼ m_ 2 h2 + m_ 3 h3
(9.2)
where x1, x2, x3 are the dryness fraction (quality) of the mixture at the corresponding states; m_ 1 , h1 are the mass flow rate and the specific enthalpy of the saturated geothermal fluid at the inlet of the separator; m_ 2 , h2 are the mass flow rate and the specific enthalpy of the saturated vapor at the outlet of the separator; and m_ 3 , h3 are the mass flow rate and specific enthalpy of the saturated liquid at the outlet of the separator. The power output W_ t and the isentropic efficiency T of the turbine are given as: W_ T ¼ m_ 2 ðh2 h4 Þ
(9.3)
actual output h2 h4 ¼ ideal output h2 h4s
(9.4)
T ¼
where h4 is the specific enthalpy value of the steam for the real process, and h4s is the isentropic value of the specific enthalpy of the steam for the isentropic process at the outlet of the turbine. The thermal power Q_ c rejected by the condenser is calculated as: Q_ c ¼ m_ 2 ðh4 h5 Þ
(9.5)
where h5 is the specific enthalpy value of the saturated liquid at the outlet of the condenser.
FIG. 9.14 Temperature-entropy diagram for single-flash geothermal plants [33].
144
PART II Thermodynamic analysis of geothermal power plants
The cooling water required for the condenser is given as [37]: m_ cw ¼ m_ 2
h 4 h5 Ccw DΤ
(9.6)
where Ccw is the specific heat and DT is the temperature increase of the cooling water passing through the condenser. The power W_ cp consumed by the cooling system is determined by the equation: 1 m_ cw Dpcp W_ cp ¼ rcw cp
(9.7)
where rcw is the density of the cooling water, Dpcp is the pressure increase provided by the pump, and cp is the isentropic efficiency of the pump. The energy conversion efficiency o for the heat engine is calculated by: o ¼
W_ net Q_ in
(9.8)
where W_ net is the net power output of the turbine (excluding the parasitic load of the cycle), and Q_ in is the thermal power provided by the geothermal fluid, given by: Q_ in ¼ m_1 ðh2 h1 Þ
9.4.2
(9.9)
Double-flash geothermal unit
The double-flash power cycle is an improvement of the single-flash cycle because it can produce 15%–25% more energy for the same geothermal fluid conditions. It is a more complex and expensive installation with higher maintenance requirements compared to the single-flash units, although this is compensated by the increased energy production. Double-flash units have similar features to
single-flash units but with key differences such as the existence of a second separator. The second separator is responsible for the additional production of steam and its injection in the turbine from the returning low-pressure geothermal fluid of the first separator. For the metric model, it is assumed that the turbine is a dual-admission, single-flow machine where the lowpressure steam is added to the steam path at an appropriate stage to merge smoothly with the partially expanded highpressure steam. Other designs, for example, two separate turbines, one for the high-pressure steam and one for the low-pressure steam, are also possible [36]. The processes of this design are shown in Fig. 9.15. The geothermal fluid in the saturated liquid phase after the first separator (point 3) is flashed. This is done through a throttle valve, and a low-pressure water-steam mixture of steam and brine is produced, leading to the second separator. It is worth noting that flashing process 3–6 is considered to be isenthalpic. The mass and energy balance in the flasher are expressed as: m_ 6 ¼ m_ 7 + m_ 8
(9.10)
m_ 6 h6 ¼ m_ 7 h7 + m_ 8 h8
(9.11)
By Eqs. (9.10) and (9.11), the mass flow rate m_ 8 at point 8 is calculated as: m_ 8 ¼ m_ 6
h 6 h7 h 8 h7
(9.12)
where m_ 6 and h6 are the mass flow rate and the specific enthalpy value at the outlet of the second separator; h7 is the specific enthalpy value of the saturated liquid at the outlet of the second separator; and h8 is the specific enthalpy of the saturated vapor at the outlet of the second separator.
FIG. 9.15 Temperature-entropy diagram for double-flash geothermal plants [36].
Solar-geothermal power plants Chapter 9
145
The low-pressure steam from the first separator is added to the turbine at the same time as the high-pressure steam gradually expanded at state 5. The energy balance for state 9 is then applied where the partially expanded high-pressure steam (state 5) and the low-pressure dry saturated steam (state 8) coexist within the low-pressure stages of the turbine. Therefore, the mass flow rate m_ 9 and the energy balance are given as:
where v_ a is the volumetric flow rate through the air-cooled condenser; DP is the pressure differential between the inlet and outlet of the condenser; is is the isentropic efficiency; and m is the efficiency of the fan. The overall efficiency o of the cycle is calculated by:
m_ 9 ¼ m_ 8 + m_ 5
(9.13)
m_ 9 h9 ¼ m_ 8 h8 + m_ 5 h5
(9.14)
where m_ w, f is the mass flow of the water in the flash tank, and hin and hout are the specific enthalpy values at the inlet and outlet of the thermal cycle (state 1 is the inlet and states 11 and 7 are the outlet). Streams from points 11 and 7 (Fig. 9.15) are mixing before injecting back to the geothermal reservoir.
where m_ 5 and h5 are the mass flow rate and the specific enthalpy value of the high-pressure steam; m_ 8 is the mass flow rate of the saturated vapor at the outlet of the second separator; and h9 is the specific enthalpy value at the inlet of the low-pressure stages of the turbine. The power output of a dual admission and double-flow turbine with distinct high- (W_ HT ) and low-pressure (W_ LT ) stages is given by: W_ HT ¼ m_ 4 ðh4 h5 Þ
(9.15)
W_ LT ¼ m_ 9 ðh9 h10 Þ
(9.16)
where h10 is the specific enthalpy value of the steam at the outlet of the turbine. According to [30, 32], the isentropic efficiencies for high- (HTis) and low- (LTis) pressure turbines are calculated as: x + x 4 5 HTis ¼ Td (9.17) 2 x + x 9 10 (9.18) LTis ¼ Td 2 where Td is the turbine efficiency operating under dry steam conditions, and x4, x5, x9, x10 are the dryness fraction (quality) of the mixture at the corresponding points. Furthermore, the overall power production W_ T from a dual admission turbine is: W_ T ¼ W_ HT + W_ LT
o ¼
9.4.3
W_ net m_ w, f ðhin hout Þ
(9.22)
Binary organic cycle
In the binary organic geothermal cycle, the geothermal fluid from the production wells exchanges heat through a heat exchanger with the heat transfer fluid of the organic cycle, which is usually fluid with improved thermodynamic properties and a low boiling temperature. Organic fluids such as isobutane, isopentane, etc., are ideal solutions for sites with low-enthalpy geothermal fluid or with limitations to the geothermal fluid mass flow. It is worth noting that the geothermal fluid does not change phase from liquid to steam in the exchanger, and therefore the scale formation is minimum [28, 38–40]. As shown in Fig. 9.16, the energy balance to the heat exchanger is expressed as: m_ g ðha hb Þ ¼ m_ w, f ðh1 h2 Þ
(9.23)
where m_ g is the mass flow of the geothermal fluid; m_ w, f is the mass flow rate of the organic heat transfer fluid; and ha, hb, h1, and h2 are the specific enthalpy values of the respective currents.
(9.19)
The net power output W_ net of the geothermal plant is calculated by: W_ net ¼ W_ T W_ cp W_ p
(9.20)
where W_ cp and W_ p are the power consumed by the cooling system and the power consumption of the pump of the thermal storage network, correspondingly. In the case of an air-cooled system, the W_ cp is calculated as: v_a DP W_ cp ¼ is m
(9.21)
FIG. 9.16 The two currents of the geothermal brine and the organic heat transfer fluid of the heat exchanger.
146
PART II Thermodynamic analysis of geothermal power plants
The heat transfer flow Q_ between the two currents is calculated as: ðTa T1 Þ ðTb T2 Þ Q_ ¼ U A Ta T1 ln ðTb T2 Þ
(9.24)
where U is the total heat transfer coefficient; A is the surface of the heat exchanger; and Ta, Tb, T1, and T2 are the temperature values at the respective currents, as shown in Fig. 9.16. The overall theoretical efficiency th of the cycle is given as: W_ net th ¼ (9.25) Q_ where W_ net is the net power output of the turbine.
9.4.4
Solar field
The thermal efficiency of the solar field strongly depends on the parabolic trough solar collectors’ performance. The net heat Qa absorbed by the receiver per unit of area is calculated as [19, 30]: Qa ¼ Ib g r τ a IAM TE GA MC DE
(9.26)
where Ib is the beam normal irradiance; g is the intercept factor; r is the reflectivity of the collector’s reflective surface; τ is the transmissivity of the glass cover; a is the absorptivity of the receiver selective coating; TE, GA, MC, DE are the efficiency values taking into consideration the tracking and twist error losses, the geometric accuracy losses, the mirror clearness losses, and the dust losses on the absorber, correspondingly; and IAM is the incident angle modifier, which includes the optical and geometric losses for an incidence angle higher than 0 degree, calculated as a function of the transversal and longitudinal angles. The heat losses Ql, r of the receiver per unit of area can be calculated as: Ql, r ¼ 0:141 Tm + 6:48 109 Tm 4
(9.27)
and Tm ¼
Tf in + Tf out Tair 2
(9.28)
where Tfin and Tfout are the temperature values (in K) of the thermal transfer fluid before and after the exchanger and Tair is the air temperature. The piping heat losses Ql, p per unit area are given as: Ql, p ¼ 10 0:001693 Tm 1:683 105 Tm2 + 6:78 108 Tm3 (9.29)
By taking into consideration Eqs. (9.26)–(9.29), the required solar collectors’ reflective area Ar is calculated by: Ar ¼
Qd Qa Ql, r Ql, p
(9.30)
where Qd is the heat load demand of the power cycle.
9.5 Comparative analysis of the hybrid designs To compare the hybrid configurations, the thermodynamic efficiency for both units (solar and geothermal) when operating simultaneously must be determined. However, as the solar thermal unit operates only when solar energy is available, only the geothermal unit operates during low or no sunshine periods. The thermodynamic performance is evaluated by an exergy analysis adopted by [29]. In the direct steam generation systems, the rate of exergy _ DE g of the geothermal fluid from the production wells flowing directly in the power plant under steady conditions or flowing in the heat exchanger in indirect steam generation systems is calculated as: (9.31) DE_ g ¼ m_ g hog hig T0 sog sig and correspondingly, DE_ f for the heat transfer fluid of the solar field: (9.32) DE_ f ¼ m_ f hof hif Τ 0 sof sif where m_g is the mass flow rate of the geothermal brine; hig and hog are the specific enthalpy values at the production and injection wells accordingly; sig and sog are the specific entropy at the production and injection wells accordingly; T0 is the dead-state temperature; m_ f is the mass flow rate of the heat transfer fluid; hif and hof are the specific enthalpy values at the entrance and exit of the solar-geothermal heat exchanger accordingly; and sif and sof are the specific entropy at the inlet and outlet of the solar-geothermal heat exchanger accordingly. The exergy efficiency h compares the energy production of the hybrid system with the exergy provided by both the geothermal brine and solar field heat transfer fluid. The h evaluates the performance of the hybrid system, calculated as: h ¼
W_ net DE_ g + DE_ f
(9.33)
where W_ net is the net power production of the thermal cycle. The exergy efficiency u compares the energy produced by the hybrid system due to the solar field contribution, with
Solar-geothermal power plants Chapter 9
the exergy provided by both the geothermal and solar energy. The u evaluates the performance of the solar field as part of the hybrid system, expressed as: u ¼
W_ net W_ g DE_ g + DE_ f
(9.34)
where W_ g is the geothermal power produced when operating without the solar field contribution. The efficiency of the solar field exclusively is expressed by the exergy efficiency u, s, which compares the energy produced by the hybrid system due to the solar field contribution with the exergy provided by the solar field heat transfer fluid, calculated as: u, s ¼
W_ net W_ g DE_ f
(9.35)
Accordingly, the efficiency of the geothermal plant is expressed by the exergy efficiency u,g, which compares the energy produced by the hybrid system due to the geothermal field contribution with the exergy provided by the geothermal brine, calculated as: W_ g u, g ¼ DE_ g
W_ net Q_ fg
(9.37)
where Q_ fg is the heat transferred to the thermal cycle by the combination of the geothermal brine and the solar heat transfer fluid.
9.6
The geothermal unit produces 95 MW using two singleflash units and one double-flash unit as thermal cycle vaporizers [43]. The pressure of the geothermal fluid in the production wells is between 4 and 7 bar, the brine temperature 154–160°C, and the quality of the steam of the geothermal fluid 15%–20%. The arrangement of the parabolic trough collectors consists of a solar field with dimensions of 300 m 400 m where steam is produced at a rate of 5.8 kg/s operating during the day from 9 a.m. to 5 p.m. The solar field capacity, if it were operated separately from the geothermal plant, would produce 2.5 MW of electricity considering an inlet pressure of 4.4 bar for the steam turbine. During plant operation, the steam generated from the solar field is added to separated dry steam from the production wells of the geothermal unit, thus increasing the amount of steam at the turbine (Figs. 9.17 and 9.18). Due to the high solar radiation values and the long sunshine duration combined with the rich geothermal potential available in El Salvador, the hybrid plant is favored.
9.6.2 Geothermal field “Stillwater” (9.36) at Nevada, United States
Finally, the theoretical efficiency th of the hybrid power plant is: th ¼
147
Hybrid solar-geothermal power projects
The most recent research and publications [26, 41, 42] focus on hypothetical scenarios rather than utilized power plants because a lot of challenges need to be addressed. Meanwhile, hybrid installations already exist around the world and are briefly presented and analyzed.
9.6.1 Geothermal field “Ahuachapan” at El Salvador The first attempt to combine a geothermal unit with solar parabolic trough collectors for thermal amplification was in 2008 at the Ahuachapan geothermal field in El Salvador by the public electricity company Lageo [42]. The maximum temperature of the geothermal field is 225°C.
The Enel Green Power plant in Stillwater, Nevada, is the first hybrid plant in the world to combine three renewable energy technologies in the same area, geothermal (binary organic cycle), solar thermal, and photovoltaic (Fig. 9.19). 17 MW of parabolic trough collectors were added in 2015 to the existing 33 MWe geothermal plant. The PTC field added 2 MWe to the existing installation. The thermal power system consists of a parabolic trough collectors with a central vacuum tube from which the demineralized water transfers thermal energy to the geothermal power cycle. The thermal energy is transferred to the geothermal fluid from the production wells to amplify the thermal load and eventually is transferred via a heat exchanger to the binary organic cycle with isobutane as the working fluid (Fig. 9.20). According to [23], the solar field of the parabolic collectors consists of 22 lines (11 loops) of parabolic concentrating solar panels with each line at 200 m and each parabolic mirror 6 m in length. The solar field is composed of 2772 parabolic mirror panels of 24,778 m2 reflective surface with a concentration ratio of 75. The total aperture area of the project is about 81,000 m2. In the same area, a 26 MW photovoltaic park has been installed since 2012. The solar PTC field increases energy production (when the thermal efficiency of the installation is at a minimum due to the air-cooled condenser of the binary plant), decreases the environmental footprint, and reduces emitted gases (Fig. 9.21).
148
PART II Thermodynamic analysis of geothermal power plants
FIG. 9.17 Solar field’s heat transfer fluid circuit of the Ahuachapan geothermal field at El Salvador [42].
FIG. 9.18 Thermal cycle’s circuit of the Ahuachapan geothermal unit at El Salvador [42].
Solar-geothermal power plants Chapter 9
149
FIG. 9.19 Enel Green Power’s Stillwater Hybrid Power Plant at Nevada, United States, the first triple hybrid plant worldwide. (Source: Enel Green Power. Stillwater solar geothermal hybrid project in Fallon, USA (Eng) [Video]. YouTube; 2016. Retrieved from https://youtu.be/U_kayrbkkus (3:23).)
FIG. 9.20 Hybrid configuration of Enel Green Power’s Stillwater Hybrid Power Plant at Nevada, United States. (Source: Wendt D, Mines G, Turchi C, Zhu G. Geothermal risk reduction via geothermal/solar hybrid power plants. Final Report. 2015. doi:10.2172/1245529.)
150
PART II Thermodynamic analysis of geothermal power plants
FIG. 9.21 Solar thermal field of Enel Green Power’s Stillwater Hybrid Power Plant at Nevada, United States. (Source: Enel Green Power. Enel green power stillwater (3:23) [Video]. YouTube; 2016. Retrieved from https://youtu.be/_mHDHJu5ZSE (0.22).)
9.7
Closing remarks
As mentioned earlier, the hybridization of solar and geothermal technologies is an ideal solution due to the high availability and strong correlation through the direct production of thermal energy. Advantages, disadvantages, and challenges have been reported in the literature [30, 44]. The major advantages of the combination of the two technologies include: l
l
l
l
l
The temperature increase of the geothermal fluid or the generated steam in case of overheating. For example, PTCs can reach temperatures above 500°C, increasing the steam temperature at the turbine inlet. Higher temperatures provide more options for thermal plants and higher efficiency values. Moreover, increased power generation will eventually result in higher energy production and higher capacity factor values. Sources of lower enthalpy can be utilized because the thermal power required is assisted by the available thermal power from solar panels. By reheating the geothermal fluid, a mean of thermal storing of the generated thermal energy of solar panels is provided. In binary organic cycle units that typically use aircooled condensers and their discharged load tends to decrease during the day and increase at night as the air cools, the solar thermal power plant minimizes their intermittent operation by adjusting the generated load and thus increasing the power generation, especially during the day when the greatest demand occurs. The hybridization of the two technologies reduces the space required to produce the same power from a PTC-only installation.
l
l
l
The temperature increase of the incoming geothermal brine from the solar field prevents scale formation, which is common in geothermal installations. The ability to extend the solar field reduces the uncertainty caused by the gradual reduction of the geothermal reservoir efficiency, increasing the flexibility of the hybrid installation as the power generated does not depend only on the geothermal source, with all the negative consequences caused. Many parts of solar and geothermal hybrids are common (power plant, auxiliary structures, administration building, cooling system, staff, etc.); therefore, reduced specific operation and maintenance costs and total specific costs are achieved.
On the other hand, disadvantages should be also considered, studied, and overcome to achieve higher reliability and penetration of solar-geothermal systems. Some of the disadvantages include: l
l
During the operation of the hybrid unit, continuous monitoring and regulation of the geothermal energy from the production well concerning the solar-thermal energy supplied to the system by the solar field is necessary, requiring advanced equipment and control systems. So far, the initial cost of the investment tends to be quite high, making the project noncompetitive in the short term. Moreover, in the low pressure and temperature of underground reservoirs, more energy from the solar field and consequently a larger solar field is required, which further increases the initial cost.
From the above analysis, it is concluded that hybrid solargeothermal plants have many benefits and can be considered competitive against other renewable energy sources under favorable conditions. They are more efficient than the
Solar-geothermal power plants Chapter 9
respective individual geothermal and solar thermal power units, as long as they add up their advantages and overlap their negative elements, in terms of economic profit and thermal efficiency. However, hybrid units of this type have not been implemented beyond a minimum because significant technical and nontechnical challenges must be overcome. These challenges include: l
l
l
l
The configurations of hybrid units vary depending on the characteristics of the geothermal field and their correlation with solar thermal power production. Despite the recent years of extensive design studies of such units, their implementation remains low mainly due to their complexity and high initial costs. A hybrid installation aims to increase the temperature of the geothermal fluid, maintaining the capacity factor of the solar field at high levels. However, there is a limitation in geothermal fluid temperature increase. Moreover, high-cost materials are needed to withstand and maintain high temperatures in the solar field and the geothermal unit components. Although solar-geothermal systems can increase the capacity factor of solar thermal units, they are prone to weather conditions and especially solar resource availability. Long periods of cloud cover pose a risk for the investment, especially in cases where the technical minimum limits of the power plant are not met. The integration of the solar thermal power plant into enhanced geothermal systems is challenging in combining geothermal sources with no sufficient geothermal energy operating a single geothermal power plant. Solar thermal power reduces the storage needs and the equipment costs. Nevertheless, the cost reduction may not be enough to make such a hybrid unit economically viable.
According to the above, solar-geothermal systems have multiple advantages over individual geothermal and solar thermal power units. It is necessary to address challenges and problems to achieve a high penetration ratio in the worldwide electrical power generation system.
REFERENCES [1] Anon. Global energy transformation: a roadmap to 2050 (2019 edition). IRENA; 2019. [2] WEC. World energy resources 2016. World Energy Council; 2017. [3] Roberts VW. Geothermal energy prospects for the next 50 years. California, USA: Electric Power Research Institute; 1978. [4] Stefansson V. World geothermal assessment. In: Proceedings World Geothermal Congress 2005, Antalya, Turkey; 2005. [5] IRENA. Geothermal power: technology brief. International Renewable Energy Agency (IRENA); 2017. [6] Dickson MH, Fanelli M, editors. Geothermal energy: utilization and technology. Paris: Unesco Publishing; 2003.
151
[7] Bertani R. Geothermal power generation in the world 2010–2014 update report. Geothermics 2016;60:31–43. https://doi.org/10.1016/ j.geothermics.2015.11.003. [8] Bromley CJ, Mongillo M, Hiriart G, Goldstein B, Bertani R, Huenges E, et al. Contribution of geothermal energy to climate change mitigation: the IPCC renewable energy report. In: Proceedings World Geothermal Congress, vol. 2010; 2010. [9] Anon. Data and statistics—IRENA resource, http://resourceirena. irena.org/gateway/dashboard/?topic¼4&subTopic¼16; 2020. [Accessed 15 July 2020]. [10] Zarrouk SJ, Moon H. Efficiency of geothermal power plants: a worldwide review. Geothermics 2014;51:142–53. https://doi.org/ 10.1016/j.geothermics.2013.11.001. [11] Kavadias KA. Stand-alone, hybrid systems. In: Sayigh A, editor. Comprehensive renewable energy, vol. 2. Elsevier B.V; 2012. p. 623–56. https://doi.org/10.1016/B978-0-0808-7872-0.00222-5. [12] Kalogirou SA, Kambezidis HD. Solar radiation. In: Sayigh A, editor. Comprehensive renewable energy, vol. 2. Elsevier B.V; 2012. p. 623–56. https://doi.org/10.1016/B978-0-0808-7872-0.00222-5. [13] Soteris A K. Solar energy engineering. 2nd ed. Academic Press; 2013. [14] Axaopoulos PJ. Solar thermal conversion. Active solar systems. Greece: Simmetria Publications; 2010. [15] Trieb F, Schillings C, O’Sullivan M, Pregger T, Hoyer-Klick C. Global potential of concentrating solar power. German Aerospace Centre (DLR); 2009. [16] REN21. Renewables 2019—global status report; 2019. [17] IEA. SolarPACES—solar power & chemical energy systems. https:// www.solarpaces.org/ (Accessed 18 July 2020). [18] IEA. CSP projects around the world. SolarPaces; 2020. https://www. solarpaces.org/csp-technologies/csp-projects-around-the-world/. [Accessed 18 July 2020]. [19] Lovegrove K, Stein W. Concentrating solar power technology: principles, developments and applications. Elsevier; 2012. [20] G€unther M, Joemann M, Csambor S, Guizani A, Kr€uger D, Hirsch T. Chapter 5: Parabolic trough technology, In: Advanced CSP teaching materials. enerMENA; 2011. [21] Venegas R, Kuravi S, Kota K, McCay M. Comparative analysis of designing solar and geothermal power plants: a case study. Int J Renew Energy Res 2018;8:625–34. [22] Geothermal Technologies Office. GETEM—geothermal electricity technology evaluation model. US Department of Energy; 2012. [23] Wendt D, Mines G, Turchi C, Zhu G. Geothermal risk reduction via geothermal/solar hybrid power plants. Final report; 2015. https://doi. org/10.2172/1245529. [24] Kavadias KA, Alexopoulos P, Charis G, Kaldellis JK. Sizing of a solar–geothermal hybrid power plant in remote island electrical network. Energy Procedia 2019;157:901–8. https://doi.org/10.1016/ j.egypro.2018.11.256. [25] Kavadias KA, Alexopoulos P, Charis G. Techno-economic evaluation of geothermal-solar power plant in Nisyros island in Greece. Energy Procedia 2019;159:136–41. https://doi.org/10.1016/j. egypro.2018.12.031. [26] DeLovato N, Sundarnath K, Cvijovic L, Kota K, Kuravi S. A review of heat recovery applications for solar and geothermal power plants. Renew Sustain Energy Rev 2019;114:109329. https://doi.org/ 10.1016/j.rser.2019.109329. [27] Gurgenci H. Options to use solar heat to enhance geothermal power plant performance. In: Proceedings world geothermal congress 2015, Melbourne. Australia; 2015. p. 1–6.
152
PART II Thermodynamic analysis of geothermal power plants
[28] Bonyadi N, Johnson E, Baker D. Technoeconomic and exergy analysis of a solar geothermal hybrid electric power plant using a novel combined cycle. Energy Convers Manage 2018;156:542–54. https://doi.org/10.1016/j.enconman.2017.11.052. [29] Greenhut A, Tester JW, DiPippo R, Field R, Love C, Nichols K, et al. Solar-geothermal hybrid cycle analysis for low enthalpy solar and geothermal resources. In: Proceedings world geothermal congress 2010, Bali. Indonesia; 2010. p. 1–10. [30] Wendt D, Huang H, Sharan P, Kitz K, Green S, McLennan J, et al. Flexible geothermal power generation utilizing geologic thermal energy storage. Idaho National Laboratory (INL); 2019. [31] DiMarzio G, Angelini L, Price W, Chin C, Harris S. The stillwater triple hybrid power plant: integrating geothermal, solar photovoltaic and solar thermal power generation. In: Proceedings World Geothermal Congress 2015, Melbourne, Australia; 2015. p. 19–25. [32] Mctigue J, Castro J, Turchi C, Mungas G, Kramer N, King J, et al. Techno-economic assessment of geothermal power plants hybridized with solar heat and thermal storage. In: Proceedings 44th workshop on geothermal reservoir engineering, Stanford, California; 2019. p. 1–11. [33] DiPippo R. Geothermal power plants: evolution and performance assessments. Geothermics 2015;53:291–307. https://doi.org/ 10.1016/j.geothermics.2014.07.005. [34] Cakici DM, Erdogan A, Colpan CO. Thermodynamic performance assessment of an integrated geothermal powered supercritical regenerative organic Rankine cycle and parabolic trough solar collectors. Energy 2017;120:306–19. https://doi.org/10.1016/j. energy.2016.11.083. [35] Cardemil JM, Cortes F, Dı´az A, Escobar R. Thermodynamic evaluation of solar-geothermal hybrid power plants in northern Chile. Energy Convers Manage 2016;123:348–61. https://doi.org/10.1016/ j.enconman.2016.06.032.
[36] Radmehr B, Jalilinasrabady S. Modeling of the Single and double flash cycles and comparing them for power generation in sabalan geothermal field, Iran. In: Proceedings World Geothermal Congress 2015, Melbourne, Australia; 2015. p. 6. [37] DiPippo R. Geothermal power plants. In: Comprehensive renewable energy, vol. 7. Elsevier; 2012. p. 208–36. [38] DiPippo R. Geothermal power plants: principles, applications, case studies and environmental impact. 4th ed. Butterworth-Heinemann; 2015. [39] Ciani Bassetti M, Consoli D, Manente G, Lazzaretto A. Design and off-design models of a hybrid geothermal-solar power plant enhanced by a thermal storage. Renew Energy 2018;128:460–72. https://doi. org/10.1016/j.renene.2017.05.078. [40] McTigue JD, Wendt D, Kitz K, Gunderson J, Kincaid N, Zhu G. Assessing geothermal/solar hybridization—integrating a solar thermal topping cycle into a geothermal bottoming cycle with energy storage. Appl Therm Eng 2020;171:115121. https://doi.org/10.1016/j. applthermaleng.2020.115121. [41] Olabi AG, Mahmoud M, Soudan B, Wilberforce T, Ramadan M. Geothermal based hybrid energy systems, toward eco-friendly energy approaches. Renew Energy 2020;147:2003–12. https://doi.org/ 10.1016/j.renene.2019.09.140. [42] Alvarenga Y, Handal S, Recinos M. Solar steam booster in the Ahuachapan geothermal field. Geoth Resour Counc Trans 2008;32:395– 400. [43] Sander M. Combining renewable energy technologies with a geothermal focus. In: Proceedings of the 4th African Rift Geothermal Conference; 2012. p. 21–3. [44] Li K, Liu C, Jiang S, Chen Y. Review on hybrid geothermal and solar power systems. J Clean Prod 2020;250:119481. https://doi.org/ 10.1016/j.jclepro.2019.119481.
Chapter 10
Thermodynamic analysis of a transcritical CO2 geothermal power plant Anil Erdogana, Onder Kizilkanb, Mehmet Akif Ezana,c, and C. Ozgur Colpana,c a
The Graduate School of Natural and Applied Sciences, Dokuz Eylul University, Buca, Izmir, Turkey b Department of Mechanical Engineering, Faculty of
Technology, Isparta University of Applied Sciences, Isparta, Turkey c Faculty of Engineering, Department of Mechanical Engineering, Dokuz Eylul University, Buca, Izmir, Turkey
Nomenclature _ D Ex _ Q Ex _ W Ex _ Q W_ exf m_ h P RC s T
exergy destruction rate (MW) exergy transfer rate by heat (MW) exergy transfer by power (MW) heat transfer rate (MW) power (MW) specific flow exergy (kJ/kg) mass flow rate (kg/s) specific enthalpy (kJ/kg) pressure (kPa) Rankine cycle specific entropy (kJ/kg K) temperature (K)
Greek letters h
efficiency
Subscripts 0 acc air cr cv e en ex geo i ORC regen tCO2-RC TIP wf
10.1
dead state air-cooled condenser air critical control volume exit energy exergy geothermal inlet organic Rankine cycle regenerator transcritical CO2 Rankine cycle turbine inlet pressure cycle working fluid
Introduction
The growing energy demand due to the increasing population worldwide and the developments in technology, the
increasing concerns regarding air pollution and global warming, and the depletion of fossil fuels have led to the transition to renewable (e.g., geothermal, solar, hydro, ocean, wave, biomass, and wind) and alternative energy resources (e.g., nuclear) and technologies (e.g., parabolic trough solar collectors, photovoltaic panels, biomass gasifier, organic Rankine cycle (ORCs), fuel cells, and nuclear power plants). Geothermal energy, which is the thermal energy derived from the subsurface of the Earth, is one of the clean energy resources. Geothermal resources can be classified according to the production well temperature. Low-enthalpic resources are below 100°C, mediumenthalpic resources are between 100°C and 180°C, and high-enthalpic resources are above 180°C [1, 2]. Lowenthalpic resources can be directly used for district heating, whereas geothermal brine extracted from medium- and high-enthalpic resources can be used for power generation [2]. In the IEA’s World Energy Outlook 2019 report [3], it is stated that global electricity generation from geothermal energy was 90 TWh in 2018; according to different scenarios, this amount will be increased to 258 TWh (current policies), 316 TWh (stated policies), or 552 TWh (sustainable development) by 2040. According to the REN21’s Key Findings of the Renewables 2020 Global Status Report [4], the global geothermal power capacity was 13.9 GW in 2019. This report also shows that the top five countries with the highest geothermal power capacity at the end of 2019 are the United States, Indonesia, Philippines, Turkey, and New Zealand. For 2019, the most investment in geothermal power installation was from Turkey, which was followed by Indonesia, Kenya, Costa Rica, and Japan [4]. Geothermal power plants are mostly based on the organic Rankine cycle (ORC), which uses carbon-based working fluids to produce electricity by utilizing the heat of the hot geothermal brine [5, 6]. There are three basic power plant types for producing electricity from geothermal energy: the dry steam, the flash steam, and the binary cycle.
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00014-7 Copyright © 2021 Elsevier Inc. All rights reserved.
153
154
PART
II Thermodynamic analysis of geothermal power plants
In a dry steam power plant, steam from the geothermal reservoir is fed directly to the turbine. In a flash steam plant, high-pressure geothermal brine is extracted and converted to steam using a separator before sending it to the turbine. After that, the steam condenses to water and is sent back to the reinjection well. In a binary cycle power plant, geothermal heat is transferred to a secondary fluid (the working fluid). This heat provides the secondary fluid to be vaporized, which is then sent to the turbine. The International Renewable Energy Agency (IRENA) indicates that flash technologies, including double and triple flashes, dry steam, and the binary cycle cover 58%, 25%, and 17% of the global electricity production by geothermal energy, respectively [7, 8]. Some geothermal power plants use a combined cycle in which an ORC is coupled with a Brayton cycle or a Kalina cycle. On the other hand, to increase the efficiency of the system, a concentrated solar power (CSP) technology could be integrated (e.g., [9]). In the last decades, geothermal-based multigeneration systems have been proposed to produce not only electricity but also heat, fuel, and chemicals (e.g., hydrogen [10–12] and biogas [13]). The ORC is not only used in geothermal power plants, but also to produce electrical and thermal from other renewables (e.g., solar and biomass), waste heat from industrial processes, and other recoverable heat sources. The selection of the working fluid of an ORC is essential in the design of ORCs. In general, the working fluid should have the following properties: low evaporating pressure and temperature, low environmental impact (e.g., low ODP and GWP values), noncorrosive and nontoxic nature, high thermal conductivity, and low viscosity [9]. The following working fluid types are generally used in commercial ORCs: isobutane, pentane, n-butane, R134a, and R245fa [14, 15]. Recently, carbon dioxide (CO2) has been used as a working fluid in Rankine cycles (RCs) under the transcritical conditions. The reasons for using CO2 as the working fluid can be given as follows: it has superior thermophysical properties, and it is a nonflammable, nontoxic, harmless, and ecofriendly fluid with an ODP of 0 and a GWP of 1 [16]. In addition to these unique characteristics, CO2 has a high critical pressure (7.38 MPa) but a low critical temperature (31.18°C). Thanks to its properties, it is attractive for utilizing low- or medium-level heat sources for generating power. Furthermore, it is easily accessible and inexpensive. The properties of CO2 are given in Table 10.1. There are some studies on the thermodynamic assessment of geothermal-based tCO2-RCs in the literature. Zare and Takleh [18] proposed a geothermal-driven multigeneration system that included a transcritical CO2 and a subcritical RC. The system proposed by the authors is modified by changing the gas cooler with an internal heat exchanger in order to make the system more efficient. They performed a parametric analysis to investigate the effect of turbine inlet and outlet pressures, evaporator temperature,
TABLE 10.1 The properties of CO2 [17]. Properties
Values
Critical pressure
7.38 MPa
Critical temperature
31.18°C
Molar mass
44 kg/kmol
Volumetric refrigerant capacity
22,454 kJ/m3
ODP
0
GWP
1
ASHRAE safety group
A1
and heater outlet temperature on the performance of the system. The performance parameters were taken as the first and second law efficiencies and the net power developed by the system. The results demonstrated that when an internal heat exchanger was used instead of the gas cooler in the proposed system, the exergetic efficiency and the net power output were improved by 30.9% and 49.1%, respectively. Yu et al. [19] developed a mathematical model for a geothermal-based combined Kalina and transcritical CO2 cycles. A parametric analysis was performed to assess the effect of ammonia-water concentration, turbine inlet temperature and pressure, and condenser pressure on the performance of the combined cycle. The results indicated that the net power output and the energy and exergetic efficiencies of the combined system are 2808 kW, 28.28%, and 44.47%, respectively. Ahmadi et al. [8] analyzed a geothermal-based transcritical CO2 cycle to assess its performance. They conducted a parametric analysis to investigate the effect of turbine inlet temperature and pressure as well as the back pressure of the turbine on the system performance. The results showed that a lower turbine inlet pressure or a higher turbine inlet temperature caused higher energy and exergetic efficiencies. Shengjun et al. [20] conducted a performance comparison between the subcritical ORC and transcritical ORC in low-temperature level binary geothermal power systems. They also performed an optimization study to minimize the levelized energy cost and maximize the thermodynamic performance. Furthermore, they conducted a parametric analysis to investigate the effect of some operational parameters, which are the working fluid type, the evaporating temperature, and the condensing temperature on the exergetic efficiency, energy efficiency, and recovery efficiency. The results illustrated that R123 used in the subcritical ORC system yields the highest thermal and exergetic efficiencies of 11.1% and 54.1%, respectively, whereas R125 used in the transcritical cycle yields a thermal efficiency as 20% and an exergetic efficiency of 46.4%. Yang and Yeh [21] assessed the
Transcritical CO2 geothermal power plant Chapter 10
transcritical Rankine cycle integrated with a geothermal energy source from a thermodynamical point of view. A parametric analysis was performed to assess the effect on the system performance of turbine outlet pressure and the ratio of turbine inlet pressure on the condenser pressure. In addition, they conducted an economic optimization to maximize the net power output index and minimize the ratio of net power output to the total cost. The results indicated that the system with R125 gave the most satisfactory performance, followed by the system with R41 and CO2. The authors concluded that the transcritical Rankine cycle using CO2 as the working fluid has the highest temperature difference in the heat exchangers and thus is the potential system of the near future for producing electricity from low-temperature heat sources. Guo et al. [22] developed a model for a geothermal-driven CO2-based transcritical Rankine cycle and a subcritical Rankine cycle with R245fa as the working fluid A parametric analysis was conducted to assess the effect of geothermal resource temperature on the thermal efficiency and net power output. The results demonstrated that the thermal efficiency increases from 18% to 27% when the transcritical Rankine cycle is used instead of subcritical ORC under the same geothermal resource conditions. The study conducted by Guo et al. [23] presents an analysis of the geothermal-based transcritical CO2 Rankine cycle with the pinch point methodology. A parametric study was performed to assess the effect of the type of working fluid and the geothermal brine temperature on the performance of the system. The results showed that R125 yields the best performance among the other types of working fluids. It demonstrated a 10% higher net power output and 22% lower total heat transfer area under the same baseline conditions (CO2 as the working fluid and 90°C of geothermal source temperature). Li et al. [24] analyzed two cycles, the transcritical CO2 RC and the subcritical RC driven by the low-temperature geothermal source with the temperature between 90°C and 120°C. They assessed and compared two cycles considering the net power output, thermal efficiency, exergetic efficiency, and cost per net power output. The results indicated that when R600a is used as the working fluid, the highest net power output is obtained; however, R601 gives the highest thermal and exergetic efficiencies among the other working fluid types. In addition, it was found that the CO2 transcritical power cycle has a better economic performance than the ORC under the same turbine inlet pressure. Mokarram and Mosaffa [25] proposed the integration of a single geothermal system with a transcritical ORC working fluid with R245fa. To investigate the feasibility and thermodynamic performance, their proposed system was compared with a subcritical ORC with the integration of a single flash geothermal system. They performed a parametric analysis to investigate the effect of heat source temperature, separator pressure, and geofluid condenser pressure on the system
155
performance. In addition, they conducted a thermoeconomic analysis to find the total cost rate and levelized cost of energy for the considered systems. The results illustrated that when the separator and flash condenser pressures are kept constant, the heat source temperature plays a vital role in the performance of the system. On the other hand, the power output of the tCO2-RC cycle is 7.2% higher than the subcritical ORC. Baik et al. [26] compared a R125based transcritical cycle with subcritical ORCs using R134a, R245fa, and R152a. An optimization study was performed using a simulation method to maximize the power output with a fixed overall conductance of the heat exchangers. Their results illustrated that when R125a is used as the working fluid in the transcritical cycle, the power output of the cycle is greater than that of subcritical ORC. In this chapter, a comparative performance assessment between a geothermal-based tCO2-RC and an ORC using R134a as the working fluid is carried out. As a case study, the geothermal brine conditions taken from Aydın, Turkey, which are given in the paper by Cakici et al. [9], are used to compare the performance of an ORC operating with R134a with a hypothetical tCO2-RC. In addition, a parametric study is performed to assess the effect of the geothermal brine temperature entering the heat exchanger group, the turbine inlet pressure, and the pump inlet pressure on the performance of these systems. The performance parameters are taken as the net power output obtained by the system, the energy and exergy efficiencies, and the total exergy destruction, respectively. The following section provides a system description. The third section includes the mass, energy, and exergy balance equations for the system’s components and the equations for the performance assessment parameters of the system. The fourth section provides the results and discussions for the thermodynamic analyses of the tCO2-RC and the ORC operating with R134a. In addition, the results and discussion of the parametric analyses are given in this section. The final section includes the conclusions derived from the study.
10.2
System description
The layouts of the ORC operating with R134a as the working fluid and the tCO2-RC, which are compared in this study, are taken as the same and shown in Fig. 10.1. The working principle of either of these cycles is as follows. The geothermal brine extracted from the production wells enters the heat exchanger group (state 7), where the brine transfers its heat to the ORC working fluid. Then, the heated working fluid goes through the turbine (state 1), where the fluid is expanded and the power is produced. After leaving the turbine (state 2), the fluid goes through the regenerator; at this point, this increases the fluid temperature and thus
156
PART
II Thermodynamic analysis of geothermal power plants
air, out 3
TABLE 10.2 The input parameters of the system studied. Geothermal brine and ambient conditions [9]
Air Cooled Condenser
4 Pump air, in
5 Regenerator
Working Fluid Geothermal Brine 7
Production well
Turbine
2
1
Heat Exchanger Group
6 FIG. 10.1 The layout of the ORC system.
8
Re-injection well
increases the efficiency of the system. After that, the fluid enters the air-cooled condenser (ACC) (state 3), where the fluid leaves at the saturated liquid state (state 4). As it leaves the ACC, the fluid is sent to the pump and fed to the regenerator again (state 5). Finally, the fluid is sent to the heat exchanger group to complete the cycle (state 6). The geothermal brine is sent back to the reinjection well at the relevant temperature in order to provide the sustainability of the wells (state 8). The main difference of tCO2-RC with the ORC operating with R134a is that the pressure of the working fluid (CO2) in the heat exchanger group (P1 or P6) is greater than the critical pressure of that fluid in the tCO2-RC. The input parameters for the systems studied are given in Table 10.2 for the baseline conditions. The conditions for the geothermal brine are taken from Cakici et al. [9]. On the other hand, the operating conditions of the tCO2-RC are taken from Kizilkan [27]. The T-s diagrams of the ORC operating with R134a and the tCO2-RC for the baseline conditions are presented in Fig. 10.2A and B, respectively.
10.3
Mathematical modeling
In this section, the equations used in the thermodynamic analyses of the ORC operating with R134a and tCO2-RC are presented. The assumptions made in these analyses are given below. l
The systems operate under steady-state conditions.
l
l
l
l
The temperature of geothermal brine leaving the production well (T7)
160.6°C
The temperature of geothermal brine entering the reinjection well (T8)
69.5°C
The pressure of the geothermal brine leaving the production well (P7)
625.3 kPa
The mass flow rate of the geothermal brine
120.7 kg/s
Ambient temperature
20°C
Ambient pressure
101.325 kPa
Main operating parameters of the ORC
ORC operating with R134a [9]
tCO2-RC [27]
Working fluid
R134a
CO2
The mass flow rate of the working fluid
213.4 kg/s
1218 kg/s
Turbine inlet temperature (T1)
156.8°C
146°C
Turbine inlet pressure (P1)
2500 kPa (assumed)
9647 kPa
Turbine outlet pressure (P2)
947 kPa
6431 kPa
Isentropic turbine efficiency
0.80
0.80 (assumed)
Isentropic pump efficiency
0.80
0.80
The pressure drops across all the heat exchangers (the heat exchange group, the air-cooled condenser, and the regenerator) are ignored. The outer walls of all components are thermally insulated. The thermophysical properties of geothermal brine are assumed to be pure water. The kinetic and potential energies and exergies are neglected.
The steady-state mass (Eq. 10.1) and energy (Eq. 10.2) balance equations in rate forms are applied to the control volumes enclosing the cycle components to find the unknown thermodynamic properties of each state and the work and heat transfer interactions. X X (10.1) m_ m_ 0¼ i i e e X X (10.2) 0 ¼ Q_ cv W_ cv + m_ h m_ h i i i e e e
Transcritical CO2 geothermal power plant Chapter 10
157
FIG. 10.2 The temperature-specific entropy diagram for the (A) ORC operating with R134a and (B) tCO2-RC for the baseline conditions.
250 200 1
150
2
T [°C]
100
3 2500 kPa 6
50
947 kPa
5
4
0 −50 −100 −0.25
0.00
0.25
(A)
0.50
0.75
1.00
1.25
1.50
s [kJ/kg-K] 175 150
1
125 2
100
9647 kPa
T [°C]
75 50
3
6 5
25
4
6431 kPa
0 −25 −50 −75 −2.25
(B)
−2.00
−1.75
−1.50
−1.25
−1.00
−0.75
−0.50
s [kJ/kg-K]
Here, m_ i and m_ e are the mass flow rates of the working fluid at the inlet and outlet, respectively; Q_ cv is the heat transfer rate into the control volume (this term is neglected); W_ cv is the power output produced by the system; and hi and he are the specific enthalpies of the working fluid at the inlet and outlet, respectively. After determining all the state properties, the steadystate exergy balance equation (Eq. 10.3) in rate form is applied to the control volumes enclosing cycle components to find the exergy destructions in these components. X X _ W+ _ D (10.3) _ Q Ex m_ ex m_ ex Ex 0 ¼ Ex i i f ,i e e f ,e _ Q is defined as the rate of exergy transfer by heat where Ex _ W is the rate of exergy transfer (this term is neglected), Ex _ rate by work, Ex D is the rate of exergy destruction, and
exf, i and exf, e are the specific flow exergy at the inlet and outlet, respectively. This term can be found using Eq. (10.4). exf ¼ ðh h0 Þ T0 ðs s0 Þ
(10.4)
Here, h0 and s0 are the specific enthalpy and entropy at the dead-state conditions, respectively. The energy and exergy balance equations for each component are given in Table 10.3. The performance indicators, which are the energy and exergy efficiencies, can be found using Eqs. (10.5) and (10.6), respectively. en ¼
W_ net W_ net ¼ _ Q geo m_ geo ðh7 h8 Þ
(10.5)
TABLE 10.3 Balance equations for each component. Component name
Component schematic
Heat exchanger
1 7
Energy balance equation
Exergy balance equation
0 ¼ m_ wf ðh6 h1 Þ + m_ geo ðh7 h8 Þ
_ D, hex 0 ¼ m_ wf ex f , 6 ex f , 1 + m_ geo ex f , 7 ex f , 8 Ex
0 ¼ W_ turbine + m_ wf ðh1 h2 Þ
_ W , turbine + m_ wf ex f , 1 ex f , 2 Ex _ D, turbine 0 ¼ Ex
0 ¼ m_ wf ðh2 h3 Þ + m_ wf ðh5 h6 Þ
_ D, regen 0 ¼ m_ wf ex f , 2 ex f , 3 + m_ wf ex f , 5 ex f , 6 Ex
0 ¼ W_ pump + m_ wf ðh4 h5 Þ
_ W , pump + m_ wf ex f , 4 ex f , 5 Ex _ D, pump 0 ¼ Ex
0 ¼ m_ wf ðh3 h4 Þ + m_ air ðhair, i hair, e Þ
_ D, acc 0 ¼ m_ wf ex f , 3 ex f , 4 + m_ air ex f , air, in ex f , air, out Ex
8 Heat Exchanger Group
6 Turbine
2
Turbine
1 Regenerator
5 3
2 Regenerator
6 Pump
4 Pump 5
Air-cooled condenser
air, out 3 Air cooled condenser 4 air, in
Transcritical CO2 geothermal power plant Chapter 10
ex ¼
10.4
W_ net W_ net ¼ _ f ,8 _ geo Ex _ f ,7 Ex Ex
(10.6)
Results and discussion
In this section, the results on the parametric studies on the ORC operating with R134a and the tCO2-RC are presented. Parametric studies are performed to assess the effect of the geothermal source temperature extracted from the production well and the turbine inlet temperature on the energy and exergy efficiencies and the net power obtained by the systems. In addition, the exergy destruction rates of each component for the baseline conditions are compared. The thermodynamic properties are evaluated using the Engineering Equation Solver (EES). The thermodynamic properties and the specific flow exergies of each state are
159
given in Tables 10.4 and 10.5 for the ORC operating with R134a and the tCO2-RC, respectively. The results of the energy and exergy analyses for the ORC operating with R134a and the adapted tCO2-RC are given in Table 10.6. As seen from the table, tCO2-RC produces a higher power output compared to the ORC operating with R134a under the same geothermal heat input. Considering the power consumption of the pumps, the net power output is calculated as 2.566 MW for the ORC operating with R134a and 4.029 MW for the tCO2-RC. On the other hand, the results of the exergy analysis show that the total exergy destruction rates are found as 4.885 and 6.936 MW for the existing ORC and the tCO2-RC, respectively. Furthermore, tCO2-RC has higher energy and exergy efficiencies than the ORC operating with R134a. The energy efficiency is calculated as 6.64% for the actual ORC and 8.20% for the tCO2-RC. Although the exergy destruction rate is higher in the tCO2-RC, the exergy efficiency of this
TABLE 10.4 Thermodynamic properties of each state of the ORC operating with R134a. State number
Pressure (kPa)
Temperature (°C)
Specific enthalpy (kJ/kg)
Specific entropy (kJ/kg K)
Specific flow exergy (kJ/kg)
1
2500
156.8
383.6
1.158
91.67
2
947
123.7
360.3
1.173
64.03
3
947
104.8
340.5
1.122
59.45
4
947
37.5
104.3
0.3825
43.58
5
2500
38.8
106
0.3836
44.93
6
2500
52
125.9
0.446
46.21
7
625.3
160.6
677.4
1.947
101.7
8
625.3
69.5
290.9
0.9487
13.19
TABLE 10.5 Thermodynamic properties of each state of the tCO2-RC. State number
Pressure (kPa)
Temperature (°C)
Specific enthalpy (kJ/kg)
Specific entropy (kJ/kg K)
Specific flow exergy (kJ/kg)
1
9647
150.6
67.69
0.6261
251.1
2
6431
118
46.24
0.6123
225.6
3
6431
49.14
45.66
0.8723
209.4
4
6431
24.98
232.1
1.491
203
5
9647
31.92
226.9
1.488
207.5
6
9647
46.18
135
1.195
214
7
625.3
160.6
677.4
1.947
101.7
8
625.3
69.5
290.9
0.9487
13.19
160
PART
II Thermodynamic analysis of geothermal power plants
TABLE 10.6 Performance comparison of the ORC operating with R134a and tCO2-RC systems. ORC operating with R134a
tCO2RC
Power produced by the turbine (MW)
3.417
5.317
Net power output (MW)
2.566
4.029
Total exergy destruction rate (MW)
4.885
6.936
Energy efficiency (%)
6.64
8.20
Exergy efficiency (%)
26.86
35.03
FIG. 10.3 The exergy destruction ratios for each component for the baseline conditions in the (A) ORC operating with R134a and (B) tCO2-RC.
system is still higher than the ORC operating with R134a for the given conditions. The main reason for this result is that the net power output is considerably higher in the tCO2-RC. Fig. 10.3A and B show the exergy destruction ratios for each component for the baseline conditions. As can be seen from the figures, in the ORC operating with R134a, the highest and lowest exergy destructions occur in the aircooled condenser (3.101 MW with 68% ratio) and the pump (0.476 MW with 1% ratio), respectively. On the other hand, the highest and lowest exergy destructions occur in the heat exchanger group (2.009 MW with 49% ratio) and the pump (0.189 MW with 2% ratio), respectively, for the tCO2-RC. The reason for the highest exergy destruction in the aircooled condenser for ORC operating with R134a is that the condensation temperature of the working fluid for the ORC operating with R134a is higher than the tCO2-RC.
Transcritical CO2 geothermal power plant Chapter 10
161
FIG. 10.4 The effect of geothermal source temperature on the net power output and the energy efficiency for the ORC operating with R134a and tCO2-RC.
10.4.1 The effect of the geothermal source temperature on the performance of the systems Fig. 10.4 shows the effect of the geothermal source temperature on the net power output for the ORC operating with R134a and the tCO2-RC. As the source temperature increases, the net power output increases for both cycles. It can be seen that under the same source temperature (e.g., 140°C), the power output of the tCO2-RC is 44.54% higher than the ORC operating with R134a. It can be seen from the figure that as the source temperature increases, the net power difference between the two systems increases. The reason for this trend is the higher enthalpy in the supercritical state. On the other hand, when the source temperature varies between 117°C and 160°C, the energy efficiency increases by 6.08%–6.63% for the ORC operating with R134a and 6.68%–8.03% for tCO2-RC. It can be seen from the figure when the source temperature is taken as 140°C that the energy efficiency of tCO2-RC is 13.88% higher than that of the ORC operating with R134a. In addition, it can be seen that the change in the energy
efficiency of the ORC operating with R134a is less significant when the source temperature is between 140°C and 160°C. Thus, the tCO2-RC has better performance compared to the ORC operating with R134a. In Fig. 10.5, the total exergy destruction rate is presented as a function of the geothermal source temperature. As the source temperature increases, the total exergy destruction rate increases for both tCO2-RC and ORC operating with R134a. Under the same source temperature (e.g., 140°C), the results indicate that the total exergy destruction of the tCO2-RC is 18.5% higher than the ORC operating with R134a. In addition, Fig. 10.5 demonstrates the effect of the geothermal source temperature on the exergy efficiencies of the ORC operating with R134a and tCO2-RC. As the temperature increases, the exergy efficiency of ORC operating with R134a decreases. This trend is mainly due to the fact that as the temperature increases, the flow exergy change between the inlet and exit streams of the geothermal fluid increases. Thus, the exergy efficiency decreases. On the other hand, as the source temperature increases, the exergy efficiency of tCO2-RC changes little. In addition, it can be seen that the exergy efficiency of the FIG. 10.5 The effect of the geothermal source temperature on the total exergy destruction rate and the exergy efficiency for the ORC operating with R134a and the tCO2-RC.
162
PART
II Thermodynamic analysis of geothermal power plants
tCO2-RC is less sensitive to the source temperature compared to the ORC operating with R134a.
10.4.2 The effect of the turbine inlet pressure on the performance of the systems Another operating parameter that affects the performance of the ORC operating with R134a and tCO2-RC is the turbine inlet pressure. For this aim, in order to compare the systems studied, a nondimensional parameter, namely reduced pressure (Pr), is used. This parameter, as shown in Eq. (10.7), is defined as the ratio of the turbine inlet pressure (PTIP) to the critical pressure of the respective working fluid (Pcr). Pr ¼
PTIP Pcr
(10.7)
FIG. 10.6 The effect of turbine inlet pressure on the net power output and the efficiency of the systems for the ORC operating with R134a and the tCO2-RC.
FIG. 10.7 The effect of turbine inlet pressure on the total exergy destruction rates and the exergy efficiency for the ORC operating with R134a and t-CO2-RC.
The effect of the turbine inlet pressure on the net power output of the ORC operating with R134a and the tCO2RC is depicted in Fig. 10.6. As the pressure increases, the net power output for both systems increases. It can be seen from the figure that higher power is produced in tCO2-RC when the turbine inlet pressure increases. On the other hand, as the pressure increases, the energy efficiency increases. Fig. 10.7 gives the variation of the total exergy destruction rate with the turbine inlet pressure. Contrary to the net power production and energy and exergy efficiencies, the total exergy destruction rate decreases with the increase in the turbine inlet pressure. According to the results of the exergy analysis, the irreversibilities that occurred in the tCO2-RC are higher compared with the ORC operating with R134a. On the other hand, as the pressure increases, the exergy efficiencies increase. It can be seen from this figure that the tCO2-RC has higher performance than that of another system.
Transcritical CO2 geothermal power plant Chapter 10
163
10.4.3 The effect of the pump inlet pressure The reason for this trend is that the fluid is close to the supercritical state. In addition, it can be seen from the figure that on the performance of the systems Another parameter that can affect the performance of the systems is the inlet pressure of the pump. The pump inlet pressure in the tCO2-RC is higher than the ORC operating with R134a. To compare the systems effectively, a term called “reduced pump inlet pressure” is used. This term is designated as the ratio of the pump inlet pressure to the critical pressure of the respective working fluid. Fig. 10.8 shows the effect of reduced pump inlet pressure on the net power produced by the systems. As the reduced pump inlet pressure increases, the net power decreases sharply for the ORC operating with R134a. This trend is mainly due to the fact that when the reduced pump inlet pressure increases, the pump power consumption increases due to the enthalpy differences enclosing the pump. On the other hand, as the reduced pump inlet pressure increases, the net power developed slightly increases for the tCO2-RC.
if the reduced pump inlet pressure is 0.2 for the ORC operating with R134a and 0.9 for tCO2-RC, the net power outputs can be found as 2.983 and 4.08 MW, respectively. In Fig. 10.8, the influence of the reduced pump inlet pressure on the energy efficiencies of the systems with the ORC operating with R134a and the tCO2-RC are also presented. As the pressure increases, the energy efficiency decreases sharply for the ORC operating with R134a, whereas the energy efficiency increases slightly for the tCO2-RC. According to the results of the energy analysis, the reduced pump inlet pressure has a significant negative impact on the performance of the ORC operating with R134a, whereas it has a small positive impact on the performance of the tCO2-RC. The variation of the reduced pump inlet pressure on the total exergy destruction rate is plotted in Fig. 10.9. As the reduced pump inlet pressure increases, the total exergy FIG. 10.8 The effect of the reduced pump inlet pressure on the net power produced by the systems and the energy efficiencies for the ORC operating with R134a and the tCO2-RC.
FIG. 10.9 The effect of the reduced pump inlet pressure on the total exergy destruction rate and the exergy efficiencies for the ORC operating with R134a and tCO2-RC.
164
PART
II Thermodynamic analysis of geothermal power plants
destruction rate of the ORC operating with R134a increases due to the decreasing of the net power output, as discussed in Fig. 10.8. For the tCO2-RC, when the reduced pump inlet pressure is increased, the total exergy destruction decreases slowly. In addition, it can be seen from the figure that when the reduced pump inlet pressure increases from 0.18 to 0.49 for the ORC operating with R134a and 0.61 to 0.95 for the tCO2-RC, the total exergy destruction rate increases by 56.12% for the ORC operating with R134a whereas it decreases by 4.21% for the tCO2-RC. The change of the reduced pump inlet pressure on the exergy efficiency of the system is also depicted in Fig. 10.9. As seen in the figure, the reduced pump inlet pressure is varied between 0.18 and 0.5 for the ORC operating with R134a, and the exergy efficiency decreases from 35% to 3%. On the other hand, for the tCO2-RC, when the reduced pump inlet pressure is altered between 0.6 and 0.95, the exergy efficiency slightly increases from 30% to 33%.
10.5
Concluding remarks
In this chapter, the mathematical models for the geothermalbased ORC operating with R134a and tCO2-RC were developed, and their performances were compared. For this purpose, the thermodynamic modeling approach was used in which the energy and exergy balance equations were considered. In addition, parametric analyses were conducted to assess the effects of the geothermal source temperature, turbine inlet pressure, and pump inlet pressure on the performances of systems. Furthermore, the exergy destruction ratios for each component for the baseline conditions were presented. The major results of the study are given below: l
l
l
l
l
The tCO2-RC is a better option for producing more power compared to the ORC operating with R134a. The tCO2-RC is suitable for high-enthalpic geothermal resources. For the baseline conditions, the turbine power, the net power output, the total exergy destruction rate, and the energy and exergy efficiencies were found to be 3.417 MW, 2.566 MW, 4.885 MW, 6.64%, and 26.86%, respectively, for the ORC operating with R134a; they were 5.317 MW, 4.029 MW, 6.936 MW, 8.20%, and 35.03%, respectively, for the tCO2-RC. As the geothermal source temperature increases, the net power output increases for both systems. However, the exergy efficiency decreases with the increase in the temperature for the ORC operating with R134a, whereas the exergy efficiency changes less significantly for the tCO2-RC. In addition, the total exergy destruction increases for both systems studied. When the turbine inlet pressure increases, the net power output as well as the energy and exergy efficiencies
l
increase for both systems. However, the total exergy destruction rate decreases for both systems. When the pump inlet pressure increases, the net power output as well as the energy and exergy efficiencies decrease for the ORC operating with R134a. However, the total exergy destruction increases in this system. On the other hand, the net power output as well as the energy and exergy efficiencies increase slowly, and the total exergy destruction decreases for the tCO2-RC.
It can be concluded that the higher net power outputs can be realized when tCO2-RC is used instead of an ORC operating with R134a. New geothermal plants can be designed to be operated with transcritical cycles to obtain high performance.
References [1] Gitonga G. Geothermal power plants for medium and high temperature steam and an overview of wellhead power plants. In: SDG Short Course II Exploration and Development of Geothermal Resources, Kenya; 2017. p. 1–7. [2] Sigfusson B, Uihlein A. 2014 JRC geothermal energy status report; 2015. https://doi.org/10.2790/460251. [3] IEA. World energy outlook 2019. Paris; 2019. p. 2019. [4] REN21. Renewables 2020 global status report; 2020. [5] Macchi E, Astolfi M. Organic Rankine cycle (ORC) power systems technologies and applications. Elsevier; 2017. [6] Anon. Geothermal explained: geothermal power plants, https://www. eia.gov/energyexplained/geothermal/geothermal-power-plants.php; 2019. [Accessed 22 July 2020]. [7] IRENA. Technology brief: geothermal power; 2017. p. 111–3. [8] Ahmadi MH, Mehrpooya M, Pourfayaz F. Exergoeconomic analysis and multi objective optimization of performance of a carbon dioxide power cycle driven by geothermal energy with liquefied natural gas as its heat sink. Energy Convers Manage 2016;119:422–34. https://doi. org/10.1016/j.enconman.2016.04.062. [9] Cakici DM, Erdogan A, Colpan CO. Thermodynamic performance assessment of an integrated geothermal powered supercritical regenerative organic Rankine cycle and parabolic trough solar collectors. Energy 2017;120:306–19. https://doi.org/10.1016/j.energy. 2016.11.083. [10] Yilmaz C, Kanoglu M, Abusoglu A. Exergetic cost evaluation of hydrogen production powered by combined flash-binary geothermal power plant. Int J Hydrogen Energy 2015;40:14021–30. https://doi. org/10.1016/j.ijhydene.2015.07.031. [11] Gholamian E, Habibollahzade A, Zare V. Development and multiobjective optimization of geothermal-based organic Rankine cycle integrated with thermoelectric generator and proton exchange membrane electrolyzer for power and hydrogen production. Energy Convers Manage 2018;174:112–25. https://doi.org/10.1016/j. enconman.2018.08.027. [12] Karapekmez A, Dincer I. Modelling of hydrogen production from hydrogen sulfide in geothermal power plants. Int J Hydrogen Energy 2018;43:10569–79. https://doi.org/10.1016/j.ijhydene.2018.02.020. [13] Mohan AR, Turaga U, Subbaraman V, Shembekar V, Elsworth D, Pisupati SV. Modeling the CO2-based enhanced geothermal system
Transcritical CO2 geothermal power plant Chapter 10
[14]
[15] [16]
[17]
[18]
[19]
[20]
(EGS) paired with integrated gasification combined cycle (IGCC) for symbiotic integration of carbon dioxide sequestration with geothermal heat utilization. Int J Greenh Gas Control 2015;32:197– 212. https://doi.org/10.1016/j.ijggc.2014.10.016. Calnetix ORC, https://www.calnetix.com/sites/default/files/CalnetixAE ORC 125MT Brochure_web_spread_pages.pdf; 2019. [Accessed 25 July 2020]. Ormat, https://www.ormat.com/en/renewables/geothermal/main/; 2020. [Accessed 25 July 2020]. Kizilkan O. Performance assessment of steam Rankine cycle and sCO2 Brayton cycle for waste heat recovery in a cement plant: a comparative study for supercritical fluids. Int J Energy Res 2020. https:// doi.org/10.1002/er.5138. er.5138. Linde Industrial Gas n.d. https://www.linde-gas.com/en/products_ and_supply/refrigerants/natural_refrigerants/r744_carbon_dioxide/ index.html (Accessed 19 July 2020). Zare V, Takleh HR. Novel geothermal driven CCHP systems integrating ejector transcritical CO2 and Rankine cycles : thermodynamic modeling and parametric study. Energy Convers Manage 2020;205:112396. https://doi.org/10.1016/j.enconman.2019.112396. Yu Z, Su R, Feng C. Thermodynamic analysis and multi-objective optimization of a novel power generation system driven by geothermal energy. Energy 2020. https://doi.org/10.1016/j. energy.2020.117381. Shengjun Z, Huaixin W, Tao G. Performance comparison and parametric optimization of subcritical organic Rankine cycle ( ORC ) and transcritical power cycle system for low-temperature geothermal power generation. Appl Energy 2011;88:2740–54. https://doi.org/ 10.1016/j.apenergy.2011.02.034.
165
[21] Yang M-H, Yeh R-H. Economic performances optimization of the transcritical Rankine cycle systems in geothermal application. Energy Convers Manage 2015;95:20–31. https://doi.org/10.1016/j. enconman.2015.02.021. [22] Guo T, Wang H, Zhang S. Comparative analysis of CO2-based transcritical Rankine cycle and HFC245fa-based subcritical organic Rankine cycle using low-temperature geothermal source. Sci China Technol Sci 2010;53:1638–46. https://doi.org/10.1007/s11431-0103123-4. [23] Guo T, Wang H, Zhang S. Comparative analysis of natural and conventional working fluids for use in transcritical Rankine cycle using low-temperature geothermal source. Int J Energy Res 2011;35:530– 44. https://doi.org/10.1002/er.1710. [24] Li M, Wang J, Li S, Wang X, He W, Dai Y. Thermo-economic analysis and comparison of a CO2 transcritical power cycle and an organic Rankine cycle. Geothermics 2014;50:101–11. https://doi. org/10.1016/j.geothermics.2013.09.005. [25] Mokarram NH, Mosaffa AH. Investigation of the thermoeconomic improvement of integrating enhanced geothermal single flash with transcritical organic Rankine cycle. Energy Convers Manage 2020;213. https://doi.org/10.1016/j.enconman.2020.112831. [26] Baik Y, Kim M, Chang K, Lee Y, Yoon H. A comparative study of power optimization in low-temperature geothermal heat source driven R125 transcritical cycle and HFC organic Rankine cycles. Renew Energy 2013;54:78–84. https://doi.org/10.1016/j.renene.2012.08.055. [27] Kizilkan O. Evaluation of transcritical Rankine cycle driven by lowtemperature geothermal source for different supercritical working fluids. Int J Technol Sci 2019;11:155–69.
Chapter 11
Double-flash enhanced Kalina-based binary geothermal power plants Hadi Rostamzadeha, Milad Feilib, and Hadi Ghaebib a
Energy and Environment Research Center, Niroo Research Institute (NRI), Tehran, Iran b Department of Mechanical Engineering, Faculty of
Engineering, University of Mohaghegh Ardabili, Ardabil, Iran
Nomenclature Symbols _ Ex ex h M m_ p Q_ T XB
exergy rate (kW) exergy per mass (kJ kg1) specific enthalpy (kJ kg1) molar mass (kg kmol1) mass flow rate (kg s1) pressure (bar) heat transfer rate (kW) temperature (K) basic ammonia concentration
Greek letters « h
effectiveness efficiency (%)
Subscripts and superscripts C ch D EV ex F FT geo GPP IHX in is L M m mtd net out P Ph Pr RIP
condenser chemical destruction expansion valve exergy fuel flash tank geothermal geothermal power plant internal heat exchanger inlet isentropic loss mixer medium minimum temperature difference net value outlet pump physical product reinjection pump
separator simple KC self-superheating heater steam turbine turbine total two-stage KC vapor generator
S SKC SSH ST T tot TSKC VG
Abbreviations B-GPP BKC/DFB-GPP DF-GPP DFB-GPP DKC/DFB-GPP E-GPP ETKC/DFB-GPP EV FT GPP IHX KC RIP SF-GPP SKC SSH TSKC
11.1
binary GPP basic KC/DFB-GPP double-flash GPP double-flash binary GPP double KC/DFB-GPP enhanced GPP enhanced triple KC/DFB-GPP expansion valve flash tank geothermal power plant internal heat exchanger Kalina cycle reinjection pump single-flash GPP simple KC self-superheating heater two-stage KC
Introduction
Despite the high investment cost of geothermal power plants (GPPs), recent advances have led to a low operating cost for such power plants by increasing the conversion efficiency of the GPPs via appropriate and highly efficient process integration schemes [1]. The low conversion efficiency of the existing GPPs mainly stems from the fact that many of them are working on low-temperature wells in comparison with the combustion temperature of fossil fuels. Hence, thermal performance enhancement of GPPs via highly efficient process integration methods are very necessary, especially for low-temperature wells.
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00009-3 Copyright © 2021 Elsevier Inc. All rights reserved.
167
168
PART II Thermodynamic analysis of geothermal power plants
Currently, four types of GPPs are conventional worldwide: single-flash GPP (SF-GPP), double-flash GPP (DF-GPP), binary GPP (B-GPP), and enhanced GPP (E-GPP). Of course, there are many newly developed configurations over the recent years that are a mix of these four conventional types. Due to an urgent need for high-capacity power plants with population growth, double-flash binary GPPs (DFB-GPPs) are highly recommended mainly due to their high performance [2], low unit cost of operation, and low environmental impact [3], to mention but a few. Of course, there are several dilemmas with regard to the installment of DFB-GPPs, such as the high content of NCGs (noncondensable gases), which adversely affects the sustainability of the set-up [4]. However, viable solutions in this regard are also proposed by scholars that ease the environmental burdens on the environmental activists. Over the past years, the applications of DF-GPPs have been extended for power augmentation [5], hydrogen coproduction [6], freshwater coproduction [2, 7], cooling/heating cogeneration [2, 8, 9], etc. All investigated studies have unanimously substantiated the promising role of process integration in DF-GPPs for thermal enhancement, lowering operation unit cost, reducing the entropy generation rate, decreasing environmental impacts, etc. Because the focus of this chapter is on the power enhancement of a conventional DFB-GPP, the literature review is directed toward recent studies in integrating appropriate power generation systems with the DFB-GPP. Abdolalipouradl et al. [10] used a Kalina cycle (KC) and a transcritical Rankine cycle (TRC) as binary cycles of a DF-GPP and presented an exergoeconomic analysis and multiobjective optimization of the devised configuration for the Sabalan geothermal power plant in Iran. They found that the use of TRC for the hightemperature part of the geofluid contributed nearly 20% to the overall net power of the system, which can be increased by 4.68% when a KC is added to the low-temperature part of the geofluid. Mokarram and Mosaffa [11] integrated a KC as a binary cycle with a single-flash B-GPP (SFB-GPP) and a DFB-GPP in four scenarios. They used a part of the geofluid heat from the wellhead to increase the energy of the steam at the inlet of each steam turbine of the GPP system. They found a 6% improvement in the optimum power generation of the basic system. Aali et al. [12] proposed a new DFBGPP for the Sabalan geothermal well with an ORC (organic Rankine cycle) as the binary cycle. They found an optimum exergy efficiency of 54.87% when R141b is used in the ORC. More recently, Feili et al. [13] used a novel zeotropic refrigeration cycle as binary cycle of the enhanced DFBGPP proposed by Mokarram and Mosaffa [11] and achieved energy and exergy efficiencies of 20.37% and 20.95%, respectively. In light of the reviewed literature, it is evident that using KC as a binary cycle of the DFB-GPP can be an outstanding selection due to the low-temperature characteristics of KC
as well as its zeotropic mixture feature, which has made KC a safer and more reliable power system at low temperatures in comparison with ORC. Although several previous studies have proposed the use of KC in the DFB-GPPs such as the study carried out by Mokarram and Mosaffa [11], the present study demonstrates that more highly efficient KCbased DFB-GPPs can be proposed in more innovative ways. Mokarram and Mosaffa [11] presented a basic and an enhanced KC-based DFB-GPP, which only used simple KC in both simulated set-ups with the difference of preheating the inlet steam of each steam turbine via the high heat of the liquid discharged from each flash tank. However, it is believed that the high thermal energy of the highpressure geofluid before throttling the first flash tank and the thermal heat of the liquid discharging from the second flash tank of a DFB-GPP can be used independently to drive a KC. This integrated plant is called a double KC-based DFB-GPP (DKC/DFB-GPP) because two KCs are used in a more efficient way. In addition, a modified two-stage KC proposed by Bahrampoury and Behbahaninia [14] is used to capture the thermal energy of the discharging liquid from the first flash tank in the next phase of the current investigation. This integrated plant is called a triple KCbased DFB-GPP (TKC/DFB-GPP) because three KCs are used in a more efficient way. The superiority of the devised two KC-based DFB-GPPs over the best set-up proposed by Mokarram and Mosaffa [11] is proven in this study using two viewpoints, energy and exergy analysis. The rest of this chapter is arranged in the following order. In Section 11.2, a brief description of the layouts is presented. In Section 11.3, all employed mathematical relations and presumptions are displayed. In the Section 11.4, the results are presented and discussed extensively. Finally, some concluding comments are listed in the last part.
11.2
Description of the plants
Before the presentation of the new proposed KC-based DFB-GPPs, the thermodynamic cycles of the KC and twostage KC (TSKC) are separately illustrated in Fig. 11.1. According to Fig. 11.1A, the NH3-H2O blend at the saturated pressure of the vapor generator is split into lean and rich mixtures, where the rich mixture is rotated through the turbine to generate electricity while the lean mixture loses its temperature prior to a throttling process via an IHX (internal heat exchanger) and then is merged with the expanded rich mixture. The merged stream is next liquefied at the condensation temperature and then is pressurized to the saturated pressure of the vapor generator via a pump, ending the KC flow process. Fig. 11.1B illustrates a sketch of the selected TSKC proposed by Bahrampoury and Behbahaninia [14] (in the study of these authors, the superiority of this configuration to other similar ones has been demonstrated). In this system, the basic NH3 solution has been
Kalina-based binary geothermal power plants Chapter 11
169
FIG. 11.1 Layout of (A) a simple KC (SKC) and (B) a two-stage KC (TSKC) [14] used in the proposed DFB-GPPs.
enriched via two separators at two different processes, where the enriched solution is expanded separately via two turbines. A more involved explanation of the selected configuration can be found in Ref. [14]. Fig. 11.2 displays the basic KC-based DFB-GPP proposed by Mokarram and Mosaffa [11] along with two newly developed KC-based DFB-GPPs: a double KC-based DFB-GPP (DKC/DFB-GPP) illustrated in Fig. 11.2B and an enhanced triple KC-based DFB-GPP (ETKC/DFBGPP) illustrated in Fig. 11.2C. In Fig. 11.2A, two selfsuperheating heaters (SSHs) are used to preheat the inlet steam of the steam turbines by the thermal heat of the liquid at the outlet of each flash tank. Further explanation of this
system can be found in Ref. [11]. Because this system is the most highly efficient DF-GPP with KC as the binary cycle available in the literature, this configuration is selected as the basic system, and the modification results are compared with this scenario. The thermal heat of the geofluid at the liquid state before throttling to flash tank-1 pressure can be used as a prime mover of a KC instead of being used for preheating. It is noteworthy that the idea of using an extra KC at the outlet of the liquid fraction of the first flash tank is also investigated in this study, where the results were not appealing to the present configuration, so they are excluded here. The configuration associated with this idea is displayed in Fig. 11.2B. In the third phase of this study, the performance
170
PART II Thermodynamic analysis of geothermal power plants
FIG. 11.2 Layout of the: (A) basic KC-based DFB-GPP (BKC/DFB-GPP) proposed by Mokarram and Mosaffa [11], (B) new double KC-based DFBGPP (DKC/DFB-GPP), and (C) enhanced triple-KC-based DFB-GPP (ETKC/DFB-GPP).
of the DKC/DFB-GPP is further improved by employing a modified two-stage KC at the outlet of the first flash tank in a more novel way in which the low-temperature thermal energy of this modified KC is supplied by the geofluid exiting from the vapor generator in the third KC (Fig. 11.2C).
11.3
Materials and methods
In this section, the simulation procedure, presumptions, and relations are presented. In the first subsection, the primary thermodynamic presumptions needed through the analysis are pinpointed. In the second subsection, all relations required for the simulation of the devised power plants in
terms of energy and exergy are presented. Ultimately, a comprehensive parametric study is outlined for three systems under a constant external condition.
11.3.1 Thermodynamic presumptions and evaluation The subsequent presumptions are assumed in the present analysis: l l
The condition through the simulation is the steady state. Pumps and turbines operate with an isentropic efficiency.
Kalina-based binary geothermal power plants Chapter 11
l l
Expansion devices work isenthalpically. The outlet streams of the vapor generators and condensers are assumed saturated.
The input data used in the analyses are listed in Table 11.1. The general form of the governing equations at the steady state for the thermodynamic evaluation of a unit can be articulated as: X X m_ out ¼ 0 (11.1) m_ in X X _ Þout _ Þin ðmh ðmh (11.2) Q_ c:v: W_ c:v: ¼
171
In terms of the Second Law of Thermodynamics, the balance relation of a unit may be articulated as: _ D,k ¼ Ex
k X
_ in,i Ex
i¼1
k X
_ out,i Ex
(11.3)
i¼1
or, in terms of the exergy of the fuel, product, loss, and destruction, Eq. (11.3) can be reexpressed as follows: _ Fu,k ¼ Ex _ _ _ Ex Pr,k + ExL,k + ExD,k
(11.4)
The overall exergy of the fluid stream is declared as:
TABLE 11.1 Required thermodynamic input parameters for the simulation. Parameter
Value
References
Reference temperature, T0 (°C)
25
[11]
Reference pressure, P0 (kPa)
101.325
[11]
Steam turbine isentropic efficiency, is,ST (%)
85
[11]
KC turbine isentropic efficiency, is,KC,T (%)
75
[11]
Pumps isentropic efficiency, is,P (%)
85
[11]
The temperature of geothermal water at the wellhead, Tin,geo (°C)
200
[11]
The temperature of cooling water at the condenser inlet, Tin,C (°C)
25
[11]
The temperature of water at the condenser outlet, Tout,C (°C)
35
[11]
The temperature of the mixture at the KC condenser outlet, TKC,C(°C)
30
[11]
The minimum temperature difference between the geofluid and the AWM stream, DTmtd (K)
5
[11]
The temperature of the geofluid injected to the wellhead, Tout,geo (°C)
70
[11]
The pressure of the first flash tank, PFT1 (kPa)
1000
[11]
The pressure of the second flash tank, PFT2 (kPa)
780
–
The pressure of the geothermal power plant water at the outlet of the condenser, PGPP,C (kPa)
10
[11]
The pressure of the geothermal water at the wellhead, Pin,geo (kPa)
1554
[11]
The upper pressure of the SKC-2, PSKC2(kPa)
2000
[11]
The upper pressure of the SKC-1, PSKC1 (kPa)
3000
–
The medium pressure of the TSKC, Pm,TSKC (kPa)
1500
[14]
The upper pressure of the TSKC, PTSKC (kPa)
2500
[14]
Ammonia concentration of the SKC-2, XSKC2
0.65
[11]
Ammonia concentration of the SKC-1, XSKC1
0.36
–
0.36
–
1
[11]
0.3
[14]
0.5
[14]
Effectiveness of IHX, EIHX
0.80
[11]
Effectiveness of SSH, ESSH
080
[15]
Ammonia concentration of the main stream of the TSKC, XTSKC _ in, geo The mass flow rate of geothermal water at the wellhead, m _ 27 kg s1 Main stream mass flow rate of TSKC, m _ r¼ The mass ratio of TSKC, m
_ 28 m _ 27 m
kg s1
172
PART II Thermodynamic analysis of geothermal power plants
_ ph,k _ k ¼ Ex Ex
(11.5)
where, _ ph,k ¼ m_ ðh h0 T0 ðs s0 ÞÞ Ex k
(11.6)
11.3.2 Main performance assessment parameters The energy efficiency of the simulated KC-based DFBGPPs is expressed as below: en ¼
The exergetic efficiency of the kth element is declared as: ex,k ¼
_ out Ex _ Pr,k Ex ¼ _ in Ex _ Fu, k Ex
(11.7)
Some mass-, energy-, and exergy-based balance relations for each constituent of the assumed set-ups are enumerated in Table 11.2.
W_ net m_ 1 ðh1 h0 Þ
(11.8)
where W_ net for the basic KC-, double KC-, and enhanced triple KC-based DFB-GPPs can be expressed respectively as: W_ net,BKC=DFBGPP ¼ W_ ST1 + W_ ST2 + W_ SKC,T W_ SKC,P W_ RIP1 W_ RIP2 (11.9)
TABLE 11.2 Energy and exergy balance equations obtained from the exergy and energy evaluations of the devised systems. Component
Energy balance equation
Exergy balance equation
ST1
W_ ST1 ¼ m_ 4 ðh4 h5 Þ, is,ST ¼ (h4 h5)/(h4 h5s)
_ _ _ E_ x ST1 D ¼ Ex 4 E x 5 W ST1
ST2
W_ ST2 ¼ m_ 12 ðh12 h13 Þ, is,ST ¼ (h12 h13)/(h12 h13s)
_ _ _ E_ x ST2 D ¼ Ex 12 E x 13 W ST2
Condenser1
Q_ GPP, C1 ¼ m_ 5 ðh5 h6 Þ, Q_ GPP, C1 ¼ m_ 20 ðh20 h19 Þ
, C1 ¼ Ex 5 E_ x 6 _E_ x 20 E_ x 19 E_ x GPP D
Condenser2
Q_ GPP, C2 ¼ m_ 13 ðh13 h14 Þ, Q_ GPP, C2 ¼ m_ 18 ðh18 h17 Þ
, C2 ¼ Ex 13 E_ x 14 _ E_ x 18 E_ x 17 E_ x GPP D
RIP1
W_ RIP1 ¼ m_ 7 ðh7 h6 Þ, is,P ¼ (h7s h6)/(h7 h6)
RIP2
W_ RIP2 ¼ m_ 15 ðh15 h14 Þ, is,P ¼ (h15s h14)/(h15 h14)
EV1
h1 ¼ h2
, EV1 ¼ E_ x 1 E_ x 2 E_ x GPP D
EV2
h3 ¼ h8
FT1
m_ 3 h3 + m_ 4 h4 ¼ m_ 2 h2 , m_ 3 + m_ 4 ¼ m_ 2
, EV2 ¼ E_ x 3 E_ x 8 E_ x GPP D E_ x FT1 ¼ E_ x 2 E_ x 3 + E_ x 4
FT2 SSH1
m_ 12 h12 + m_ 9 h9 ¼ m_ 8 h8 , m_ 12 + m_ 9 ¼ m_ 8 m_ 1 h1 h1 ¼ m_ 4 h4 h4
SSH2
m_ 3 h3 h3 ¼ m_ 12 h12 h12
_ ¼ Ex 3 E_ x 3 E_ x12 E_ x 12 E_ x SSH2 D
Turbine
W_ SKC, T ¼ m_ 23 ðh23 h24 Þ, is,KC,T ¼ (h23 h24)/(h23 h24s)
,T ¼ Ex 23 E_ x 24_ W_ KC, tur E_ x SKC D
Condenser
Q_ SKC, C ¼ m_ 27 ðh27 h28 Þ Q_ SKC, C ¼ m_ 32 ðh32 h31 Þ
,C ¼ Ex 27 E_ x 28 _ E_ x 32 E_ x 31 E_ x SKC D
GPP
E_ x RIP1 ¼ W_ RIP1 E_ x 7 E_ x 6 D E_ x RIP2 ¼ W_ RIP2 E_ x 15 E_ x 14 D
D
_ _ _ E_ x FT2 D ¼ E x 8 E x 9 + E x 12 _ ¼ Ex 1 E_ x 1 E_ x 4 E_ x 4 E_ x SSH1 D
SKC
Kalina-based binary geothermal power plants Chapter 11
173
TABLE 11.2 Energy and exergy balance equations obtained from the exergy and energy evaluations of the devised systems— cont’d Component
Energy balance equation
Pump
W_ SKC, P ¼ m_ 29 ðh29 h28 Þ, is,P ¼ (h29s h28)/(h29 h28)
EV
h25 ¼ h26
Separator
m_ 22 h22 + m_ 23 h23 ¼ m_ 21 h21 , m_ 22 + m_ 23 ¼ m_ 21
Mixer
m_ 24 h24 + m_ 26 h26 ¼ m_ 27 h27 , m_ 24 + m_ 26 ¼ m_ 27
IHX
Q_ SKC, IHX ¼ m_ 22 ðh22 h25 Þ, Q_ SKC, IHX ¼ m_ 30 ðh30 h29 Þ
Exergy balance equation ,P ¼ W_ SKC, P E_ x 29 E_ x 28 E_ x SKC D , EV ¼ E_ x 25 E_ x 26 E_ x SKC D ,S ¼ E_ x 21 E_ x 23 + E_ x 24 E_ x SKC D
,M ¼ E_ x 24 + E_ x 26 E_ x 27 E_ x SKC D , IHX E_ x SKC ¼ Ex 22 E_ x 25 _ E_ x 30 E_ x 29 D
EIHX ¼(T22 T25)/(T22 T29) VG
Q_ SKC, VG ¼ m_ 21 ðh21 h30 Þ
, VG ¼ from geothermal E_ x 21 E_ x 30 E_ x SKC D
TSKC Turbine1
W_ TSKC, T1 ¼ m_ 23 ðh23 h24 Þ, KC,is,T ¼(h23 h24)/(h23 h24s)
, T1 ¼ Ex 23 E_ x 24 _ W_ TSKC, T1 E_ x TSKC D
Turbine2
W_ TCKC, T2 ¼ m_ 34 ðh34 h35 Þ, KC,is,T ¼ (h34 h35)/(h34 h35s)
, T2 ¼ Ex 34 E_ x 35 _ W_ TSKC, T2 E_ x TSKC D
Condenser
Q_ TSKC, C ¼ m_ 42 ðh42 h25 Þ, Q_ TSKC, C ¼ m_ 44 ðh44 h43 Þ
,C E_ x TSKC ¼ Ex 42 E_ x 25 _ E_ x 44 E_ x 43 D
Pump1
W_ TSKC, P1 ¼ m_ 26 ðh26 h25 Þ, is,P ¼ (h26s h25)/(h26 h25)
Pump2
W_ TSKC, P2 ¼ m_ 29 ðh29 h28 Þ, is,P ¼ (h29s h28)/(h29 h28)
EV1
h36 ¼ h37
, EV1 ¼ E_ x 36 E_ x 37 E_ x TSKC D
EV2
h40 ¼ h41
Separator1
m_ 22 h22 + m_ 23 h23 ¼ m_ 21 h21 , m_ 22 + m_ 23 ¼ m_ 21
, EV2 ¼ E_ x 40 E_ x 41 E_ x TSKC D , S1 ¼ E_ x 21 E_ x 23 + E_ x 24 E_ x TSKC D
Separator2
m_ 33 h33 + m_ 38 h38 ¼ m_ 32 h32 , m_ 33 + m_ 38 ¼ m_ 32 m_ 33 X33 + m_ 38 X38 ¼ m_ 32 X32
Mixer1
m_ 37 h37 + m_ 38 h38 ¼ m_ 39 h39 , m_ 37 + m_ 38 ¼ m_ 39 m_ 37 X37 + m_ 38 X38 ¼ m_ 39 X39
Mixer2
m_ 35 h35 + m_ 41 h41 ¼ m_ 42 h42 , m_ 35 + m_ 41 ¼ m_ 42
IHX1
Q_ TSKC, IHX1 ¼ m_ 22 ðh22 h36 Þ Q_ TSKC, IHX1 ¼ m_ 32 ðh32 h31 Þ
, IHX1 E_ x TSKC ¼ Ex 22 E_ x 36 _ E_ x 32 E_ x 31 D
Q_ TSKC, IHX2 ¼ m_ 39 ðh39 h40 Þ, Q_ TSKC, IHX2 ¼ m_ 27 ðh27 h26 Þ
, IHX2 E_ x TSKC ¼ Ex 39 E_ x 40 _ E_ x 27 E_ x 26 D
, P1 ¼ W_ TSKC, P1 E_ x 26 E_ x 25 E_ x TSKC D , P2 ¼ W_ TSKC, P2 E_ x 29 E_ x 28 E_ x TSKC D
, S2 ¼ E_ x 32 E_ x 33 + E_ x 37 E_ x TSKC D , M1 ¼ E_ x 37 + E_ x 38 E_ x 39 E_ x TSKC D , M2 ¼ E_ x 35 + E_ x 41 E_ x 42 E_ x TSKC D
EIHX1 ¼ (T22 T36)/(T22 T31)
IHX2
EIHX2 ¼ (T39 T40)/(T39 T26)
VG1
Q_ TSKC, VG1 ¼ m_ 21 ðh21 h29 Þ
VG2
Q_ TSKC, VG2 ¼ m_ 31 ðh31 h30 Þ,
, VG1 ¼ from geothermal E_ x 21 E_ x 29 E_ x TSKC D , VG2 ¼ from geothermal E_ x 31 E_ x 30 E_ x TSKC D
174
PART II Thermodynamic analysis of geothermal power plants
W_ net,DKC=DFBGPP ¼ W_ ST1 + W_ ST2 + W_ SKC1, T + W_ SKC2, T W_ SKC1,P W_ SKC2,P W_ RIP1 W_ RIP2 (11.10) W_ net,ETKC=DFBGPP ¼ W_ ST1 + W_ ST2 + W_ SKC1,T + W_ SKC2,T + W_ TSKC,T1 + W_ TSKC,T2 W_ SKC1,P W_ SKC2,P W_ TSKC, P1 W_ TSKCP2 W_ RIP1 W_ RIP2 (11.11) The exergy efficiency of the devised KC-based DFB-GPPs is articulated as below: ex ¼
11.4
_ Pr,tot Ex _ Fu,tot Ex
(11.12)
Results and discussion
In this part, the results of the simulation are displayed in the form of figures and tables for various central parameters. Table 11.3 lists the main performance parameters for the three simulated DFB-GPPs under the constant input conditions given in Table 11.1. According to Table 11.3, the total power produced by the installment of the BKC/DFB-GPP is calculated as 66.1 kW, which is mainly contributed to the two main steam turbines of the geothermal plant. As the DKC/DFB-GPP is employed instead of the basic system, the total power is increased by 2.98%. In this case, the energy and exergy efficiencies are increased from 8.84% to 9.1% (an increase around 2.94%) and from 41% to 42.35% (an increase around 3.29%), respectively. The high power production of the DKC/DFB-GPP can be ascribed to
TABLE 11.3 Comparison of the main performance parameters defined for different simulated GPPs.
Parameter
Basic KC-based DFB-GPP
Double KCbased DFB-GPP
Enhanced triple KCbased DFB-GPP
W_ GPP ðkWÞ
37.9
25.22
25.22
W_ SKC1 ðkWÞ
–
7.03
3.48
W_ SKC2 ðkWÞ
28.2
35.81
26.95
W_ TSKC ðkWÞ
–
–
15.06
W_ net ðkWÞ
66.1
68.07
70.72
en (%)
8.84
9.10
9.46
ex (%)
41
42.35
44.11
the second Kalina cycle, which constitutes nearly 52.6% of the overall power generation of this power plant. This indicates the main shift of power generation from the steam turbines to the employed KCs. The performance of the proposed DKC/DFB-GPP can further be improved by proposing an enhanced triple KCbased DFB-GPP instead of the basic power plant. According to Table 11.3, the role of the second KC is highlighted in this power plant. Although the power generation of the first and third KCs is sacrificed in favor of the second KC, it can be seen that total power is increased around 3.89% in comparison with the DKC/DFB-GPP and 6.98% in comparison with the BKC/DFB-GPP. In terms of thermal performance, the energy efficiency is improved by 7.01%, and hence has become 9.46% (the maximum thermal efficiency attained in this study). In terms of second-law evaluation, the exergy efficiency is enhanced by around 4.15% in comparison with the DKC/DFB-GPP and 7.58% in comparison with the BKC/DFB-GPP. Fig. 11.3 displays the effects of flash tank-1 pressure on the total power, energy efficiency, and exergy efficiency of the three power plants. With the rise of flash tank-1 pressure, the total power for each of the three power plants is increased, where the augmentation of the ETKC/DFBGPP is more substantial than the two other ones. With the rise of flash tank-1 pressure, the TSKC produces more power, while the first SKC generates a low amount of electricity. In addition, the power value generated by the steam turbines increases with the rise of flash tank-1 pressure while the output power of the second SKC remains unvaried during this change. A close investigation uncovers that the total power production of the steam turbines and the TSKC is higher than the power decrement of the SKC-1, and hence the total power of the plants increases with the rise of flash tank-1 pressure. It should also be noted that because a modified KC is used as the second KC in the ETKC/DFB-GPP, the power value generated in the TSKC is very significant, which results in a significantly high power improvement in comparison with the other two power plants. The same justification is also true for the alteration trend of energy and exergy efficiencies because they are directly related to the total power value. Fig. 11.4 investigates the effects of flash tank-2 pressure on the total power, energy efficiency, and exergy efficiency of the simulated power plants. With the rise of flash tank-2 pressure, the total power for each of the three power plants is decreased, where the decrement rate of the ETKC/ DFB-GPP is more substantial than the other two. With the rise of flash tank-2 pressure in the basic system (Fig. 11.1A), the SKC produces more power. The second steam turbine generates a low amount of electricity, which is more considerable than the increment power of the SKC. However, in the DKC/DFB-GPP, with the rise of flash tank2 pressure, the second SKC power slightly increases
Kalina-based binary geothermal power plants Chapter 11
BKC/DFB-GPP
DKC/DFB-GPP
BKC/DFB-GPP
ETKC/DFB-GPP
DKC/DFB-GPP
175
ETKC/DFB-GPP
11.0
80 78
10.5
74
h en (%)
Wnet (kW)
76
72 70
10.0
9.5
68 9.0
66 64
8.5
1000
(A)
1020
1040
1060
1080
1100
1000
PFT1 (kPa)
1020
DKC/DFB-GPP
1060
1080
1100
PFT1 (kPa)
(B) BKC/DFB-GPP
1040
ETKC/DFB-GPP
50 49 48
h ex (%)
47 46 45 44 43 42 41 1000
(C)
1020
1040
1060
1080
1100
PFT1 (kPa)
FIG. 11.3 Effects of flash tank-1 pressure on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
because more thermal heat is supplied to this KC while steam turbine-2 outlet power decreases significantly. The same factor can be reasoned for the third power plant. All in all, these variations lead to a decrease in the total power of all three power plants with the rise of flash tank-2 pressure. As pinpointed previously, because the energy and exergy efficiencies are directly related to the total power value, the same justification can be applied to them. Hence, the energy and exergy efficiencies decrease with the rise of flash tank-2 pressure. Fig. 11.5 displays the effects of the upper pressure of the first SKC (PSKC1) on the total power, energy efficiency, and exergy efficiency of the two new proposed power plants. With the rise of PSKC1, the total power for each power plant is maximized at a specific point. With the rise of PSKC1, the first SKC power reveals a maximum point versus the PSKC1 while the power rate of the other turbines remains constant. As a result, the total power and energy and exergy
efficiencies reveal a maximum point at a specific value of PSKC1. Fig. 11.6 displays the effects of the upper pressure of the second SKC (PSKC(2)) on the total power, energy efficiency, and exergy efficiency of each power plant. With the rise of PSKC(2), the total power for each power plant is increased because the expansion ratio of the turbine of the SKC-2 increases, resulting in the power increment of the SKC-2. In contrast, the power generation rate of other turbines remains unchanged. The same justification is also true for the alteration trend of energy and exergy efficiencies because they are directly related to the total power value. Therefore, the energy and exergy efficiencies also increase with the rise of the upper pressure of the SKC-2. Fig. 11.7 displays the effects of the high and medium pressures of the TSKC (PTSKC) on the total power, energy efficiency, and exergy efficiency of the ETKC/DFB-GPP. With the rise of the upper pressure of the TSKC or with
176
PART II Thermodynamic analysis of geothermal power plants
DKC/DFB-GPP
BKC/DFB-GPP
ETKC/DFB-GPP
12.5
92
12.0
88
11.5
84
11.0
h en (%)
Wnet (kW)
BKC/DFB-GPP
96
80
ETKC/DFB-GPP
10.5
76
10.0
72
9.5
68
9.0
64
8.5 620 640 660 680 700 720 740 760 780 800
(A)
DKC/DFB-GPP
620 640
(B)
PFT2 (kPa) BKC/DFB-GPP
DKC/DFB-GPP
660 680 700 720 740
760 780 800
PFT2 (kPa) ETKC/DFB-GPP
58 56 54
h ex (%)
52 50 48 46 44 42 40 620 640 660 680 700 720 740 760 780 800
(C)
PFT2 (kPa)
FIG. 11.4 Effects of flash tank-2 pressure on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
the decrease of its medium pressure, the total power for the proposed power plant increases because more power is produced by the TSKC, and hence the energy and exergy efficiencies will increase. Fig. 11.8 displays the effects of the basic NH3 concentration of the first SKC (XSKC1) on the total power, energy efficiency, and exergy efficiency of two new proposed power plants. With the rise of XSKC1, the total power for each power plant is minimized to 66.34 kW for the DKC/ DFB-GPP at XSKC1 ¼ 0.66 and 69.7 kW for the ETKC/ DFB-GPP system at XSKC1 ¼ 0.62. As a result, the energy and exergy efficiencies reveal a maximum point at the same concentration. Hence, any design consideration around this point should be avoided. Similar to Fig. 11.8, the effects of the basic NH3 concentration of the SKC-2 and TSKC on the total power, energy
efficiency, and exergy efficiency are also investigated, and results are shown in Figs. 11.9 and 11.10, respectively. According to the results, the main performance criteria also reveal a minimum value with respect to the NH3 concentration of the other two KCs. Fig. 11.11 illustrates the exergy balance diagram, also called the Grassmann diagram, for the whole unit of each simulated power plant, treating each as a unit control volume. According to Fig. 11.11A, under a total input exergy of 162.6 kW for the BKC/DFB-GPP, the exergy destruction of the SKC is more appreciable than the GPP itself while the exergy of loss of the GPP accounts for a higher portion than the exergy of loss of the SKC. Quantitatively speaking, the exergy destruction of the SKC and GPP contributes around 36.82% (59.86 kW) and 9.3% (15.12 kW) of the overall exergy of the fuel, respectively.
Kalina-based binary geothermal power plants Chapter 11
ETKC/DFB-GPP
DKC/DFB-GPP
9.5
71
9.4
70
9.3
h en (%)
Wnet (kW)
DKC/DFB-GPP
72
69
68
ETKC/DFB-GPP
9.2
9.1
67
9.0
2500
(A)
177
3000
3500
4000
2500
PSKC1 (kPa)
3000
4000
PSKC1 (kPa)
(B) DKC/DFB-GPP
3500
ETKC/DFB-GPP
45.0 44.5
h ex (%)
44.0 43.5 43.0 42.5 42.0 41.5 41.0 2500
3000
(C)
3500
4000
PSKC1 (kPa)
FIG. 11.5 Effects of SKC-1 pressure on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
The exergy loss of the SKC and GPP contributes around 3.7% (6.02 kW) and 9.17% (14.91 kW) of the overall exergy of the fuel, respectively. Only 41% of the overall exergy of the fuel (66.66 kW) is converted into the exergy of the product, in which the SKC and GPP constitute around 23.35% (37.97 kW) and 17.65% (28.69 kW), respectively. Therefore, employing SKC in the BKC/DFB-GPP highly contributes to the overall power augmentation of the basic DF-GPP. By considering the DKC/DFB-GPP, the exergy balance for this plant can be depicted and illustrated in Fig. 11.11B. The total exergy of the product is increased to 42.34% (22.31% of SKC-2, 15.51% of GPP, and 4.52% of SKC-1) of the overall exergy rate of the fuel of 163 kW, in comparison with the BKC/DFB-GPP. Also, during this modification, the exergy destruction of SKC after flash tank-2 (SKC in the BKC/DFB-GPP and SKC-2 in the DKC/DFB-GPP) and the exergy loss of GPP are decreased from 59.86 to 53.74 kW
and from 15.12 to 14.47 kW, respectively, underlining the fact that such a modification also improves the lost and destructed exergy of the basic unit. Finally, the exergy balance of the ETKC/DFB-GPP is depicted in Fig. 11.11C in terms of the Grassmann diagram. After this state-of-the-art modification, the total exergy of the product is substantially increased while the exergy destruction of SKC-2 is reduced significantly in comparison with both BKC/DFB-GPP and ETKC/DFB-GPP. Quantitatively speaking, the total exergy of the product is increased to 72.05 kW, indicating that it is improved approximately 8.08% in comparison with the BKC/DFB-GPP and 5.84% in comparison with the DKC/DFB-GPP. In terms of exergy destruction, the destruction exergy of SKC-2 is decreased to 48.52 kW, indicating that it is improved approximately 18.9% in comparison with the BKC/DFB-GPP and 9.71% in comparison with the DKC/DFB-GPP.
BKC/DFB-GPP
DKC/DFB-GPP
BKC/DFB-GPP
ETKC/DFB-GPP
DKC/DFB-GPP
ETKC/DFB-GPP
11.5
86 84
11.0
82
10.5
78
h en (%)
Wnet (kW)
80 76 74
10.0
72
9.5
70 68
9.0
66 64
8.5 2000
(A)
2500
3000
3500
4000
PSKC2 (kPa)
2000
2500
(B) BKC/DFB-GPP
PSKC2 (kPa)
DKC/DFB-GPP
ETKC/DFB-GPP
54 52
h ex (%)
50 48 46 44 42 40 2000
(C)
2500
3000
PSKC2 (kPa)
FIG. 11.6 Effects of SKC-2 pressure on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
3000
3500
4000
3500
4000
ETKC/DFB-GPP
ETKC/DFB-GPP
75
ETKC/DFB-GPP
10.0
High pressure (Ph)
High pressure (Ph)
74 9.8
73
9.6
Medium pressure (Pm)
h (%) en
.
Wnet (kW)
72 71
ETKC/DFB-GPP
70 69
Medium pressure (Pm)
9.4 9.2
68 9.0
67 66
8.8
65 8.6
64 1500
(A)
1800
2100
2400
2700
1500
3000
PTSKC (kPa)
1800
2100
PTSKC (kPa)
(B) ETKC/DFB-GPP
ETKC/DFB-GPP
47 High pressure (Ph)
46 45
h ( %) ex
Medium pressure (Pm)
44 43 42 41 40 1500
(C)
1800
2100
2400
PTSKC (kPa)
FIG. 11.7 Effects of TSKC pressure on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
2400
2700
3000
2700
3000
ETKC/DFB-GPP
DKC/DFB-GPP
9.5
71
9.4
70
9.3
h en (%)
Wnet (kW)
DKC/DFB-GPP
72
69 68
ETKC/DFB-GPP
9.2 9.1 9.0
67
8.9
66
8.8 8.7
65 0.3
(A)
0.4
0.5
0.6
0.7
0.8
0.3
0.9
0.4
0.5
XSKC1
(B)
XSKC1 DKC/DFB-GPP
ETKC/DFB-GPP
44.5 44.0
h ex (%)
43.5 43.0 42.5 42.0 41.5 41.0 40.5 0.3
(C)
0.4
0.5
0.6
0.6
0.7
0.8
XSKC1
FIG. 11.8 Effects of SKC-1 basic NH3 concentration on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
0.9
0.7
0.8
0.9
BKC/DFB-GPP
DKC/DFB-GPP
ETKC/DFB-GPP
BKC/DFB-GPP
74
DKC/DFB-GPP
ETKC/DFB-GPP
10.0 9.8
72
hen (%)
Wnet (kW)
9.6
70 68
9.4 9.2 9.0
66
8.8
64 62 0.60
(A)
8.6
0.65
0.70
0.75
0.80
0.85
0.90
8.4 0.60
0.95
XSKC2
0.65
0.70
0.75
XSKC2
(B) BKC/DFB-GPP
DKC/DFB-GPP
ETKC/DFB-GPP
45
hex (%)
44 43 42 41 40 39 0.60
(C)
0.65
0.70
0.75
0.80
0.80
0.85
0.90
XSKC2
FIG. 11.9 Effects of SKC-2 basic NH3 concentration on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
0.95
0.85
0.90
0.95
ETKC/DFB-GPP
ETKC/DFB-GPP
74
10.0
72
h (%) en
Wnet (kW)
9.5
70 68
9.0
66 8.5
64 8.0
62 0.3
(A)
0.4
0.5
0.6
0.7
0.8
0.3
0.9
0.4
0.5
ETKC/DFB-GPP
45 44
hex (%)
43 42 41 40 39 0.3
(C)
0.4
0.5
0.6
0.6
XTSKC
(B)
XTSKC
0.7
0.8
XTSKC
FIG. 11.10 Effects of TSKC basic NH3 concentration on (A) total power, (B) energy efficiency, and (C) exergy efficiency.
0.9
0.7
0.8
0.9
Kalina-based binary geothermal power plants Chapter 11
28.69 kW ExPr,GPP (17.65%) Ex C K
Ex
PP G L,
S L,
Ex
ExFu,tot
162.6 kW (100%)
37.97 kW Ex Pr,SKC (23.35%)
PP ,G D
Ex
ExPr,tot
,S D C K
66.66 kW (41%)
15.12 kW (9.303%)
59.86 kW (36.82%)
6.028 kW (3.707%)
14.91 kW (9.17%)
SK C2
,S C1
K
D PP
,G
L, PP
G
C2
K
,S
D
14.47 kW (8.882%)
53.7 kW (32.98%)
25.22 kW (15.51%)
ExPr,GPP
7.037 kW (4.525%)
ExPr,SKC1
ExPr,tot
0.623 kW 6.046 kW (0.382%) (3.71%)
9.248 kW (5.675%)
ExPr,SKC2
C1
L,
D
Ex
Ex
Ex
9.82 kW (6.026%)
35.81 kW (22.31%)
SK
L,
Ex
Ex
ExFu,tot
Ex
163 kW (100%)
(A)
68.07 kW (42.345%)
C
7.239 kW (4.432%)
FIG. 11.11 Grassmann diagram for the exergy balance of the power plants.
C1
K
TS C1
K
,S
PP
C2
K ,S
14.46 kW (8.851%)
9.896 kW (6.058%)
4.55 kW (2.785%)
SK
L,
D
PP
,G
C
SK
D
G L,
D
48.52 kW (29.71%)
C2
SK
L,
,T
D
Ex
Ex
Ex
(C)
L,
Ex Ex Ex
Ex
ExFu,tot
Ex
163.3 kW (100%)
(B)
4.534 kW (2.777%)
ExPr,tot 72.05kW
0.3161 kW (44.12%) 1.783 kW (0.1935%) (1.092%)
27.36 kW (16.75%)
ExPr,SKC2
25.27 kW (15.47%)
ExPr,GPP
15.84 kW (9.70%)
ExPr,TSKC
3.579 kW (2.20%)
ExPr,SKC1
183
184
11.5
PART II Thermodynamic analysis of geothermal power plants
Closing remarks
This study used the Kalina cycle (KC) as a binary cycle of a DF-GPP in two innovative scenarios due to the lowtemperature characteristics of KC. The present study demonstrated that present devised enhanced KC-based geothermal power plants could supersede the previous ones in terms of the First and Second Laws of Thermodynamics. Quantitatively speaking, it was demonstrated that the energy and exergy efficiencies of the best scenario could be enhanced by 7.01% and 7.58% when two simple KCs and a two-stage KC were used in the DF-GPP in the proposed configuration. Also, the total exergy of the product was improved by approximately 8.08% in comparison with the BKC/DFB-GPP and 5.84% in comparison with the DKC/DFB-GPP. In terms of exergy destruction, the destruction exergy of SKC-2 was decreased by approximately 18.94% in comparison with the BKC/DFB-GPP and 9.71% in comparison with the DKC/DFB-GPP.
References [1] Lee I, Tester JW, You F. Systems analysis, design, and optimization of geothermal energy systems for power production and polygeneration: state-of-the-art and future challenges. Renew Sust Energy Rev 2019;109:551–77. [2] Gholizadeh T, Vajdi M, Rostamzadeh H. A new trigeneration system for power, cooling, and freshwater production driven by a flash-binary geothermal heat source. Renew Energy 2019;148:31–43. [3] Wang Y, Du Y, Wang J, Zhao J, Deng S, Yin H. Comparative life cycle assessment of geothermal power generation systems in China. Resour Conserv Recycl 2020;155:104670. [4] Manente G, Bardi A, Lazzaretto A, Paci M. Low emission flashbinary and two-phase binary geothermal power plants with water absorption and reinjection of non-condensable gases. Geothermics 2019; 80:155–69.
[5] Kolahi M-R, Nemati A, Yari M. Performance optimization and improvement of a flash-binary geothermal power plant using zeotropic mixtures with PSO algorithm. Geothermics 2018;74:45–56. [6] Ratlamwala T, Dincer I. Comparative efficiency assessment of novel multi-flash integrated geothermal systems for power and hydrogen production. Appl Therm Eng 2012;48:359–66. [7] Salehi S, Mahmoudi SMS, Yari M, Rosen M. Multi-objective optimization of two double-flash geothermal power plants integrated with absorption heat transformation and water desalination. J Clean Prod 2018;195:796–809. [8] Ashraf MA, Liu Z, Li C, Peng W-X, Ghaebi H. Proposal and comprehensive analysis of an innovative CCP plant based on an internal integration of double flash power system and ejector refrigeration cycle. Energy Convers Manage 2020;203:112232. [9] Wang J, Ren C, Gao Y, Chen H, Dong J. Performance investigation of a new geothermal combined cooling, heating and power system. Energy Convers Manage 2020;208:112591. [10] Abdolalipouradl M, Khalilarya S, Jafarmadar S. Exergoeconomic analysis of a novel integrated transcritical CO2 and Kalina 11 cycles from Sabalan geothermal power plant. Energy Convers Manage 2019;195:420–35. [11] Mokarram NH, Mosaffa A. A comparative study and optimization of enhanced integrated geothermal flash and Kalina cycles: a thermoeconomic assessment. Energy 2018;162:111–25. [12] Aali A, Pourmahmoud N, Zare V. Exergoeconomic analysis and multi-objective optimization of a novel combined flash-binary cycle for Sabalan geothermal power plant in Iran. Energy Convers Manage 2017;143:377–90. [13] Feili M, Rostamzadeh H, Ghaebi H. A new high-efficient cooling/ power cogeneration system based on a double-flash geothermal power plant and a novel zeotropic bi-evaporator ejector refrigeration cycle. Renew Energy 2020. [14] Bahrampoury R, Behbahaninia A. Thermodynamic optimization and thermoeconomic analysis of four double pressure Kalina cycles driven from Kalina cycle system 11. Energy Convers Manage 2017;152:110–23. [15] Mathieu-Potvin F. Self-superheating: a new paradigm for geothermal power plant design. Geothermics 2013;48:16–30.
Chapter 12
Combined cooling and power production from geothermal resources Tahir Abdul Hussain Ratlamwala, Uzair Aziz Suria, Rao Hamza Jamil, Mohammad Mustafa Pardesi, and Mohammad Ashar Jamal Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Nomenclature m_ S P T x h s v Tc Te xLiBr Q_ c Q_ e ex _ dest Ex _ W h
mass flow rate (kg/s) salinity pressure (kPa) temperature (K) steam quality specific enthalpy (kJ/kg) specific entropy (kJ/kg K) specific volume (m3/kg) condenser temperature (K) evaporator temperature (K) LiBr mass fraction condenser heat rate (kW) evaporator heat rate (kW) specific exergy (kJ/kg) exergy destruction rate (kW) work rate (kW) efficiency
Subscripts cond turb sat p ph 0 ev sc hx b th gen a c e SHX
condenser turbine saturation pump preheater (for ORC) used to denote dead state of a fluid expansion valve separation chamber heat exchanger boiler (for ORC) thermal generator (vapor absorption cycle) absorber (vapor absorption cycle) condenser (vapor absorption cycle) evaporator (vapor absorption cycle) solution heat exchanger (vapor absorption cycle)
Acronyms EES GHG
Engineering Equation Solver greenhouse gases
GPD IAPWS IPCC LiBr, LiBr Water MGD NH3 - H2O ORC SEARS TTD VAC
12.1
gallons per day International Association for the Properties of Water and Steam Intergovernmental Panel on Climate Change aqueous lithium bromide million gallons per day ammonia-water mixture, or aqueous ammonia organic Rankine cycle single effect absorption refrigeration system terminal temperature difference vapor absorption cycle
Introduction
Before the industrial revolution of 1760, the mechanized power required for various industrial processes was obtained through water and wind potential [1]. One of the most iconic examples of wind power utilization is the windmill towers, which converted wind power to the rotational power required to run the milling process (of grains, etc.). Due to the limited understanding of scientific principles, especially in the field of fluid mechanics, these methods of harnessing energy were very inefficient, unreliable, costly, and with very little degree of control. This was because vanes designed to harness wind energy (or hydropotential) were not very efficient or optimized, and due to the absence of knowledge about advanced fluid mechanic principles, optimization of the previous design was a very difficult if not impossible task. Moreover, humans have very little control over the speed of the wind striking the windmills, or the direction from which the wind flows, rendering mechanized power generated from windmills to be quite unreliable in terms of continuity. These problems, coupled with the fact that manufacturing such devices was quite costly, made this an unattractive choice. Therefore, the inventors of that period were constantly trying to develop new methods of generating mechanized power.
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00018-4 Copyright © 2021 Elsevier Inc. All rights reserved.
185
186 PART
II Thermodynamic analysis of geothermal power plants
Fossil fuels and especially coal provided an attractive alternative that was quite reliable due to its abundance. The initial designs of coal-powered steam engines were very inefficient and hence required a huge amount of fuel. These initial designs were only used where circumstances forced their use or where convenience was more important than cost [1]. With time, the engine design became more efficient, and the use of steam engines became economical. Since then, human comfort needs kept increasing, and the unabated use of fossil fuel to meet those demands has led to a significant reduction of fossil fuel reserves across the world. Apart from depleting fossil reserves, the world has just begun to understand the unintended consequences to the Earth’s environment of using fossil fuels. One of the major concerns is the global warming phenomenon. Global warming refers to the increase in average atmospheric temperature resulting in more extreme and severe weather conditions such as frequent droughts, heavy downpours, severe hurricanes, and extreme heat waves [2]. This severe weather will shape the future of the global climate, which will undoubtedly be a very hostile one. Global warming has also been linked to rising sea levels and the depletion of arctic ice [3–5] and glaciers across the world [6, 7]. The Intergovernmental Panel on Climate Change (IPCC) has determined that primarily human activities—and more specifically the production of greenhouse gases (GHG) from fossil fuel consumption—are responsible for the observed global warming phenomenon [8]. In the United States from 1990 to 2017, GHG emissions due to electricity production, heat production, transportation, and industrial processes (primarily by burning fossil fuels for energy) accounted for approximately 78% of the total US GHG emissions [9]. Globally in 2010, these same parameters contributed toward approximately 60% of the total GHG emissions [10]. Thus, using renewable energy sources such as geothermal can lead to a significant reduction in GHG emissions and consequently help in recovery from the adverse effects of global warming. Geothermal energy is essentially the thermal energy stored and generated inside the Earth’s crust due to the decaying radioactive process of matter inside the crust. It is considered one of the most clean, renewable, and sustainable forms of thermal energy available [11, 12]. This chapter deals with the design and analysis of a geothermal power generation system that, in addition to generating electric power, provides space cooling, space heating, water heating, and water desalination. The main focus of this design shall be to utilize the geothermal potential to its maximum and to produce electrical energy. This will be achieved by using three different power generation cycles: high-pressure steam power generation, medium-pressure steam power generation, and the organic Rankine cycle.
The organic Rankine cycle shall also produce one secondary output, that is, heating water at ambient temperature to 60°C. Afterward, the remaining thermal energy from the geothermal fluid shall be used to power two different secondary cycles: the thermal flash desalination cycle and the lithium bromide vapor absorption cycle. The outputs expected from these secondary systems are water desalination and space heating/cooling. The upcoming sections of this chapter give a detailed overview of the whole system, the mathematical modeling of the system in Engineering Equation Solver, and finally the thermodynamic analysis of the multigeneration plant.
12.2
System description
The whole process starts with the extraction of geothermal fluid from a high-pressure well at the saturated liquid state. It is then passed through multiple subsystems to produce the required outputs. Essentially, there are five subsystems of the whole plant (refer to Fig. 12.1) and they are as follows 1. 2. 3. 4. 5.
High-pressure steam power generation cycle. Medium-pressure steam power generation cycle. Organic Rankine cycle. Water desalination cycle. LiBr vapor absorption refrigeration cycle.
The outputs expected from the above subsystems are as follows 1. 2. 3. 4. 5.
Electrical power. Desalinated water. Water heating. Space cooling. Space heating.
12.2.1 High-pressure steam power generation cycle The thermodynamic state of incoming high-pressure geothermal fluid usually varies from saturated liquid to saturated vapor. In this instance, the saturated liquid state at 60 bars (548.7 K) is assumed. Fig. 12.2 shows the highpressure steam power cycle scheme. All further explanations regarding the process refer to the state numbers defined in the figure. To generate electricity from a steam turbine, it is essential that the saturated liquid (at state 1) must be converted to vapor, and this is accomplished by flashing the saturated liquid (state 1 to state 2) at a pressure of 31 bars (saturation temperature of 509 K). After flashing, the saturated liquid-vapor mixture is passed through a separation chamber 1, where the mixture is separated into saturated vapor and saturated liquid (state 2 to state 3 and state 4).
Combined cooling and power production from geothermal resources Chapter
12
187
Condenser/ r Evaporator (unless labelled otherw r ise)
Turbine 1
Turbine 2 Expansion Va V lve 11
Qc1
Pump
5 12
The heat in to generator is supplied by geothermal fluid (states [25] to [37])
Qg *
13
Separation Chamber 1
3
Atmosphere 6 4
Heat Exchanger 1
2 EV1
38 7
EV6
Separation Chamber 2
8 EV2
Qc
9
40 Qg * Generator
1
46
25 37
10 41 16
39
Boiler PreHeater
Heat Exchanger
18
17
EV5 42
43
14
47
EV4
24
44 23
Cold Water in
45
ORC Turbine 21
Absorber Qa
19 20
48 Evaporator
Hot Water out
22 Qe 15
32
31
Fresh Water Storage
Heat Exchanger 2
30 28
EV3 Separation Chamber 3
33
29
Heat Exchanger 3
27 26
24
35
Mixing Chamber
34
EV7
36
Saline Water Storage
Atmosphere
FIG. 12.1 Schematic representation of a complete multigeneration plant.
The saturated vapor is then expanded in the high-pressure steam turbine (state 3 to state 5), where the enthalpy of the saturated vapor is converted to mechanical energy. The outlet pressure of the turbine (700.5 kPa) is selected such that the outlet fluid does not have a vapor quality less than 90%. The outlet fluid from the turbine is then sent to the shell side of condenser 1. The saturated liquid from the
separation chamber (state 4) is transmitted to the tube side of condenser 1. After passing through the condenser (state 4 to state 7), it is flashed (state 7 to state 9) to pressure that is equal to the high-pressure turbine outlet (700.5 kPa). The saturated mixture is separated in separation chamber 2 (state 9 to state 8 and state 10), and the vapor is sent to the shell side of condenser 1.
188 PART
II Thermodynamic analysis of geothermal power plants
the saturated liquid at 509 K (state 4 to state 7) from separation chamber 1. The superheated vapor is then expanded in a medium-pressure turbine (state 6 to state 11), and the enthalpy of the vapor is converted to mechanical energy. The outlet pressure of the turbine is selected such that the outlet fluid does not have a saturated vapor quality less than 90%. The outlet from the medium-pressure turbine is then condensed (state 11 to state 12), and the condensate is pumped up to atmospheric pressure (state 12 to state 13).
Turbine 1
5
Separation Chamber 1
3
6 4
Heat Exchanger 1
2 EV1
7
Separation Chamber 2
8 EV2 9
1 10
FIG. 12.2 High-pressure steam power generation cycle schematic representation.
12.2.2 Medium-pressure steam power generation cycle Fig. 12.3 provides a schematic overview of the second phase of the multigeneration system. 90% vapor from the highpressure turbine outlet and the saturated vapor from separation chamber 2 are heated to 506 K (state 5 and state 8 to state 6), and the thermal energy required is obtained from FIG. 12.3 Medium-pressure steam power generation cycle schematic representation.
12.2.3
Organic Rankine cycle
Fig. 12.4 gives a pictorial view of the processes involved in the organic Rankine cycle integrated with geothermal steam power plants. The working fluid (organic) for this cycle was selected to be n-pentane. The saturated liquid fluid (state 20) at a pressure of 75.94 kPa is pumped to the pressure of 101.3 kPa (state 23). Afterward, the organic fluid is passed through a preheater where the fluid is preheated (state 23 to state 17), and then through the boiler where the fluid is superheated (state 17 to state 18). The thermal energy supplied to the organic fluid in the boiler comes from the saturated geothermal liquid coming from separation chamber 2 (state 10) at 438 K, and after the boiler (state 14), the geothermal fluid is divided into three fractions. They subsequently go on to provide thermal energy to the preheater in the organic Rankine cycle, the LiBr absorption refrigeration cycle, and the water desalination process. After the boiler, the superheated organic fluid is expanded in the turbine (state 18 to state 19), and then the fluid is condensed to a saturated liquid state (state 19 to state 20). Because the condenser transfers heat from the organic fluid to the cold water at ambient temperature (state 21 to state 22), and the organic fluid has to be saturated liquid to ensure proper pump functioning, the temperature of the incoming water (or the ambient temperature) determines the outlet pressure of the turbine.
6
Turbine 2
11
Qc1 12
13
Atmosphere
Combined cooling and power production from geothermal resources Chapter
12
189
FIG. 12.4 Schematic representation of the organic Rankine cycle with n-pentane as the working fluid.
10
16
Boiler PreHeater 18
17
14
25
24 15 23
Cold Water in
ORC Turbine
21 19
20
Hot Water out
12.2.4
22
Water desalination process
12.2.5
The desalination methodology used in this phase is known as flash desalination (see Fig. 12.5). The saline water at atmospheric pressure and temperature (state 26) is pumped up to a higher pressure of 200 kPa (state 27). The high-pressure saline water is then passed through a condenser (state 27 to state 30), where it absorbs heat from the incoming saturated vapor of pure water. Afterward, the saline water is passed through a heat exchanger where it absorbs heat (state 30 to state 31) from the incoming geothermal fluid (one of the fractions after the ORC boiler, state 15 to state 32). The saline water is then flashed (state 31 to state 33) to produce saturated vapor and liquid (state 33 to state 25 and state 34). The saturated vapor is then condensed (state 28 to state 29) by passing through the condenser and transferring heat to high-pressure saline water.
15
Lithium-bromide absorption cycle
Even after the geothermal fluid has gone through three different cycles to produce electricity, it still possesses enough specific enthalpy to power other processes. Therefore, instead of utilizing electricity to produce space cooling through the vapor compression cycle, it is more sensible to utilize the heat to general space cooling through the vapor absorption cycle. The vapor absorption cycle has a complex design and thus requires greater initial investment, space, and technical expertise to operate. On the other hand, the vapor absorption cycle is especially advantageous in this case because it requires very minimal electrical power to operate its refrigerant pump, which contrasts with the vapor compression cycle that consumes high electrical power to operate its refrigerant compressor. But before moving on to the
32
31
Fresh Water Storage
Heat Exchanger 2
30 28
EV3
Separation Chamber 3
33
29
Heat Exchanger 3
27 26
24
35
Mixing Chamber
34
FIG. 12.5 Thermal flash desalination process diagram.
EV7
36
Atmosphere
Saline Water Storage
190 PART
II Thermodynamic analysis of geothermal power plants
FIG. 12.6 LiBr absorption cooling cycle. Note that in the case of a generator, the heat supplied to the refrigerant is provided by the incoming geothermal fluid.
Atmosphere
38
EV6
Qc 40 Qg * Generator
46
25 37 41
39
Heat Exchanger EV5 42
43 47
EV4 44
45
Absorber Qa 48 Evaporator Qe
processes (see Fig. 12.6) required for vapor absorption cooling, two extremely important considerations should be explained. First, for the vapor absorption cycle to function properly, the generator must be at temperatures above 80°C. The second consideration, which relates directly to the selection of the type of vapor absorption cycle, is the purpose of space cooling. There are essentially two types of vapor absorption cycles. The first is the ammonia-water absorption cycle (also known as the NH3-H2O absorption cycle), which utilizes a mixture of ammonia and water in its cycle while the extracted ammonia is used as the refrigerant. The other type is the lithium-bromide absorption cycle (also known as the LiBr absorption cycle), which utilizes aqueous lithium bromide in its cycle, and the extracted water is used as the refrigerant. For the current system, the purpose of space cooling is to provide air conditioning to human beings. Therefore, due to its harmful effects, the ammonia-water solution cannot be used as a refrigerant. Hence, the aqueous lithium bromide solution is chosen due to its unharmful nature. The geothermal fluid fraction from the ORC boiler outlet transfers heat to the LiBr-water solution inside the
generator (state 25 to state 37). The weak LiBr-water solution (state 39) gets heated, and as a result, a fraction of water absorbed in the LiBr is released in vapor form (state 40). The remaining solution becomes more concentrated due to the extraction of water from it (state 41). The water vapor coming from the generator is then condensed to a saturated liquid state in a condenser (state 40 to state 46), where the heat rejected is utilized for space heating. The condensed liquid is then throttled to a lower pressure. This pressure is determined by setting the evaporator temperature to be the saturation temperature and determining the corresponding saturation pressure. After throttling, the refrigerant is in the saturated liquid-vapor state (state 47). Afterward, the refrigerant is passed through the evaporator, where the refrigerant absorbs heat from the space being cooled. After the evaporator, the refrigerant is sent to the absorber (state 48). The strong solution exiting from the generator is passed through a heat exchanger and recovers some heat (state 41 to state 42) that was provided to the solution in the generator. And then, the heat is transferred to the incoming weak solution (state 43 to state 39). After the heat exchanger, the strong solution is then throttled to
Combined cooling and power production from geothermal resources Chapter
a lower pressure of the absorber (state 44). The strong solution (state 44) coming into the absorber is cooled by rejecting some heat into the atmosphere. The lower temperature of the solution increases its capability to absorb water vapor, and therefore, the vapor incoming from the evaporator is absorbed to make a weak LiBr-Water solution (state 45). This weak solution is then pumped to a higher pressure of the generator (state 43) and is passed through the heat exchanger to gain some heat from the outcoming strong solution from the generator.
12.3 Thermodynamic analysis Whenever an energy system is designed, it is necessary to perform thermodynamic analysis to determine whether the design is feasible. Traditionally, thermodynamic analysis usually refers to the energy analysis of the system. The First Law of Thermodynamics, also known as the law of conservation of energy, states that energy can neither be created nor destroyed, but can change its form (e.g., chemical energy to thermal energy). Thus, the energy analysis allows a designer to account for all the energy going into the system and coming out of the system. If these quantities do not match each other, it is an indication that either the energy analysis was done incorrectly or the suggested design is not possible. However, the energy analysis does not take irreversibility or entropy generation into account, and the determined energy efficiency may give the wrong picture. Consider, as an example, a large body of water at environmental temperature (and with no gravitational potential). This body will also have a huge amount of energy (enthalpy), but little or no work can be extracted from it; therefore, any system design to use the energy of this body will have an efficiency that is either zero or very close to zero. Therefore, along with energy analysis, there is a need to analyze the system in a way which accounts for the quality of the energy available. This is where exergy analysis comes in. Exergy analysis uses the concept of entropy generation, along with the concept of energy, to specify the quality of the energy that is available. The specific flow exergy at any state can be defined by the following equation exi ¼ ðhi h0 Þ T0 ðsi s0 Þ
(12.1)
Another important formulation is the thermal exergy rate rejected in a condenser/evaporator unit, which is represented by T0 _ _ Ex th,c ¼ 1 (12.2) Qc Tc _ th,e ¼ 1 T0 Q_ e Ex (12.3) Te
12
191
The following sections discuss the equations and relations used to develop the mathematical model for the whole system. All state properties can be evaluated using property tables, or as in this case, a state property database with appropriate software. The specific flow exergy of a given state can be calculated using Eq. (12.1). Once all the states of the system have been defined, the next step is to evaluate the efficiencies of the system. Again, there are two ways to define the efficiency of the system; it is either done with reference to incoming energy or with reference to incoming exergy. For energy (or First Law) efficiency evaluation, the following formula may be used X Energy Output Extracted (12.4) I ¼ X Energy Input Required while for exergy (or Second Law) efficiency, the following formula may be used. X Exergy Output Extracted II ¼ X (12.5) Input Exergy Streams + W_ in It is worth mentioning here that due to the complex nature of the design, quite a few results must be calculated through iterative numerical methods. Therefore, the software Engineering Equation Solver (EES) is required to reduce manual labor and errors. Before discussing each subcomponent in isolation, the assumptions and generalizations on which the mathematical model is based upon should be made clear: l
l
l
l l
l
l
Atmospheric temperature is assumed to be 298 K (25°C), and atmospheric pressure is assumed to be 101.325 kPa. Geothermal fluid is available at a pressure of 60 bar, in a saturated liquid state with a mass flow rate of 15 kg/s, and the initial flashing is done at 3100 kPa (just above the limit of minimum pressure for a high-pressure steam turbine). Isentropic turbine and pump efficiencies are assumed to be 90%. Saline water has a salinity of 35 g/kg. The terminal temperature difference (TTD) in heat exchangers is 5°F or 2.8°C (not applicable in a vapor absorption cycle). In a vapor absorption cycle only, the heat exchanger has an efficiency of 80%. The thermodynamic states of geothermal fluids are evaluated from the IAPWS steam/water library.
12.3.1 High-pressure steam power generation cycle As discussed in Section 12.2.1, the geothermal fluid from the well comes at a pressure of 60 bar (or 6000 kPa) at a
192 PART
II Thermodynamic analysis of geothermal power plants
saturated liquid state. Therefore, the saturation phase at the pressure of 60 bar and the quality of “0” fixes the thermodynamic state of the incoming fluid. The mass balance equation of the fluid across the throttling valve is given by m_ 1 ¼ m_ 2
(12.6)
Similarly, because the enthalpy of the fluid remains unchanged after passing through the throttling valve, the enthalpy relation is given by h1 ¼ h2
(12.7)
The exergy balance for the throttling valve can be written as _ dest,ev1 m_ 1 ex1 ¼ m_ 2 ex2 + Ex
(12.8)
Afterward, the fluid enters the separation chamber, and the mass balance relation for the separation chamber is given by the following equations
vapor, and the vapor is sent to the heat exchanger to mix with high-pressure turbine exit fluid and gain heat from saturated liquid from separation chamber 1. Hence, the mass balance can be written as m_ 4 + m_ 5 + m_ 8 ¼ m_ 6 + m_ 7
(12.17)
The energy and exergy balances of the heat exchanger are represented by the following m_ 4 h4 + m_ 5 h5 + m_ 8 h8 ¼ m_ 6 h6 + m_ 7 h7
(12.18)
_ dest,hx1 m_ 4 ex4 + m_ 5 ex5 + m_ 8 ex8 ¼ m_ 6 ex6 + m_ 7 ex7 + Ex (12.19) The exiting fluid (at state 7) is then throttled to turbine 1 pressure. The mass balance for the throttling valve is given by m_ 7 ¼ m_ 9
(12.20)
m_ 2 ¼ m_ 3 + m_ 4
(12.9)
Similarly, because the enthalpy of the fluid remains unchanged after passing through the throttling valve, the enthalpy relation and the exergy balance are given by
m_ 3 ¼ m_ 2 x2
(12.10)
h7 ¼ h9
(12.21)
m_ 4 ¼ m_ 2 ð1 x2 Þ
(12.11)
_ dest,ev2 m_ 7 ex7 ¼ m_ 9 ex9 + Ex
(12.22)
The overall energy balance of the separation chamber is given by m_ 2 h2 ¼ m_ 3 h3 + m_ 4 h4
(12.12)
While the exergy balance for the separation chamber may be written as _ dest,sc1 m_ 2 ex2 ¼ m_ 3 ex3 + m_ 4 ex4 + Ex
(12.13)
The saturated vapor from the separation chamber is then expanded through the turbine to generate mechanical energy. The mass balance for the turbine is m_ 3 ¼ m_ 5
(12.14)
The energy and exergy balance equations for the turbine are given by m_ 3 h3 ¼ m_ 5 h5 + W_ turb1 _ dest,turb1 m_ 3 ex3 ¼ m_ 5 ex5 + W_ turb + Ex
Afterward, the fluid (at state 9) enters separation chamber 2, and the mass balance relation for the separation chamber is given by the following equations m_ 9 ¼ m_ 8 + m_ 10
(12.23)
m_ 8 ¼ m_ 9 x9
(12.24)
m_ 10 ¼ m_ 9 ð1 x9 Þ
(12.25)
The overall energy balance of the separation chamber is given by m_ 9 h9 ¼ m_ 8 h8 + m_ 10 h10
(12.26)
The exergy balance for the separation chamber may be written as _ dest, sc1 m_ 9 ex9 ¼ m_ 8 ex10 + m_ 8 ex10 + Ex
(12.27)
(12.15) (12.16)
It should be noted again that for the durability of turbine blades, the quality of vapor exiting the turbine must not go below 90%, and therefore the outlet pressure is determined by this factor through iterative solutions. After the fluid exits from the turbine at the quality of 90%, it is sent to a heat exchanger, where it gains heat from the saturated liquid coming from separation chamber 1. Meanwhile, the exiting fluid (at state 7) from the same heat exchanger is flashed to a lower pressure (equal to turbine exit), and the resulted fluid is separated into liquid and
12.3.2 Medium-pressure steam power generation cycle In this cycle, the superheated steam exiting heat exchanger 1 (state 6) is expanded through a medium-pressure turbine. Similar to the high-pressure turbine, the outlet pressure is set such that the outlet fluid does not go below the quality of 90%. The mass balance for the medium-pressure turbine is as follows m_ 6 ¼ m_ 11
(12.28)
Combined cooling and power production from geothermal resources Chapter
The energy and exergy balances of the medium-pressure turbine are given by
12
193
The energy and exergy balances for the condenser are given by
and the vapor used to generate electricity, but this flashing will again result in a saturated liquid at even lower temperatures. Therefore, to reduce the complexity and cost of the system, the ORC is used. In this instance, the working fluid of the cycle is an organic fluid called n-pentane, and the cycle consists of a preheater, boiler, turbine, condenser, and liquid pump. Because the turbine outlet must be condensed by rejecting heat to the environment, the outlet pressure of the turbine is defined by the environmental temperature. That pressure is evaluated by setting T20 as the saturation temperature and finding the corresponding saturation pressure. The main source of heat energy required to drive this cycle comes from the saturated liquid from separation chamber 2. This liquid goes into a heat exchanger serving as a boiler. The mass balance equations of this boiler are
m_ 11 h11 ¼ m_ 12 h12 + Q_ cond
(12.32)
m_ 10 ¼ m_ 14
(12.38)
(12.33)
m_ 17 ¼ m_ 18
(12.39)
m_ 6 h6 ¼ m_ 11 h11 + W_ turb2 _ dest,turb2 m_ 6 ex6 ¼ m_ 11 ex11 + W_ turb + Ex
(12.29) (12.30)
The outlet pressure, in this case, goes below the atmospheric pressure. So, for the fluid to be rejected to the atmosphere, the pressure must be greater than or equal to that of the atmosphere. Hence, to reduce pressurization energy, the fluid is first condensed to the saturated liquid form, and afterward, it is pumped to the atmospheric pressure. The mass balance for the condenser is m_ 11 ¼ m_ 12
_ th,cond + Ex _ dest,cond m_ 11 ex11 ¼ m_ 12 ex12 + Ex where
(12.31)
The energy and exergy balances of this boiler can be written as
T0 _ Ex th,cond ¼ 1 Q_ cond Tcond
m_ 10 h10 + m_ 14 h14 ¼ m_ 17 h17 + m_ 18 h18
_ dest,b (12.41) m_ 10 ex10 + m_ 14 ex14 ¼ m_ 17 ex17 + m_ 18 ex18 + Ex
The mass balance for the pump is given by m_ 12 ¼ m_ 13 ¼ m_ p
(12.34)
The power consumed by the pump is given by v12 ðP13 P12 Þ W_ p ¼ m_ p p
(12.35)
It is worth noting that liquid pumping requires considerably lower power consumption (due to a negligible change in the volume of the liquid during pumping) than the power required for gas compression (or the power generation of gas expansion through the turbine). Hence the expansion to subatmospheric pressure in the turbine is justified for this design. The overall energy and exergy balances for the pumping process can be written as m_ 12 h12 + W_ p ¼ m_ 13 h13 _ dest,p m_ 12 ex12 + W_ p ¼ m_ 13 ex13 + Ex
12.3.3
(12.40)
The superheated organic fluid then expands in the turbine to generate mechanical energy. The following relations are necessary to model the turbine Mass balance m_ 18 ¼ m_ 19
(12.42)
Energy and exergy balances m_ 18 h18 ¼ m_ 19 h19 + W_ ORC _ dest,turb, ORC m_ 18 ex18 ¼ m_ 19 ex19 + W_ ORC + Ex
(12.43) (12.44)
The organic fluid exiting from the turbine is then condensed by passing through a heat exchanger and transferring heat to incoming cold water. The mass balances for this process are m_ 19 ¼ m_ 20
(12.45)
(12.36)
m_ 21 ¼ m_ 22
(12.46)
(12.37)
The energy and exergy relations that describe this process can be written as
Organic Rankine cycle
The organic Rankine cycle (ORC) is mainly beneficial in the recovery of heat energy at low temperatures. The saturated liquid at separation chamber 2 has high enthalpy but a relatively low temperature (hence a lower quality). Theoretically, this fluid can further be flashed to a lower pressure
m_ 19 h19 + m_ 21 h21 ¼ m_ 20 h20 + m_ 22 h22
(12.47)
_ dest, cond,ORC m_ 19 ex19 + m_ 21 ex21 ¼ m_ 20 ex20 + m_ 22 ex22 + Ex (12.48) After the fluid has condensed, it is pumped to a higher pressure and the necessary relations to model the pumping process are as follows.
194 PART
II Thermodynamic analysis of geothermal power plants
The mass balance of the process is given by m_ 20 ¼ m_ 23 ¼ m_ p,ORC
(12.49)
The power consumed by the pump given by v20 ðP23 P20 Þ W_ p,ORC ¼ m_ p,ORC p
(12.50)
The overall energy and exergy balances for the pumping process can be written as m_ 20 h20 + W_ p,ORC ¼ m_ 23 h23 _ dest,p,ORC m_ 20 ex20 + W_ p,ORC ¼ m_ 23 ex23 + Ex
(12.51) (12.52)
Finally, the organic fluid at higher pressure is then passed through a heat exchanger functioning as a preheater to heat the fluid to a saturated liquid state. Hence, the mass fraction of the fluid (at state 14) required is determined by the amount of heat required to heat the fluid to its saturated liquid state. The mass balance equations that describe the preheating process are given by m_ 23 ¼ m_ 17
(12.53)
m_ 16 ¼ m_ 24
(12.54)
The energy and exergy balances for the preheating process can be written as m_ 23 h23 + m_ 16 h16 ¼ m_ 24 h24 + m_ 17 h17
(12.55)
_ dest, ORC,ph m_ 23 ex23 + m_ 16 ex16 ¼ m_ 24 ex24 + m_ 17 ex17 + Ex (12.56)
12.3.4
Thermal flash desalination
After passing through three different power cycles, most of the geothermal fluid’s thermal energy has been utilized, and the geothermal fluid only exists as the source of low-grade thermal energy. This low-grade energy is very suitable to drive a thermal flash desalination process [13, 14]. For the sake of simplicity, a single-effect flash desalination is considered. In this process, the heat energy of the geothermal fluid is used to raise the temperature of seawater at 200 kPa; afterward, the hot seawater is throttled to atmospheric pressure, and the saturated mixture is separated into vapor and liquid. The saturated vapor is then condensed by rejecting heat to incoming cold seawater (thermal recovery). Finally, the condensed freshwater is stored in a freshwater storage tank. The mathematical model uses seawater properties given by [15], which uses several studies [16–19] to develop the relations. The salinity (denoted by S) of saline water is taken to be 35 g/kg. It should also be noted that because the seawater library does not give state properties that lie under the saturation
dome, the quality of the liquid vapor mixture is approximated using enthalpy at the saturation temperature (0.1°C is deducted from Tsat to get outside the saturation dome), the latent heat of vaporization, and the enthalpy after the throttling process. The heating fluid exiting from the ORC boiler is divided into three fractions. The fraction going to the ORC preheater is determined by the amount of heat required. From the remaining fluid, one-fifth is used to provide heat energy to the desalination process. This heating fluid enters a heat exchanger, where its thermal energy is transferred to incoming saline water. The mass balance equations for the incoming hot fluid and the heat exchanger are as follows m_ 15 ¼ ðm_ 14 m_ 16 Þ 20%
(12.57)
m_ 15 ¼ m_ 32
(12.58)
m_ 30 ¼ m_ 31
(12.59)
The energy and exergy balances for the heat exchanger are given by m_ 15 h15 + m_ 30 h30 ¼ m_ 31 h31 + m_ 32 h32
(12.60)
_ dest, hx2 m_ 15 ex15 + m_ 30 ex30 ¼ m_ 31 ex31 + m_ 32 ex32 + Ex (12.61) The heated saline water is then passed through a throttling valve to flash it. The mass balance across the throttling valve is given by m_ 31 ¼ m_ 33
(12.62)
Moreover, because the enthalpy across a throttling valve remains constant h31 ¼ h33
(12.63)
The exergy balance across the throttling valve can be written as _ dest,ev3 m_ 31 ex31 ¼ m_ 32 ex32 + Ex
(12.64)
Separation chamber 3 is used to separate saturated vapor from the saturated liquid. The mass balance describing the separation chamber is given by m_ 33 ¼ m_ 28 + m_ 34
(12.65)
m_ 28 ¼ m_ 33 x33
(12.66)
m_ 34 ¼ m_ 33 ð1 x2 Þ
(12.67)
The energy and exergy balances can be written as m_ 33 h33 ¼ m_ 28 h28 + m_ 34 h34 _ dest,sc3 m_ 33 ex33 ¼ m_ 28 ex28 + m_ 34 ex34 + Ex
(12.68) (12.69)
The saturated vapor is then condensed by transferring its heat energy to the saline water at atmospheric temperature.
Combined cooling and power production from geothermal resources Chapter
The heat exchanger for this process can be modeled with the following necessary relations. Mass balances m_ 28 ¼ m_ 29
(12.70)
m_ 27 ¼ m_ 30
(12.71)
Energy balance m_ 28 h28 + m_ 27 h27 ¼ m_ 29 h29 + m_ 30 h30
(12.72)
Exergy balance _ dest,hx3 m_ 28 ex28 + m_ 27 ex27 ¼ m_ 29 ex29 + m_ 30 ex30 + Ex (12.73)
(12.74)
The power consumed by the pump is given by v26 ðP27 P26 Þ W_ p ¼ m_ p p
_ dest,p m_ 26 ex26 + W_ p,SW ¼ m_ 27 ex27 + Ex
12.3.5
m_ 25 ¼ m_ 37
(12.78)
m_ 39 ¼ m_ 40 + m_ 41
(12.79)
LiBr mass balance m_ 39 xLiBr 39 ¼ m_ 40 xLiBr 40 + m_ 41 xLiBr 41
(12.80)
Energy balance (12.75)
The overall energy and exergy balance for the pumping process can be written as m_ 26 h26 + W_ p,SW ¼ m_ 27 h27
195
The strong solution from the generator is throttled to the absorber pressure and flows to the absorber, where it gets mixed with the condensed water coming from the evaporator. The resulting solution has a higher concentration of water and is known as a weak solution. The weak solution is then pumped to the generator (to the same pressure as that of the condenser). To make the cycle more efficient, the weak solution is preheated before it enters the generator by exchanging heat from the strong solution leaving the generator. The equations necessary to describe the process occurring in the generator are as follows: Mass balance
The saline water at atmospheric temperature is pumped to a higher pressure, for which the mass balance can be written as m_ 26 ¼ m_ 27 ¼ m_ p,SW
12
(12.76) (12.77)
LiBr-water vapor absorption cycle
As mentioned earlier, a fraction of the geothermal fluid at state 14 is used to provide heat to drive the vapor absorption cooling (VAC) system. There are many types of absorption refrigeration systems, and their selection is dependent upon the maximum available temperature. In this case, a singleeffect absorption refrigeration system (SEARS) is chosen as the appropriate option. In the generator, heat is absorbed into the LiBr-Water solution, which results in the evaporation of some water in the solution. The resulting solution has less concentration of water and is called a strong solution. This evaporated water, used as a refrigerant, flows from the generator through the condenser where it condenses, giving out heat that is used for space heating. After the condenser, it is throttled to the evaporator pressure and then passes through the evaporator, where it absorbs heat from the surroundings resulting in space cooling. This evaporated water leaves the evaporator with 100% quality and then enters the absorber, which is also at the same pressure as that of the evaporator. In the absorber, the evaporated water condenses by rejecting the heat.
m_ 25 h25 + m_ 39 h39 ¼ m_ 37 h37 + m_ 40 h40 + m_ 41 h41
(12.81)
Exergy balance m_ 25 ex25 + m_ 39 ex39 ¼ m_ 37 ex37 + m_ 40 ex40 + m_ 41 ex41 _ dest,gen + Ex (12.82) After the generator, the refrigerant is condensed in the condenser, and the necessary balance equations to model are as below Mass balance m_ 40 ¼ m_ 46
(12.83)
m_ 40 h40 ¼ m_ 46 h46 + Q_ c
(12.84)
Energy balance
Exergy balance _ th,c + Ex _ dest,c m_ 40 ex40 ¼ m_ 46 ex46 + Ex where
(12.85)
T0 _ _ Ex th, c ¼ 1 Qc Tc
Once the refrigerant is condensed, it is passed through the throttling valve and the mass balance for the throttling valve is given by m_ 46 ¼ m_ 47
(12.86)
Moreover, because the enthalpy across a throttling valve remains constant h46 ¼ h47
(12.87)
196 PART
II Thermodynamic analysis of geothermal power plants
The exergy balance across the throttling valve can be written as
The overall energy and exergy balance for the pumping process can be written as
_ dest,ev5 m_ 46 ex46 ¼ m_ 47 ex47 + Ex
m_ 45 h45 + W_ p ¼ m_ 43 h43
(12.88)
_ dest,p m_ 45 ex45 + W_ p ¼ m_ 43 ex43 + Ex
(12.98) (12.99)
Now, the refrigerant is passed through the evaporator so that it can absorb thermal energy from the environment, providing space cooling. The following essential equations are needed to model the evaporator Mass balance
The mass balances for the solution heat exchanger (SHX) are given by m_ 43 ¼ m_ 39
(12.100)
m_ 47 ¼ m_ 48
m_ 41 ¼ m_ 42
(12.101)
(12.89)
Energy balance
Energy balance m_ 47 h47 + Q_ e ¼ m_ 48 h48
(12.90)
Exergy balance _ th, e ¼ m_ 48 ex48 + Ex _ dest, e m_ 47 ex47 + Ex where
(12.91)
Now, the refrigerant (water) is sent to the absorber, where it loses its thermal energy to the environment and mixes with the strong solution resulting in a weak solution (due to an increase in the water mass fraction). The mass balance for the absorber is given by (12.92)
_ dest,SHX m_ 41 ex41 + m_ 43 ex43 ¼ m_ 39 ex39 + m_ 42 ex42 + Ex (12.103) The strong solution coming out of the solution heat exchanger is expanded with an isenthalpic process through an expansion valve. The mass balance of the expansion process is given by m_ 42 ¼ m_ 44
h42 ¼ h44 (12.93)
The energy balance for the absorber can be written as m_ 48 h48 + m_ 44 h44 ¼ m_ 45 h45
(12.94)
where
The exergy balance of the strong solution throttling process is given by _ dest m_ 42 ex42 ¼ m_ 44 ex44 + Ex
12.4
After the absorber, the weak solution is pumped to the generator pressure and sent to a heat exchanger to recover thermal energy from a strong solution coming out of the generator. The mass balance representing the pumping process is given by (12.96)
The power consumed by the pump is given by v45 ðP43 P45 Þ W_ p ¼ m_ p p
(12.106)
(12.95)
T0 _ _ Ex th,a ¼ 1 Qa Ta
m_ 45 ¼ m_ 43 ¼ m_ p
(12.105)
_ dest,ev4 m_ 42 ex42 ¼ m_ 44 ex44 + Ex
and the exergy balance is described by _ th,a + Ex _ dest,a m_ 48 ex48 + m_ 44 ex44 ¼ m_ 45 ex45 + Ex
(12.104)
Because the enthalpy remains constant throughout the throttling process, the energy and exergy balance of the strong solution throttling process can be written as
The LiBr mass balance is represented by m_ 48 xLiBr 48 + m_ 44 xLiBr 44 ¼ m_ 45 xLiBr 45
(12.102)
Exergy balance
_ th,e ¼ 1 T0 Q_ e Ex Te
m_ 48 + m_ 44 ¼ m_ 45
m_ 41 h41 + m_ 43 h43 ¼ m_ 39 h39 + m_ 42 h42
(12.97)
Results and discussion
Using basic thermodynamic knowledge and the equations provided in Section 12.3, the whole system can be modeled using appropriate software (in this case, the Engineering Equation Solver was used). Table 12.1 shows the results of the modeled multigeneration system. Once the system has been properly modeled, the outputs of the system may be evaluated. Table 12.2 provides information regarding the essential inputs and outputs of the system. As discussed at the beginning of Section 12.3, performing only an energy analysis on the system usually does not provide the whole picture. This is especially true in the case of renewable energy systems because most of the time,
Combined cooling and power production from geothermal resources Chapter
12
197
TABLE 12.1 Evaluated thermodynamic states of the system modeled in EES. Sr. #
Ti (K)
Pi (kPa)
hi (kJ/kg)
si (kJ/kg K)
exi (kJ/kg)
0
298
101.325
104.3
0.3651
1
548.7
6000
1214
3.027
316.1
15
0
–
2
508.8
3100
1214
3.049
309.6
15
0.1102
–
3
508.8
3100
2803
6.173
968.2
1.653
1
–
4
508.8
3100
1017
2.662
228.1
13.35
0
–
5
438.1
700.477
2556
6.235
702.6
1.653
0.9
–
6
506
700.477
2917
7.035
825.3
3.307
–
–
7
495.2
3100
953.1
2.535
202.1
13.35
–
–
8
438.1
700.477
2763
6.707
768.7
1.654
1
–
9
438.1
700.477
953.1
2.576
189.9
13.35
0.1239
–
10
438.1
700.477
697.1
1.992
108
11.69
0
–
11
331.8
18.716
2370
7.219
223.7
3.307
0.9
–
12
331.8
18.716
245.4
0.8141
7.365
3.307
0
–
13
331.8
101.325
245.5
0.8141
7.45
3.307
–
–
14
383.1
700.477
461.7
1.418
43.63
11.69
–
–
15
383.1
700.477
461.7
1.418
43.63
1.754
–
–
16
383.1
700.477
461.7
1.418
43.63
2.922
–
–
17
365.9
500
167
0.5016
17.8
5.994
0
–
18
435.3
500
626.3
1.721
113.7
5.994
–
–
19
395.2
75.94
548.4
1.743
29.28
5.994
–
–
m_ i (kg/s)
xi
xLiBri –
20
300.8
75.94
4.049
0.01451
0.0105
5.994
0
–
21
298
101.325
104.3
0.3651
0
22.17
–
–
22
333.2
101.325
251.5
0.8319
8.064
22.17
–
–
23
301
500
4.812
0.01477
0.676
5.994
–
–
24
303.8
700.477
129
0.4452
0.8317
2.922
–
–
25
383.1
700.477
461.7
1.418
43.63
7.017
–
–
26
298
101.325
99.17
0.3478
0.01773
1.833
–
–
27
298
200
99.27
0.3478
0.1331
1.833
–
–
28
373.1
101.325
2676
7.355
488.4
0.02301
1
–
29
300.8
101.325
116
0.4042
0.05465
0.02301
–
–
30
306
200
131.4
0.4541
0.582
1.833
–
–
31
380.3
200
430
1.328
38.74
1.833
–
–
32
306
700.477
149.8
0.4761
12.46
1.754
–
–
33
373.6
101.325
430
1.332
37.66
1.833
0.01255
–
34
373.6
101.325
402.6
1.256
32.97
1.81
0
–
35
305.6
700.477
136.8
0.4708
1.004
4.676
–
–
36
305.8
101.325
136.8
0.4728
0.4169
4.676
–
– Continued
198 PART
II Thermodynamic analysis of geothermal power plants
TABLE 12.1 Evaluated thermodynamic states of the system modeled in EES—cont’d Sr. #
Ti (K)
Pi (kPa)
hi (kJ/kg)
si (kJ/kg K)
exi (kJ/kg)
m_ i (kg/s)
xi
xLiBri
37
366
700.477
389.3
1.226
28.32
7.017
–
–
38
366.1
101.325
389.3
1.226
28.34
7.017
–
–
39
335
4.247
137.9
0.4102
71.27
0.8401
40
332.3
4.247
2610
8.442
99.16
0.1739
–
0
41
363.2
4.247
255.2
0.4712
105.2
0.6662
–
0.6711
42
315.2
4.247
171.1
0.2228
27.2
0.6662
–
0.6711
43
303.2
4.247
71.14
0.2009
9.748
0.8401
–
0.5322
44
315.2
0.814
171.1
0.2228
27.2
0.6662
–
0.6711
45
303.2
0.814
71.14
0.2009
9.746
0.8401
–
0.5322
46
303.2
4.247
125.7
0.4365
0.08548
0.1739
–
0
47
277.2
0.814
125.7
0.4539
5.08
0.1739
0.0437
0
48
277.2
0.814
2508
9.049
184.3
0.1739
0.9998
0
the available energy is quite large but the maximum amount of work that can be extracted from it is very limited. Hence, from the energy point of view, the renewable energy system may look very inefficient, but upon performing an exergy analysis on the system, a significant improvement is usually seen. Table 12.1 also provides specific exergies of streams throughout the plant, and Table 12.3 provides information about exergy destruction in each component of the system.
TABLE 12.2 System input/output evaluations. Required output/input
Evaluated magnitude
Total enthalpy into the system
18,210 kW
Total mechanical power generated
2683 kW
Total electrical power generated
2281 kW
Total pump power required
5.076 kW
Net electric power generation
2276 kW
Space heating capacity
432 kW (122.8 tons)
Space cooling capacity
414.2 kW (117.7 tons)
Water heating capacity
3263 kW (0.508 MGD)
Desalinated water
527 GPD
Recovered heat from fresh water vapor
7.227 kW
12.4.1
0.5322
The efficiency of the system
Now that the system has been properly modeled along with the energy and exergy evaluations, the next important step is to evaluate the system efficiency. To show the benefit of designing a multigeneration system, the total power of each subsystem is taken separately and accumulated to the previous subsystem output. This methodology of efficiency evaluation will show how adding each subsystem increases the efficiency of the whole plant. Table 12.4 gives information about how adding subsystems affects the overall efficiency of the multigeneration system.
12.4.2
Varying incoming mass flow rate
One purpose of modeling the multigeneration plant on software was the ease with which the parameters can be varied and that effect examined on the whole system. The mass flow rate of geothermal fluid is varied, and the outputs examined are the mechanical work produced by turbines, the space heating and cooling, water desalination and heating, and the overall exergy destruction of the system. Fig. 12.7 shows the total output generated from the highpressure steam turbine, the medium-pressure steam turbine, and an organic Rankine cycle. It can be seen that as the mass flow rate increases, the total mechanical power generated also increases. Fig. 12.8 shows the relationship between the mass flow rate and the space heating/cooling capacity of the system. The relationship between the mass flow rate and the space heating/cooling is directly proportional.
Combined cooling and power production from geothermal resources Chapter
TABLE 12.3 Tabulation of exergy destruction in each component along with the total exergy destruction of every cycle and the whole system. Component
96.34
Separation chamber 1
0.000
Turbine 1
30.85
Heat exchanger 1
49.89
Expansion valve 2
164.0
Separation chamber 2
0
Turbine 2
180.5
Condenser 1
0.019
Pump 1
Subsystem added High-pressure power generation cycle
Cumulative energy efficiency (%) 1.905
Cumulative exergy efficiency (%) 7.317
Medium-pressure power generation cycle
10.35
39.73
Organic Rankine cycle
12.50
47.96
0.028
Thermal water desalination
15.82
60.73
ORC boiler
178.2
Absorption cycle
20.47
78.56
ORC preheater
22.43
ORC turbine
39.29
ORC condenser
3.248
Generator
402.5
_ dest, total, power gen Ex
1164
VAC generator
80.02
VAC condenser
9.887
VAC expansion valve (EV5)
0.898
VAC evaporator
0.001
VAC absorber
14.77
VAC pump
0
VAC solution EV (EV4)
0
_ dest, total, VAC Ex
105.6
Desalination pump
ffi0
Heat exchanger 2
15.24
Heat exchanger 3
10.41
Expansion valve 3
1.969
Separation chamber 3
0
_ dest, total, Desalination Ex
27.69
_ dest of the System Total Ex
1297
Fig. 12.9 shows the change in water desalination capacity with respect to a change in the mass flow rate of the incoming geothermal fluid. It can be seen that the desalination capacity has a directly proportional relationship with the mass flow rate of the incoming geothermal fluid.
199
TABLE 12.4 Cumulative First Law and Second Law efficiency of the subsystems of the multigeneration system.
Exergy destruction rate (kW)
Expansion valve 1
12
Fig. 12.10 shows the rate of exergy destruction with respect to changing the mass flow rate of the incoming geothermal fluid. The relationship shown is directly proportional, that is, as the mass flow rate is varied, the rate of exergy destruction also changes with the same ratio. Fig. 12.11 shows the relationship between the First Law efficiency of the multigeneration system with a cumulative increase in the number of systems and the mass flow rate of the incoming geothermal fluid. The energy efficiency remains constant for all values of the mass flow of the geothermal fluid. Fig. 12.12 shows the relationship between the Second Law efficiency of the multigeneration system with a cumulative increase in the number of systems and the mass flow rate of the incoming geothermal fluid. The exergy efficiency remains constant for all changes in the mass flow of the geothermal fluid. All the outputs being examined have a directly proportional relationship with the mass flow rate. This is expected because the mass flow rate is directly proportional to the available heat rate, and because the specific states (specific energy and specific exergy) do not change with varying the mass flow rate. Therefore, upon changing the mass flow rate, the parameters being observed must also increase with the same ratio, except energy and exergy efficiency, which remain the same.
12.4.3
Varying ambient temperature
Figs. 12.13–12.16 show the output variation with respect to variations in ambient temperature conditions.
200 PART
II Thermodynamic analysis of geothermal power plants
FIG. 12.7 Effect on the total turbine mechanical work output due to varying the geothermal mass flow rate.
It can be seen in Fig. 12.13 that the work output from the turbines decreases with an increase in environmental temperature. This is expected because the ORC turbine outlet pressure is dependent upon the environmental temperature. Therefore, with the increase of temperature, the turbine outlet pressure also increases, therefore decreasing the ORC turbine output. Fig. 12.14 shows that space heating and cooling capacity increases as the ambient temperature increases. This may seem counterintuitive because the performance of space cooling units usually decreases with an increase in ambient temperature, but its capacity remains the same. In this case, FIG. 12.8 Effect on the space heating/cooling output due to varying the geothermal mass flow rate.
the capacity increases because each component of the plant is parametrically linked with each other. Consider the condenser of the ORC where water at ambient temperature is heated to 60°C, and the organic fluid is cooled to near the ambient temperature. When the ambient temperature increases, the organic fluid’s condensing temperature and pressure increase, which results in a smaller heat rate required in the preheater to achieve a saturation state. Hence, the fraction of geothermal fluid exiting the ORC boiler decreases, consequently increasing the heat rate available to the vapor absorption system and the desalination plant. An increase in the thermal power available
Combined cooling and power production from geothermal resources Chapter
12
201
FIG. 12.9 Effect on water desalination and water heating capacity due to varying the geothermal mass flow rate.
2000 1800 1600 1400 1200 1000 800 600 400 200 0 0
10
20
30
40
50
60
Mass Flow Rate (kg/s) Desalination Rate (Gal/day)
Water Heating (1000 Gal/day)
to the vapor absorption cycle, more specifically the generator of the absorption cycle, increases its capacity of space heating and cooling. Fig. 12.15 shows the relationship between water desalination and water heating with respect to the ambient temperature variation. As discussed previously, the increase in ambient temperature increases the thermal power available to the desalination plant; therefore, an increase in desalination capacity is reasonable.
Moreover, because the heating temperature for the water is fixed at 60°C, as the temperature increases, the amount of heat required per kilogram of water decreases. This results in an increase in the mass flow rate of the water being heated, indicating increased capacity. As discussed earlier, exergy refers to the maximum amount of work that can be extracted. With the increase in temperature, the amount of work that can be extracted from the geothermal fluid decreases, and the geothermal FIG. 12.10 Effect on overall exergy destruction in the system due to varying the geothermal mass flow rate.
3500
Exergy Destruction (kW)
3000 2500 2000 1500 1000 500 0 0
10
20
30
40
Mass Flow Rate (kg/s) Exergy Destruction
50
60
202 PART
II Thermodynamic analysis of geothermal power plants
FIG. 12.11 Effect of varying the mass flow rate on the energy efficiency of the system.
25%
Energy Efficiency
20% 15% 10% 5% 0% 5
15
25
35
45
55
Mass Flow Rate (kg/s)
FIG. 12.12 Effect of varying the mass flow rate on the exergy efficiency of the system.
Single System
Two Systems
Four Systems
Five Systems
Three Systems
90%
Exergy Efficiency
80% 70% 60% 50% 40% 30% 20%
10% 0% 0
10
20
30
40
50
60
Mass Flow Rate (kg/s)
FIG. 12.13 Effect on the total turbine mechanical work output due to varying the geothermal environmental temperature.
Single System
Two Systems
Four Systems
Five Systems
Three Systems
2360
Turbine Work (kW)
2340 2320 2300 2280 2260 2240 2220 2200 2180 285
290
295
300
Environmental Temperature (K) Mechanical Work Out
305
310
Combined cooling and power production from geothermal resources Chapter
203
FIG. 12.14 Effect on the space heating/cooling output due to varying the geothermal environmental temperature.
440
Air Conditioning Capacity (kW)
12
435 430 425 420 415 410 285
290
295
300
305
310
Environmental Temperature (K) Space Cooling
Space Heating
FIG. 12.15 Effect on the water desalination and water heating capacity due to varying the geothermal environmental temperature.
800 700 600 500 400 300 200 100 0 285
290
295
300
305
310
Environmental Temperature (K) Desalination Rate (Gal/day)
Water Heating (1000 Gal/day)
FIG. 12.16 Effect on the overall exergy destruction in the system due to varying the geothermal environmental temperature.
890
Exergy Destruction (kW)
888 886 884 882 880 878 876 874 872
285
290
295
300
Environmental Temperature (K) Exergy Destruction
305
310
204 PART
II Thermodynamic analysis of geothermal power plants
FIG. 12.17 Effect of varying the ambient temperature on the energy efficiency of the system.
25%
Energy Efficiency
20% 15% 10% 5% 0%
285
290
295
300
305
310
Ambient Temperature (K)
fluid exiting the system is nearer to the dead state (than at lower temperatures). Hence, the exergy destruction is decreased with an increase in ambient temperature (as shown in Fig. 12.16). Fig. 12.17 shows the relationship between the ambient temperature and the energy efficiency of the system (with an increasing number of systems). An increase in ambient temperature results in a reduction in the First Law efficiency for all systems (except the single system, which is not really dependent on environmental conditions). Fig. 12.18 shows the relationship between the ambient temperature and the exergy efficiency of the system (with an increasing number of systems). An increase in ambient temperature results in an increase of the Second Law efficiency for all systems.
FIG. 12.18 Effect of varying the ambient temperature on the exergy efficiency of the system.
Single System
Two Systems
Four Systems
Five Systems
12.5
Three Systems
Closing remarks
This chapter presents a multigeneration system based on geothermal energy. The main objective was to design an optimized system that produces electricity and utilizes waste heat to produce more outputs, hence increasing the overall efficiency of the system. The system is designed so it receives incoming geothermal fluid at the pressure of 60 bar and in a saturated liquid state with a mass flow rate of 15 kg/s. The design shows the utilization of the three power generation cycles to produce approximately 2.3 MW of electrical power. The organic Rankine cycle’s condenser is used to utilize the latent heat of the organic fluid and to heat water at ambient temperature to 60°C. The waste heat is further utilized to drive the lithium bromide vapor absorption cycle with a space cooling
90%
Exergy Efficiency
80% 70% 60% 50% 40% 30% 20% 10% 0% 285
290
295
300
305
Ambient Temperature (K) Single System
Two Systems
Four Systems
Five Systems
Three Systems
310
Combined cooling and power production from geothermal resources Chapter
capacity of 117 tons of refrigeration and a space heating capacity of 122 tons of heating. Furthermore, the waste heat was also utilized to power a thermal flash desalination process with a desalination rate of 527 gal per day. In later sections, discussion regarding how multigeneration systems effectively increased the efficiency of the plant was presented. The energy and exergy efficiencies increased significantly with an increase in the number of subsystems installed to capture and utilize waste heat. Finally, it was shown how the variation in input parameters affect the output of the system. It was shown that the outputs of the system are directly proportional to the change in the mass flow rate of the incoming geothermal fluid while the energy and exergy efficiencies of the system remain unaffected by the change in the mass flow rate of the geothermal fluid. On the other hand, an increase in ambient temperature results in a decrease of total electricity generated and the total exergy destruction, but the space cooling/heating as well as the water desalination and heating capacity increase with an increase in ambient temperature. The exergy efficiency of the system was also shown to increase with an increase in ambient temperature while the energy efficiency of the system slightly decreased with an increase in the ambient temperature.
References [1] Landes DS. The unbound prometheus. Cambridge University Press; 1969. [2] Field CB, Barros VR, Mach KJ, Mastrandrea MD, van Aalst M, Adger WN, et al. Technical summary climate change 2014: impacts, adaptation, and vulnerability. Part A: global and sectoral aspects. In: Contribution of Working Group II to the fifth assessment report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press; 2014. [3] Climate change: Arctic sea ice summer minimumjNOAA Climate. gov n.d. https://www.climate.gov/news-features/understandingclimate/climate-change-minimum-arctic-sea-ice-extent Accessed 13 February 2020. [4] Arctic sea ice minimumjvital signs—climate change: vital signs of the planet n.d. https://climate.nasa.gov/vital-signs/arctic-sea-ice/ Accessed 13 February 2020. [5] The Arctic is melting much faster than Antarctic. That impacts all of us. n.d. https://www.nationalgeographic.com/science/2019/12/arctic/ (Accessed 13 February 2020).
12
205
[6] Glaciers and climate change jNational Snow and Ice Data Center n.d. https://nsidc.org/cryosphere/glaciers/questions/climate.html [Accessed 13 February 2020]. [7] Graphic: dramatic glacier melt—climate change: vital signs of the planet n.d. https://climate.nasa.gov/climate_resources/4/graphic-dra matic-glacier-melt/ [Accessed 13 February 2020]. [8] Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, et al. Summary for policymakers. In: Climate change 2013: the physical science basis. Contribution of Working Group I to the fifth assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press; 2013. [9] US EPA. Inventory of U.S. greenhouse gas emissions and sinks: 1990–2017; 2019. [10] Edenhofer O, Pichs-Madruga R, Sokona Y. Summary for Policymakers. IPCC 2014: Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. New York, NY: Cambridge University Press; 2014. p. 9. [11] Fridleifsson IB. Geothermal energy for the benefit of the people. Renew Sust Energy Rev 2001;5:299–312. https://doi.org/10.1016/ S1364-0321(01)00002-8. [12] Barbier E. Nature and technology of geothermal energy: a review. Renew Sust Energy Rev 1997;1:1–69. https://doi.org/10.1016/ S1364-0321(97)00001-4. [13] Ammar Y, Joyce S, Norman R, Wang Y, Roskilly T. Low grade thermal energy sources and uses from the process industry in the UK. Appl Energy 2012;89:3–20. https://doi.org/10.1016/j. apenergy.2011.06.003. [14] Christ A, Wang X, Regenauer-Lieb K, Chua H. Low-grade waste heat driven desalination technology. Int J Simul Multidiscip Des Optim 2014;5:A02. https://doi.org/10.1051/smdo/2013007. [15] MIT. Thermophysical properties of seawater 2017, http://web.mit. edu/seawater/. [Accessed 20 January 2020]. [16] Sharqawy MH, Lienhard VJH, Zubair SM. Thermophysical properties of seawater: a review of existing correlations and data. Desalin Water Treat 2010;16:354–80. https://doi.org/10.5004/ dwt.2010.1079. [17] Leyendekkers JV. Prediction of the heat capacities of seawater and other multicomponent solutions from the Tammann-Tait-Gibson model. Mar Chem 1980;9:25–35. https://doi.org/10.1016/0304-4203 (80)90004-3. [18] Nayar KG, Sharqawy MH, Banchik LD, Lienhard JH. Thermophysical properties of seawater: a review and new correlations that include pressure dependence. Desalination 2016;390:1–24. https:// doi.org/10.1016/j.desal.2016.02.024. [19] Wagner W, Pruß A. The IAPWS formulation 1995 for the thermodynamic properties of ordinary water substance for general and scientific use; 1996. p. 384–535.
Chapter 13
Hydrogen production from geothermal power plants Murat Ozturka and Ibrahim Dincerb a
Faculty of Technology, Department of Mechatronics Engineering, Isparta University of Applied Sciences, Isparta, Turkey,
b
Clean Energy Research Laboratory, Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON, Canada
Nomenclature E E_ ex _ Ex _ D Ex h m_ P Q_ s T W_
energy (kJ) energy rate (kW) specific exergy (kJ/kg) exergy rate (kW) exergy destruction rate (kW) specific enthalpy (kJ/kg) mass flow rate (kg/s) pressure (kPa) heat transfer rate (kW) specific entropy (kJ/kgK) temperature (°C) power (kW)
Greek letters
h c
energy efficiency exergy efficiency
Abbreviations COP DFGP HEX IS ORC PEM RO SEACE
coefficient of performance double-flash geothermal plant heat exchanger integrated system organic Rankine cycle proton exchange membrane reverse osmosis single-effect absorption cooling with ejector
13.1 Introduction During the past several decades, the need for clean power production has increased greatly. Therefore, it becomes important to design energy generation systems using clean energy sources. There are energy resources such as solar, geothermal, and wind energy that can be used as clean power in nature. With these energy resources, energy production systems, especially those producing hydrogen, have a privileged feature because hydrogen has great potential in
terms of both environmental friendliness and power obtained from the unit mass. Many studies have been carried out on integrated systems that produce beneficial outputs, including hydrogen, by utilizing clean energy sources, and studies on this subject are increasing. In particular, there are studies that provide many useful outputs, including hydrogen, using a geothermal energy source. Li et al. [1] analyzed an energy system integrated with geothermal power and proton exchange membrane (PEM) fuel cells. The analyzed system produces many useful outputs besides hydrogen. Before a thermodynamic analysis was done for this plant, a mathematical model of the integrated process is created with thermodynamic equations. After integrating the PEM fuel cell to the plant, the exergy efficiency of the whole plant increased 18.68%. Additionally, the optimization of the integrated cycle is carried out. After optimization of the system, the net output work is calculated as 1571.1 kW. Yuksel et al. [2] designed a novel integrated plant that can generate many useful products besides hydrogen. A thermodynamic analysis of the new process design is carried out. According to the thermodynamic analysis results, the exergy efficiency of the whole plant was found to be 52.63%. A parametric work is carried out to analyze the effect of some selected criteria that affect system effectiveness from the process operating parameters on the system. According to the parametric analysis, the most important parameter affecting system performance is the geothermal fluid temperature. As the geothermal source temperature varies from 130°C to 200°C, the exergy efficiency of the system increases from 38% to 64%. In the reference work, the thermodynamic and effectiveness assessment results of the multigeneration plant are given in detail. Yuksel and Ozturk [3] performed an effectiveness analysis of a combined energy generation system that utilized the geothermal energy source. In this study, the EES (Engineering Equation Solver) software package is utilized while performing the effectiveness analysis of the integrated process. The parametric works
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00003-2 Copyright © 2021 Elsevier Inc. All rights reserved.
207
208
PART
II Thermodynamic analysis of geothermal power plants
are carried out to see how the cycle works under different operating criteria. The energetic and exergetic effectiveness values of the combined process are computed as 47.04% and 32.15%. The cost analysis indicates that after increasing the geothermal water temperature to 200°C, the hydrogen production cost decreases from 4.8 to 1.1 $/kg H2. Chen et al. [4] designed a combined process integrated with geothermal energy and a PEM fuel cell to meet heating and cooling demands and to provide power output. In this work, geothermal energy is utilized as clean power. A thermodynamic model is developed to examine the impact of some critical operating conditions on the cycle effectiveness. The energetic effectiveness of the proposed new system is computed as 66.3%. With this system, many useful outputs are produced besides hydrogen. The study also shows that the proposed integrated process is an efficient and effective energy generation system by using the geothermal energy source. According to the cost analysis, the annual cost savings have reached 20.9%. Waseem et al. [5] conducted a study on the comparison of geothermal- and solar power-based integrated systems. Both these systems can produce many useful outputs, including hydrogen. They show that the selected systems differ in both performance and beneficial outputs. They performed thermodynamic analyses of both the integrated systems and their subsystems. They indicate that certain parameters have an impact on the effectiveness of the integrated plants. They also performed comparative analyses between these two systems. It is therefore mentioned that hydrogen production has an important potential for these systems. Also, this study compares the efficiencies of different systems under various working conditions. Yuksel et al. [6] designed a novel multigenerational plant that can produce different useful outputs, especially with hydrogen. Geothermal power is used as a clean and renewable energy resource in this proposed integrated system. A comprehensive mathematical model is prepared for the thermodynamic assessment of the integrated plant. The parametric works are carried out to examine the impact of some working criteria that affect the system effectiveness on system performance. In the performance assessment of the integrated plant, the energy and exergy effectiveness values of the process are computed as 39.46% and 44.27%, respectively. It is stated that the temperature increment in the geothermal energy resource contributes positively to the effectiveness of the system. Khalid et al. [7] investigated an integrated process consisting of solar and geothermal energy subsystems that can produce different useful outputs besides hydrogen. The effectiveness of the integrated process has been evaluated using both energetic and exergetic efficiencies. The results demonstrate that an electrolyzer can produce 2.7 kg/h hydrogen. In addition to that, a parametric study is carried out to see how some operating indicators affect the performance of
the integrated process. Also, the cost analysis of the integrated process has been carried out. An attempt to optimize the system was done by using a software program. In this study, the optimum values obtained in terms of cost with optimization are given. The levelized cost of electricity has been calculated as 0.089 $/kWh. Al-Hamed and Dincer [8] proposed an integrated system with a cooling cycle that uses both solar and geothermal energy as power resources. At the same time, they performed the analysis of the system they proposed by taking the thermodynamic equations as reference. The system they designed produces multiple useful outputs for small energy needs supported and calculated by the comprehensive performance analysis. The results show that the overall energetic and exergetic efficiencies of the system are 53.33% and 37.07, respectively. They carried out a parametric study to demonstrate how some selected parameters affect the efficiency of the process. At the same time, they improved the process efficiency by optimizing the system. Al-Ali and Dincer [9] proposed an integrated system that utilizes solar and geothermal power for multiple useful outputs. They conducted energetic and exergetic assessments of the process to demonstrate the system effectiveness in different production modes. In that study, the largest exergy destruction ratio belongs to the solar collector with 75%. The comparison of the energy efficiencies of the single generation and the multigeneration system indicates that the efficiency of the single generation is 16.4% while the efficiency of the multigeneration system is 78%. They used a parametric analysis to examine the impact of different criteria that have an impact on the performance of the process. Mosaffa and Zareei [10] proposed a system to increase the performance of the fluid from the geothermal power source. They tried to develop their proposed system using two different methods. They have presented the comprehensive thermodynamic assessment to analyze the impact of some parameters that affect performance on the process. They gave the outputs of two selected methods comparatively. They also conducted an optimization work to find the values of the working criteria in which the process operates most efficiently. Wan et al. [11] designed a power generation system that uses both geothermal and solar energy sources to enhance system efficiency. Also, they performed a thermodynamic analysis of the process with the determined operating parameter values. Thanks to the performance and parametric analyses of the system, they obtained alternative ways to improve system performance. The net power output of the system has been found as 12.76 MW and the energy and exergy efficiencies of the whole system are calculated as 10.74% and 23.9%, respectively. Nami et al. [12] analyzed a thermodynamical assessment of a novel integrated plant powered by geothermal power according to the energetic and exergetic approaches. The environmental impact
Hydrogen production from geothermal power plants Chapter 13
evaluation indicates that the designed system is an environmentally friendly one. They have also given the performance outputs of the combined process in detail. Also, geothermal energy is considered as the available lowtemperature heat source, and this resource can be used for hydrogen production [13]. Siddiqui et al. [14] designed a new combined plant that uses solar and geothermal power. The energetic performance of the trigeneration plant is computed to be 19.6% while the exergetic performance is calculated to be 19.1%. Furthermore, the hydrogen production capacity of the whole system is 32.1 mol/s. Also, they added a parametric analysis to this work to show the impact of some study indicators on the process efficiency. In this chapter, hydrogen generation options based on the geothermal energy source are investigated. Also, the current chapter aims to give a performance comparison of a geothermal power-based combined plant for multigeneration with hydrogen generation, which is very significant for countries with high-grade geothermal sources [15]. Furthermore, this chapter aims to model a combined geothermal-based plant thermodynamically to achieve the highest possible energetic and exergetic effectiveness for the whole plant by altering some key thermodynamic indicators. In this regard, some specific objectives of this particular chapter are: (i) to develop a comprehensive thermodynamic model for a combined multigeneration plant with hydrogen production, (ii) to determine the entropy generation of each component, (iii) to calculate the exergy destruction, exergy loss, and exergetic performance values of each component, and (iv) to analyze the impact of the dead-state temperature, temperature, and mass flow rate of the geothermal working fluid on the overall system performance.
13.2 Geothermal hydrogen production Geothermal source application can be divided into two groups: (i) heating utilization, and (ii) electrical energy generation, according to its reservoir temperature level. The geothermal source application ways are illustrated in Fig. 13.1 based on heating applications and power
209
production. The direct geothermal source utilization ways are solely according to the application of thermal power without changing the condition of geothermal power. Direct geothermal source utilization techniques are among the earliest and most prevalent methods of using geothermal power. Electricity utilization needs a power transformation process, simply to transform the thermal power into electricity. On the other hand, nowadays the numerous applications for geothermal energy worldwide present a cogeneration of power and heat energy. Six process designs can be used for geothermal energy-based electricity production: (i) dry steam, (ii) single-flash, (iii) double-flash, (iv) triple-flash, (v) binary cycle, and (vi) integrated/hybrid power generation systems. All the flash-based plants are designed on the flashing of the geothermal water for vapor production, where double- and triple-flash designs use the unutilized geothermal water further to increase the energetic and exergetic efficiencies and electricity generation. In the near future, the geothermal power-based hydrogen production potential will increase for regions with abundant geothermal energy reservoirs [16]. Geothermal power can be utilized for hydrogen generation mostly in four methods: (i) direct production, (ii) water electrolysis-based production, (iii) thermal energy-based thermochemical production, and (iv) hybrid thermochemical production, as illustrated in Fig. 13.2. The geothermal water vapor ejected to the surroundings usually contains hydrogen. This vapor generally contains different hydrogen concentrations based on the geothermal source reservoir. If the hydrogen gases that are released to the surroundings are analyzed, it can be seen that there are strictly recoverable concentrations and proportions. The hydrogen cleaning process must be applied before any use of this directly produced hydrogen in the fuel cell systems. The second option for geothermal energy-based hydrogen generation is the water electrolysis plant (electrolysis at low (40–90°C) and high (700–900°C) temperatures). As given in Fig. 13.1, in this type of plant, the
Geothermal sources
Geothermal source application Thermal Energy
Heating Applications
Mechanical work
Electrical Energy
Power productions
Geothermal heat pumps;
Dry steam power generation;
Space and district heating;
Single-flash power generation;
Greenhouse heating;
Double-flash power generation;
Aquacultural heating;
Triple-flash power generation;
Agricultural drying;
Binary cycle power generation;
Industrial process heating;
Integrated/hybrid power generation systems.
Bathing and swimming; and Others.
FIG. 13.1 Schematic diagram of the geothermal source application.
Thermal energy based thermochemical cycles
Hydrogen
Hybrid thermochemical cycles
Electrolysis
Hydrogen
Hydrogen
Direct production
Hydrogen
Hydrogen storage
FIG. 13.2 Schematic diagram of geothermal energy-based hydrogen production options.
210
PART
II Thermodynamic analysis of geothermal power plants
TABLE 13.1 Low-temperature thermochemical hydrogen production cycles. Cycle
Temperature range (°C)
Chemical reactions
Sources
Cu-Cl
430–475
2Cu(s) + 2HCl(g) ! 2CuCl(l) + H2(g)
[17]
Ambient (electrolysis)
2CuCl(s) ! 2CuCl(aq) ! CuCl2(aq) + Cu(s)
>100
CuCl2(aq) + CuCl2(s)
400–450
2CuCl2(s) + H2O(g) ! CuO ∗ CuCl2(s) + 2HCl(g)
450–500
CuO ∗ CuCl2(s) ! 2CuCl(l) + 1/2O2(g)
450–500
MgCl2(s) + H2O(g) ! 2HCl(s) + MgO(g)
450–500
MgO(s) + Cl2(g) ! MgCl2(s) + 1/2O2(g)
80–100
2HCl(s) ! Cl2(s) + H2(g)
80–90
2H2O + SO2 ! H2SO4 + H2
450–500
H2SO4 ! H2O + SO3
500–550
SO3 ! SO2 + 1/2O2
Mg-Cl
Sulfuric acid
[18]
[19]
geothermal energy is converted into mechanical energy to generate electrical energy. The produced electrical power is then utilized in the electrolysis plant to split H2O into H2 and O2. The general water electrolysis reaction should be defined as follows, (13.1) H2 O + electricity ! H2 + 1 2 O2
Different low-temperature thermochemical cycles are summarized in Table 13.1. The temperature ranges of the hightemperature geothermal sources are usually between 200°C and 250°C; therefore, the working fluid temperatures in these cycles should be increased to the desired values, that is, 550–600°C, by integrating various heat pump options.
Before entering the electrolysis plant, the freshwater temperature must be increased to the required value by using the geothermal energy through a heat exchanger. The hydrogen generation option via electrolysis has numerous benefits, such as efficiently pure hydrogen generation, an environmentally benign nature, and high efficiency due to varying voltages. The primary disadvantage of this option is the need for a higher demand for electrical energy. The third option for geothermal energy-based hydrogen generation is the thermal energy-based thermochemical plants driven by geothermal heat power. Thermochemical hydrogen generation plants contain a thermal dissociation of H2O into H2 and O2 by using a series of thermally operated chemical processes. The goal is to generate hydrogen at lower temperatures than that for the water pyrolysis mechanism, which occurs at high temperatures of 2500–3000°C. Finally, the geothermal energy integrated hydrogen production option is the hybrid thermochemical production cycle. Based on the thermodynamic point of view, the thermochemical hydrogen production reactions require combinations of power and heat in some steps of the cycle. The working temperature of the thermal energy-based or hybrid thermochemical cycles containing geothermal energybased hydrogen production plants must be below 550°C.
13.3
Case study
In this section, a case study has been carried out to illustrate the thermodynamic modeling of the geothermal power-based hydrogen production method. The case study also covers detailed energetic and exergetic efficiency assessments.
13.3.1 System description Here, the integrated plant is described and its schematic representation is also shown. For this system, the dryer and hot water production subplants are integrated into the geothermal plant utilizing low-grade geothermal energy. The aim is to select a new multigeneration power plant to generate power, produce hydrogen, and meet the demands of heating, cooling, hot freshwater, and drying. The schematic of a double-flash geothermal plant-based multigeneration system with hydrogen generation is given in Fig. 13.3. This combined plant consists of six subplants, such as the double-flash geothermal plant, hydrogen production, freshwater production, single-effect absorption cooling with ejector (SEACE), dryer process, and hot water production. The water coming from the hot water well comes to valve 1 with flow 1. The saturated water mixture is obtained with valve 1. Then, this saturated water mixture
Hydrogen production from geothermal power plants Chapter 13
PEM electrolyzer
3
Double-flash geothermal plant
Separator 1 Valve 1 1
Turbine 1
7
5
10
Production well
Hydrogen production
29 Oxygen
HEX 1
Valve 2 6
Oxygen separator
28
3 way 17 valve 3 25
4
Electricity
Turbine 2
Separator 2
Hydrogen storage tank
30
27
Filter
9 2
26
Sea water Electricity 16
211
24
18
3 way-valve 2
Fresh water production
Pump 1 8
11
3 way-valve 1
19 12
54
15
Hot water storage tank
53 Injection well
Hot water
Cold water
Hot water production
51 14
52
Fresh water 23 storage tank Fresh
water Electricity
Brine reject 21
Wet product
45
13
Dryer
RO 22 module 20 Pelton turbine
Generator
40
Dry product
31
39 Pump 2 38
SEAC with 36 ejector
42
34
43
Absorber 50
33
Valve 3
Valve 4
49
46 Condenser
Ejector
41
HEX 2
Dryer process
32
37
35 3-way valve 4
Evaporator 47 48
FIG. 13.3 Schematic diagram of the integrated system based on a double-flash geothermal plant.
is transferred to separator 1 by stream 2. Thanks to separator 1, this saturated mixture is separated as liquid water and steam. The part given as steam from separator 1 is transmitted to turbine 1 to produce electricity by stream 3. In order to preheat another water flow with the heat coming out of turbine 1, the heat coming out of turbine 1 is transmitted to heat exchanger (HEX) 1 by stream 4. The liquid water obtained from separator 1 is delivered to valve 2 with flow number 6. Thanks to valve 2, a saturated water mixture is obtained from this liquid water. The saturated water mixture is sent to separator 2 with flow 7. The saturated mixture is then separated into two parts as liquid and steam in separator 2. The steam is directed to turbine 2 with stream 9 to obtain more electrical output. The liquid water from separator 2 and fluid from turbine 2 are then transmitted to three-way valve 1 with flow 8 and flow 10, respectively. The fluid is then transferred to three-way valve 2 by flow 11. The water flow from the generator for the drying process is transmitted to the dryer with flow 13. Wet product enters the dryer with flow 51 is converted into a dry product with flow 52. The fluid coming from the hot water storage tank (HWST) with stream 14 is reinjected into the well again with flow 15. The cold water supplied to the HWST with flow 53 is heated and pumped as the hot water with flow 54. For freshwater production and hydrogen
generation, seawater is supplied to the cycle with flow 16. The seawater that is supplied to the cycle first undergoes a filtration process, and then it is transmitted to HEX 1 via flow 17. The fluid from HEX 1 is sent to the highpressure pump (pump 1) with flow 18. Pump 1 sends this fluid at high pressure to the reverse osmosis (RO) module. The freshwater obtained from this module is then transmitted to the freshwater storage tank with point 22. The water coming out of the module as brine is transmitted to a Pelton turbine with flow 20. The electrical output is obtained with the Pelton turbine. The waste part is sent from the turbine with point 21 to the reservoir. Some freshwater required for hydrogen production is directed to HEX 1 from a freshwater storage tank with point 24. The fluid adjusted temperature for electrolysis is transmitted to three-way valve 3 with point 25. Fluid entering the PEM electrolyzer with stream 26 is separated into hydrogen and oxygen. The generated hydrogen is then transferred to the hydrogen storage tank with point 30. The cooling cycle starts at the generator. Ammonia and water are used in the cooling cycle. The single effect absorption cooling with ejector (SEACE) starting in the generator is completed with flow 50. The working conditions and design criteria of the integrated system are defined in Tables 13.2 and 13.3, respectively.
TABLE 13.2 The balance equations for the components of the integrated system. For valve 1 Mass
m_ 1 ¼ m_ 2
Energy
m_ 1 h1 ¼ m_ 2 h2
Entropy
m_ 1 s1 + S_ g , Val1 ¼ m_ 2 s2
Exergy
m_ 1 ex 1 ¼ m_ 2 ex 2 + E_ xD, Val1
Energy efficiency
_ 2 h2 Val1 ¼ m m_ 1 h1
Exergy efficiency
_ 2 ex 2 cVal1 ¼ m m_ 1 ex 1
For separator 1 Mass
m_ 2 ¼ m_ 3 + m_ 6
Energy
m_ 2 h2 ¼ m_ 3 h3 + m_ 6 h6
Entropy
m_ 2 s2 + S_ g , Sep1 ¼ m_ 3 s3 + m_ 6 h6
Exergy
m_ 2 ex 2 ¼ m_ 3 ex 3 + m_ 6 h6 + E_ xD, Sep1
Energy efficiency
Sep1 ¼ m_ 3 hm3_ +2 hm2_ 6 h6
Exergy efficiency
cSep1 ¼ m_ 3 exm_32+exm_26 ex 6
For turbine 1 Mass
m_ 3 ¼ m_ 4
Energy
m_ 3 h3 ¼ m_ 4 h4 + W_ T 1
Entropy
m_ 3 s3 + S_ g , T _1 ¼ m_ 4 s4
Exergy
m_ 3 ex 3 ¼ m_ 4 ex 4 + E_ xTW1 + E_ xD, T 1 _
Energy efficiency
T1 T 1 ¼ m_ 3 hW _ 4 h4 3 m
Exergy efficiency
cT 1 ¼ m_ 3 ex 3 Tm1_ 4 ex 4
E_ x W
For three-way valve 1 Mass
m_ 8 + m_ 10 ¼ m_ 11
Energy
m_ 8 h8 + m_ 10 h10 ¼ m_ 11 h11
Entropy
m_ 8 s8 + m_ 10 s10 + S_ g , 3wv1 ¼ m_ 11 s11
Exergy
m_ 8 ex 8 + m_ 10 ex 10 ¼ m_ 11 ex 11 + E_ xD, 3wv1
Energy efficiency
3wv1 ¼ m_ 8 hm8_ +11mh_ 1110 h10
TABLE 13.2 The balance equations for the components of the integrated system.—cont’d Exergy efficiency
m_ 11 ex 11 c3wv1 ¼ m_ 8 ex _ 10 h10 8 +m
For filter Mass
m_ 16 ¼ m_ 17
Energy
m_ 16 h16 ¼ m_ 17 h17
Entropy
m_ 16 s16 + S_ g , Ft ¼ m_ 17 s17
Exergy
m_ 16 ex 16 ¼ m_ 17 ex 17 + E_ xD, Ft
Energy efficiency
_ 17 h17 Ft ¼ m m_ 16 h16
Exergy efficiency
_ 17 ex 17 cFt ¼ m m_ 16 ex 16
For HEX 1 Mass
m_ 4 ¼ m_ 5 ; m_ 17 ¼ m_ 18 , m_ 24 ¼ m_ 25
Energy
m_ 4 h4 + m_ 17 h17 + m_ 24 h24 ¼ m_ 5 h5 + m_ 18 h18 + m_ 25 h25
Entropy
m_ 4 s4 + m_ 17 s17 + m_ 24 s24 + S_ g , HEX1 ¼ m_ 5 s5 + m_ 18 s18 + m_ 25 s25
Exergy
m_ 4 ex 4 + m_ 17 ex 17 + m_ 24 ex 24 ¼ m_ 5 ex 5 + m_ 18 ex 18 + m_ 25 ex 25 + E_ xD, HEX1
Energy efficiency
ðm 25 h25 m 24 h24 Þ HEX1 ¼ ðm 18 h18 mð17m_h417h4Þ + m _ 5 h5 Þ
Exergy efficiency
17 Þ + ðm 25 ex 25 m 24 ex 24 Þ cHEX1 ¼ ðm 18 ex 18 mð17m_ex4 ex _ 5 ex 5 Þ 4 m
_
_
_
_
_
_
_
_
Pump 1 Mass
m_ 18 ¼ m_ 19
Energy
m_ 18 h18 + W_ P1 ¼ m_ 19 h19
Entropy
m_ 18 s18 + S_ g , P1 ¼ m_ 19 s19
Exergy
W ¼ m_ 19 ex 19 + E_ xD, P1 m_ 18 ex 18 + E_ xP1
Energy efficiency
P1 ¼ m_ 19 h19W_m_ 18 h18
Exergy efficiency
cP1 ¼ m_ 19 ex 19E_ xWm_ 18 ex 18 P1
P1
For RO module Mass
m_ 19 ¼ m_ 20 + m_ 22
Energy
m_ 19 h19 ¼ m_ 20 h20 + m_ 22 h22
Entropy
m_ 19 s19 + S_ g , ROm ¼ m_ 20 s20 + m_ 22 s22 Continued
TABLE 13.2 The balance equations for the components of the integrated system.—cont’d Exergy
m_ 19 ex 19 ¼ m_ 20 ex 20 + m_ 22 ex 22 + E_ xD, ROm
Energy efficiency
+ m_ 22 h22 ROm ¼ m_ 20 hm20 _ 19 h19
Exergy efficiency
cROm ¼ m_ 20 exm_2019+exm_1922 ex 22
For fresh water storage tank Mass
m_ 22 ¼ m_ 23 + m_ 24
Energy
m_ 22 h22 ¼ m_ 23 h23 + m_ 24 h24
Entropy
m_ 22 s22 + S_ g , FWST ¼ m_ 23 s23 + m_ 24 s24
Exergy
m_ 22 ex 22 ¼ m_ 23 ex 23 + m_ 24 ex 24 + E_ xD, FWST
Energy efficiency
+ m_ 24 h24 FWST ¼ m_ 23 hm23 _ 22 h22
Exergy efficiency
cFWST ¼ m_ 23 exm_2322+exm_2224 ex 24
For PEM electrolyzer Mass
m_ 26 ¼ m_ 27 + m_ 30
Energy
m_ 26 h26 + W_ PEM ¼ m_ 27 h27 + m_ 30 h30
Entropy
m_ 26 s26 + S_ g , PEM ¼ m_ 27 s27 + m_ 30 s30
Exergy
W ¼ m_ 27 ex 27 + m_ 30 ex 30 + E_ xD, PEM m_ 26 ex 26 + E_ xPEM
Energy efficiency
_ 27 h27 + m_ 30 h30 PEM ¼ mm _ h + W_
Exergy efficiency
cPEM ¼ m_m_27 exex27 ++m_E_30x Wex 30 26 26 PEM
26 26
PEM
For oxygen separator Mass
m_ 27 ¼ m_ 28 + m_ 29
Energy
m_ 27 h27 ¼ m_ 28 h28 + m_ 29 h29
Entropy
m_ 27 s27 + S_ g , OS ¼ m_ 28 s28 + m_ 29 s29
Exergy
m_ 27 ex 27 ¼ m_ 28 ex 28 + m_ 29 ex 29 + E_ xD, OS
Energy efficiency
+ m_ 29 h29 OS ¼ m_ 28 hm28 _ 27 h27
Exergy efficiency
cOS ¼ m_ 28 exm_2827+exm_2729 ex 29
For generator Mass
m_ 12 ¼ m_ 13 ; m_ 40 ¼ m_ 31 + m_ 41
Energy
m_ 12 h12 + m_ 40 h40 ¼ m_ 13 h13 + m_ 31 h31 + m_ 41 h41
TABLE 13.2 The balance equations for the components of the integrated system.—cont’d Entropy
m_ 12 s12 + m_ 40 s40 + S_ g , Gen ¼ m_ 13 s13 + m_ 31 s31 + m_ 41 s41
Exergy
m_ 12 ex 12 + m_ 40 ex 40 ¼ m_ 13 ex 13 + m_ 31 ex 31 + m_ 41 ex 41 + E_ xD, Gen
Energy efficiency
+ m 41 h41 m 40 h40 Þ Gen ¼ ðm 31ðhm31 _ 12 h12 m_ 13 h13 Þ
Exergy efficiency
31 + m 41 ex 41 m 40 ex 40 Þ cGen ¼ ðm 31 ex ðm_ 12 ex 12 m_ 13 ex 13 Þ
_
_
_
_
_
_
For ejector Mass
m_ 31 + m_ 36 ¼ m_ 32
Energy
m_ 31 h31 + m_ 36 h36 ¼ m_ 32 h32
Entropy
m_ 31 s31 + m_ 36 s36 + S_ g , Ejc ¼ m_ 32 s32
Exergy
m_ 31 ex 31 + m_ 36 ex 36 ¼ m_ 32 ex 32 + E_ xD, Ejc
Energy efficiency
_ 32 h32 Ejc ¼ m_ 31 hm31 + m_ 36 h36
Exergy efficiency
cEjc ¼ m_ 31 exm_3132+exm_3236 ex 36
For condenser Mass
m_ 32 ¼ m_ 33 ; m_ 45 ¼ m_ 46
Energy
m_ 32 h32 + m_ 45 h45 ¼ m_ 33 h33 + m_ 46 h46
Entropy
m_ 32 s32 + m_ 45 s45 + S_ g , Con ¼ m_ 33 s33 + m_ 46 s46
Exergy
m_ 32 ex 32 + m_ 45 ex 45 ¼ m_ 33 ex 33 + m_ 46 ex 46 + E_ xD, Con
Energy efficiency
m 46 h46 m 45 h45 Þ Con ¼ ððm _ 32 h32 m_ 33 h33 Þ
Exergy efficiency
m 46 ex 46 m 45 ex 45 Þ cCon ¼ ððm _ 32 ex 32 m_ 33 ex 33 Þ
_
_
_
_
For evaporator Mass
m_ 34 ¼ m_ 35 ; m_ 47 ¼ m_ 48
Energy
m_ 34 h34 + m_ 47 h47 ¼ m_ 35 h35 + m_ 48 h48
Entropy
m_ 34 s34 + m_ 47 s47 + S_ g , Eva ¼ m_ 35 s35 + m_ 48 s48
Exergy
m_ 34 ex 34 + m_ 47 ex 47 ¼ m_ 35 ex 35 + m_ 48 ex 48 + E_ xD, Eva
Energy efficiency
m 48 h48 m 47 h47 Þ Eva ¼ ððm _ 34 h34 m_ 35 h35 Þ
Exergy efficiency
m 48 ex 48 m 47 ex 47 Þ cEva ¼ ððm _ 34 ex 34 m_ 35 ex 35 Þ
_
_
_
_
Continued
TABLE 13.2 The balance equations for the components of the integrated system.—cont’d For absorber Mass
m_ 49 ¼ m_ 50 ; m_ 37 + m_ 43 ¼ m_ 38
Energy
m_ 37 h37 + m_ 43 h43 + m_ 49 h49 ¼ m_ 38 h38 + m_ 50 h50
Entropy
m_ 37 s37 + m_ 43 s43 + m_ 49 s49 + S_ g , Abs ¼ m_ 38 s38 + m_ 50 s50
Exergy
m_ 37 ex 37 + m_ 43 ex 43 + m_ 49 ex 49 ¼ m_ 38 ex 38 + m_ 50 ex 50 + E_ xD, Abs
Energy efficiency
m 49 h49 Þ Abs ¼ ðm_ 37ðhm3750+hm50 _ 43 h43 m_ 38 h38 Þ
Exergy efficiency
ðm 50 ex 50 m 49 ex 49 Þ cAbs ¼ ðm_ 37 ex _ 43 ex 43 m_ 38 ex 38 Þ 37 + m
_
_
_
_
For dryer Mass
m_ 13 ¼ m_ 14 ; m_ 51 ¼ m_ 52
Energy
m_ 13 h13 + m_ 51 h51 ¼ m_ 14 h14 + m_ 52 h52
Entropy
m_ 13 s13 + m_ 51 s51 + S_ g , Dry ¼ m_ 14 s14 + m_ 52 s52
Exergy
m_ 13 ex 13 + m_ 51 ex 51 ¼ m_ 14 ex 14 + m_ 52 ex 52 + E_ xD, Dry
Energy efficiency
m 52 h52 m 51 h51 Þ Dry ¼ ððm _ 13 h13 m_ 14 h14 Þ
Exergy efficiency
m 52 ex 52 m 51 ex 51 Þ cDry ¼ ððm _ 13 ex 13 m_ 14 ex 14 Þ
_
_
_
_
For hot water storage tank Mass
m_ 14 ¼ m_ 15 ; m_ 53 ¼ m_ 54
Energy
m_ 14 h14 + m_ 53 h53 ¼ m_ 15 h15 + m_ 54 h54
Entropy
m_ 14 s14 + m_ 53 s53 + S_ g , HWST ¼ m_ 15 s15 + m_ 54 s54
Exergy
m_ 14 ex 14 + m_ 53 ex 53 ¼ m_ 15 ex 15 + m_ 54 ex 54 + E_ xD, HWST
Energy efficiency
54 h54 m 53 h53 Þ HWST ¼ ððm m_ 14 h14 m_ 15 h15 Þ
Exergy efficiency
54 ex 54 m 53 ex 53 Þ cHWST ¼ ððm m_ 14 ex 14 m_ 15 ex 15 Þ
_
_
_
_
Hydrogen production from geothermal power plants Chapter 13
TABLE 13.3 Working conditions of the cycle. Parameters
Values
Dead-state temperature, To
25°C
Dead-state pressure, Po
101.3 kPa
Geothermal fluid temperature, Tgf
170°C
The mass flow rate of geothermal fluid, m_ gh
31.24 kg/s
Isentropic efficiencies of pumps, p
0.80
Pelton turbine isentropic efficiency, PT
0.75
PEM electrolyzer temperature, TPEM
80°C
[21]. Also, the other variables in Eq. (13.5) should be defined as: _ Q ¼ 1 To Q_ (13.9) Ex T _ Q is the exergy rate related to heat, To is the ambient Here, Ex temperature, T is the process temperature, and Q_ is the heat rate. _ W ¼ W_ Ex
13.3.2 Thermodynamic model In this subsection, the energetic and exergetic ways to analyze the geothermal energy-based combined plant thermodynamically are given. Some general assumptions that are considered in the thermodynamic model of the combined plant are as follows: l
l
All parts work under steady-state conditions. The changes in kinetic and potential energies and exergies are negligible. The pressure drops and heat losses for connecting parts and components are negligible.
The mass, energetic, entropy, and exergetic balance equations are written as (one may visit [Ref. 20] for details): X X m_ ¼ m_ (13.2) X
_ + mh
in
X
Q_ +
in
X
_ + ms
X in
X Q_
in
W_ ¼
X
out
_ + mh
out
+ S_ gen ¼
X
Q_ +
X
out
X
_ + ms
W_ (13.3)
out
X Q_
(13.4) T X X X X X _ Q+ _ W¼ _ Q _ + _ + Ex Ex Ex mex mex outX out in in in _ D _ W + Ex + Ex in
in
T
out
out
out
where ex illustrates specific exergy and should be computed as: (13.6) ex ¼ exph + exch Here, exph and exch give physical and chemical exergy.
¼
Useful energy output in product Energy input
(13.12)
Exergy output in product Exergy input
(13.13)
c¼
For each process of the combined cycle, the balance equations for mass, energy, entropy, and exergy along with the energetic and exergetic performance equations are tabulated in Table 13.2. For a multigeneration plant, it is very important to describe the efficiency that represents all outputs and inputs in the most appropriate way. Therefore, the energetic and exergetic efficiencies of the combined system and its six subplants are defined as below, For the double-flash geothermal plant: W_ T1 + W_ T2 (13.14) DFG ¼ ðm_ 2 h2 m_ 12 h12 Þ cDFG ¼
th where u0i and u00 i show the chemical potential of the i portion in the thermomechanical and chemical equilibrium
_ W _ W + Ex Ex T1 T2 ðm_ 2 ex2 m_ 12 ex12 Þ
(13.15)
for the hydrogen production plant: m_ 29 h29 + m_ 30 h30 m_ 25 h25 + W_ PEM
(13.16)
m_ 29 ex29 + m_ 30 ex30 _ W m_ 25 ex25 + Ex PEM
(13.17)
HPP ¼ cHPP ¼
for the freshwater production plant:
(13.7) (13.8)
(13.11)
_ D is the exergy destruction rate and S_ gen is the where Ex entropy generation rate. To investigate the performance of a geothermal energybased combined plant, the energetic and exergetic efficiency values are defined as
(13.5)
exph ¼ ðh ho Þ To ðs so Þ X ni u0i u00 exch ¼ i
(13.10)
_ W is the exergy rate related to work and W_ is the where Ex work rate. _ D ¼ T0 S_ gen Ex
l
217
m_ 23 h23 + m_ 24 h24 + W_ PT m_ 16 h16 + W_ P1
(13.18)
_ W m_ 23 ex23 + m_ 24 ex24 + Ex PT _ W m_ 16 ex16 + Ex P1
(13.19)
FWPP ¼ cFWPP ¼
218
PART
II Thermodynamic analysis of geothermal power plants
for the single-effect absorption cooling with ejector plant: Q_ Cooling SEACE ¼ (13.20) ðm_ 12 h12 m_ 13 h13 Þ + W_ P2 cSEACE ¼
_ Q Ex Cooling
_ W ðm_ 12 ex12 m_ 13 ex13 Þ + Ex P2
(13.21)
for the drying plant: DP ¼ cDP ¼
Q_ Drying ðm_ 13 h13 m_ 14 h14 Þ _ Q Ex Drying
ðm_ 13 ex13 m_ 14 ex14 Þ
for the hot water production plant: Q_ Hot_water HWPP ¼ ðm_ 14 h14 m_ 15 h15 Þ cHWPP ¼
_ Q Ex Hot_water ðm_ 14 ex14 m_ 15 ex15 Þ
(13.22)
(13.23)
(13.24)
(13.25)
for the integrated system: W_ net + Q_ Cooling + Q_ Heating + Q_ Hot_water + m_ 30 LHV H2 IS ¼ ðm_ 1 h1 m_ 15 h15 Þ (13.26) cIS ¼
_ Q _ Q _ Q _ Wnet + Ex _ 30 exH2 Ex IS Cooling + ExHeating + ExHot_water + m ðm_ 1 ex1 m_ 15 ex15 Þ (13.27)
For a double-flash geothermal plant-based multigeneration system, the net work generation is defined as X W_ p W_ PEM (13.28) W_ net ¼ W_ T1 + W_ T2 + W_ PT Finally, the energy- and exergy-based coefficient of performance values for the SEAC with ejector process can be described, respectively, as follows: Q_ Eva Q_ Gen + W_ P2
(13.29)
_ Q Ex Eva _ W _ Q + Ex Ex
(13.30)
COPen,SEACE ¼ COPex,SEACE ¼
Gen
13.4
P2
Results and discussion
In order to improve the understanding of the plant’s capacity, it is very significant to utilize different assessments to investigate how this capacity varies with design indicators (Table 13.4). In this section, the energetic and exergetic analyses and exergy destruction assessments for the current geothermal power-based plant are presented and discussed in detail.
TABLE 13.4 Design parameters of the integrated system. Parameters
Values
Isentropic efficiencies of turbines, Tur
0.78
Separator 1 inlet temperature, T2
162.73°C
Separator 1 inlet pressure, P2
500 kPa
Separator 2 inlet temperature, T7
154.53°C
Separator 2 inlet pressure, P7
400 kPa
Working fluid of ejector cooling cycle
Isobutane
Ejector working temperature, T31
41.92°C
Ejector working pressure, P31
486.5 kPa
Ejector entrainment ratio, fejec
0.4397
The results of the thermodynamic assessment performed for the system are illustrated in Table 13.5. In terms of energy and exergy efficiencies, it is seen that the subsystem with the highest energy and exergy effectiveness values is related to the drying subsystem. The energy and exergy efficiencies of this subsystem are obtained as 82.73% and 78.46%. Considering the energy and exergy efficiencies of the whole system, it can be seen from this table that the energy and exergy efficiencies are 57.13% and 53.42%, respectively. The useful outputs from the integrated multigeneration plant are summarized in Table 13.6. The amount of electricity produced in the system is calculated as 2034 kW. The amount of hydrogen produced per unit time in the integrated system is computed as 0.0018 kg/s. The amounts of other useful outputs obtained from this plant can be seen from this table. Considering the useful outputs, it is seen that the output obtained for hot water production is higher than the rest of the subsystems. The impacts of the ambient temperature on the energy effectiveness of the combined system and its subsystems are illustrated in Fig. 13.4. It can be clearly seen in this way that the ambient temperature contributes positively to the system’s energetic effectiveness. It appears to have only a negative effect on the SEACE. When the temperature is increased gradually from 0°C to 45°C, the energy efficiency of the entire system increases from 55.17% to 58.74%. As seen from this figure, the slope of the line showing the efficiency in hydrogen production is higher than the others. The energy effectiveness of the hydrogen production cycle increases from 65.02% to 71.11% when the temperature is gradually increased from 0°C to 45°C. Another performance criterion that is examined to assess the system performance is exergy efficiency. When the reference temperature is gradually increased from 0°C to 45°C, the changes in the exergy efficiency of the system and its subsystems are given in Fig. 13.5. Fig. 13.5 also shows the positive effect of increased reference temperature on
Hydrogen production from geothermal power plants Chapter 13
219
TABLE 13.5 Energetic and exergetic results for the integrated system. Energetic performance (%)
Exergetic performance (%)
Exergy destruction rate (kW)
Exergy destruction ratio (%)
Double-flash geothermal plant
26.72
29.15
1863
33.38
Hydrogen production
68.34
64.67
994
17.81
Freshwater production
74.08
71.93
627
11.23
Single-effect absorption cooling with ejector
18.84
16.03
859
15.39
Dryer process
82.73
78.46
642
11.5
Hot water production
78.25
73.34
597
10.7
Integrated system
57.13
53.42
5582
100
Subplants
TABLE 13.6 Useful outputs from the integrated geothermal system. Plant outputs
Values
Total electricity generation, W_ Total
2034 kW
Cooling rate, Q_ Cooling
1214 kW
Heating rate, Q_ Heating
921 kW
Water heating rate, Q_ Hot_water
2348 kW
Drying, Q_ Drying
1132 kW
Hydrogen generation, m_ Hydrogen
0.0018 kg/s
Freshwater generation, m_ 22
5.23 kg/s
1
Energy efficiency
0.8
0.6 D FGP HP FWP SEACE
0.4
DP HWP IS
0.2
0 0
5
10
15
20
25
30
35
Dead state temperature (oC) FIG. 13.4 Effect of dead-state temperature on the energy efficiency of integrated system parts.
40
45
PART
II Thermodynamic analysis of geothermal power plants
1
1
0.8
0.8
Energy efficiency
Exergy efficiency
220
0.6
0.4
DFGP
DP
HP
HWP
FWP
IS
SEACE
0.4
DFGP
DP
HP
HWP
FWP
IS
SEACE
0.2
0.2
0 0
0.6
5
10
15
20
25
30
35
40
45
0 160
165
170
175
180
185
190
195
200
Dead state temperature (oC) FIG. 13.5 Effect of dead-state temperature on the exergy efficiency of integrated system parts.
Geothermal fluid temperature (oC) FIG. 13.7 Effect of geothermal fluid temperature on the energy efficiency of integrated system parts.
the exergy effectiveness of the system and its subsystems. When the temperature is gradually increased from 0°C to 45°C, the exergy efficiency of the integrated system increases from 51.07% to 55.36%. It can be seen from Fig. 13.5 that the most significant slope in the lines representing exergy efficiency is in the hydrogen production subsystem, as in the energy-based considerations. The impact of ambient temperature on useful products produced from the plant is given in Fig. 13.6. As a result of the gradual increase of the ambient temperature, the amount of electricity obtained from the integrated plant increases from 1824 to 2218 kW. Considering the impact of a gradual temperature increase on the amount of hydrogen produced, the hydrogen produced increases from 1.59 to 1.97 g/s. The generation of hot water from the plant increases from 2116 to 2551 kW. In general, the increase in temperature can be clearly seen in this way, which has a positive effect on beneficial output production. The effect of geothermal fluid temperature on the energetic performance of the plant and its subsystems is shown in Fig. 13.7. When the geothermal fluid temperature is
gradually increased from 160°C to 200°C, the energy effectiveness of the integrated plant increases from 56.11% to 60.28%. In general, the geothermal water temperature, which has a positive impact on the integrated plant and its subsystems, has an important place in terms of system performance. When the geothermal fluid temperature is increased from 160°C to 200°C, the energy efficiency of the double-flash geothermal plant (DFGP) subsystem increases from 25.98% to 29.04%. The energy efficiency of the freshwater production subsystem increases from 72.9% to 77.7%. The contribution of geothermal fluid temperature to exergy efficiency is shown in Fig. 13.8. When the geothermal fluid temperature is increased from 160°C to 200°C, the exergy effectiveness of the integrated plant increases from 52.26% to 57.04%. With the contribution of the geothermal fluid temperature to the exergy effectiveness of the system and its subsystems in general, it can be easily seen from this form that this contribution is positive. When the geothermal fluid temperature is gradually increased from 160°C to 200°C, the exergetic effectiveness of the drying subsystem increases from 77.21% to 82.3%, and the exergy efficiency of the hot water production subsystem increases from 72.03% to 77.39%. The contribution of geothermal fluid temperature to useful outputs produced from the integrated system is illustrated in Fig. 13.9. When the geothermal fluid temperature is gradually increased from 160°C to 200°C, the amount of hydrogen produced per unit of time increases from 0.0017 to 0.00209 kg/s. The heating output from the integrated plant increases from 873 to 1075 kW. In general, it can be clearly seen that the increase in geothermal fluid temperature has a positive effect on the beneficial output produced from the integrated system. The contribution of the increase in geothermal fluid mass per unit time to the energy performance of the integrated plant and its subsystems is illustrated in Fig. 13.10.
Useful outputs (kW)
2500
0.002 WTotal QCooling QHeating QHot water QDrying
0.0019
2000
0.0018 mHydrogen
1500
0.0017
1000
0.0016
500 0
5
10
15
20
25
30
o
35
40
Hydrogen production (kg/s)
3000
0.0015 45
Dead state temperature ( C)
FIG. 13.6 Effect of dead-state temperature on useful outputs from the integrated system.
Hydrogen production from geothermal power plants Chapter 13
1
Exergy efficiency
0.8
0.6
0.4
DFGP
DP
HP
HWP
FWP
IS
SEACE
0.2
165
170
175
180
185
190
195
Geothermal fluid temperature (oC) FIG. 13.8 Effect of geothermal fluid temperature on the exergy efficiency of integrated system parts.
3000
0.0021 WTotal QCooling QHeating
2500
Useful outputs (kW)
200
QHot water QDrying 0.002
2000
0.0019 mHydrogen
1500
0.0018
1000
0.0017
500 160
165
170
175
180
185
o
190
195
Hydrogen production (kg/s)
0 160
0.0016 200
Geothermal fluid temperature ( C)
FIG. 13.9 Effect of geothermal fluid temperature on useful outputs from the integrated system.
1
Energy efficiency
0.8
0.6
0.4
DFGP
DP
HP
HWP
FWP
IS
SEACE
0.2
0 22
24
26
28
30
32
34
36
38
40
Geothermal fluid mass flow rate (kg/s) FIG. 13.10 Effect of geothermal fluid mass flow rate on the energy efficiency of integrated system parts.
42
221
222
PART
II Thermodynamic analysis of geothermal power plants
When the geothermal fluid mass per unit time is increased from 22 to 42 kg/s, the energy performance of the integrated plant increases from 54.77% to 61.1%. The energy efficiency of the hydrogen production subsystem increases from 64.73% to 74.38%. The energy performance of the SEACE subsystem increases from 18.05% to 20.34%. This figure shows that the increase in the geothermal fluid mass per unit time contributes positively to the energy effectiveness of the integrated plant and its subsystems. The effect of the increase in the geothermal fluid mass per unit time on the exergetic effectiveness of the integrated plant and its subsystems is illustrated in Fig. 13.11. When the geothermal fluid mass per unit time is increased from 22 to 42 kg/s, the exergetic efficiency of the combined plant increases from 50.82% to 57.83%. The exergy efficiency of the freshwater production subsystem increases from 68.67% to 77.36%. The exergy effectiveness of the DFGP subsystem has increased from 27.23% to 32.54%. When looking at the graph in general, it should be seen from this figure that the increase in geothermal fluid mass per unit time contributes positively to the exergy effectiveness of the integrated system and its subsystems. The contribution of the increase in geothermal water mass per unit time to the beneficial outputs obtained from the integrated system is shown in Fig. 13.12. When the
geothermal water mass per unit time is increased from 22 to 42 kg/s, the amount of electricity obtained from the integrated plant increases from 1844 to 2383 kW. The amount of hydrogen produced increases from 0.0017 to 0.00224 kg/s. The generation of hot water increases from 2136 to 2733 kW. It can be clearly seen in this way that the geothermal fluid mass increase in unit time positively contributes to the beneficial outputs obtained from the integrated system, as in the effect of the previously mentioned parameters on the beneficial outputs obtained.
13.5
Concluding remarks
Nowadays, geothermal power-based multigeneration plants offer a promising option for hydrogen production to traditional single-generation, cogeneration, and tri-generation plants. In this chapter, the geothermal energy-based hydrogen production options according to the source temperature are defined. Also, in the case study section, a combined multigeneration system is proposed and thermodynamically investigated to compare performance to provide more useful system design aims. In this combined system, geothermal power is utilized to obtain power, hydrogen, cooling-heating, and hot and desalinated water.
1
Exergy efficiency
0.8
0.6
0.4
DFGP
DP
HP
HWP
FWP
IS
SEACE
0.2
0 22
24
26
28
30
32
34
36
38
40
42
Geothermal fluid mass flow rate (kg/s) FIG. 13.11 Effect of geothermal fluid mass flow rate on the exergy efficiency of integrated system parts.
3000
0.0023
Useful outputs (kW)
2500
QHot water QDrying 0.0022 0.0021
2000 mHydrogen
0.002
1500 0.0019 1000
500 22
0.0018
24
26
28
30
32
34
36
38
40
Geothermal fluid mass flow rate (kg/s)
FIG. 13.12 Effect of geothermal fluid mass flow rate on useful outputs from the integrated system.
0.0017 42
Hydrogen production (kg/s)
WTotal QCooling QHeating
Hydrogen production from geothermal power plants Chapter 13
The following concluding points should be drawn from this chapter: l
l
l
If a multigeneration power plant is chosen instead of single generation of electricity, the overall energetic efficiency increases from 26.72% to 57.13%, where the overall exergetic efficiency increases from 29.15% to 53.42%. Additional useful outputs such as cooling, hydrogen production, drying, and desalinated water can be combined with both plants to increase further the energetic and exergetic effectiveness of the proposed system. The geothermal fluid temperature and mass flow rate have positive effects on combined system performances and useful output rates.
In closing, thermochemical hydrogen production options may offer more appealing application opportunity for geothermal-based hydrogen production to better serve the future hydrogen economy.
References [1] Li Z, Khanmohammadi S, Khanmohammadi S, Al-Rashed AA, Ahmadi P, Afrand M. 3-E analysis and optimization of an organic Rankine flash cycle integrated with a PEM fuel cell and geothermal energy. Int J Hydrogen Energy 2020;45(3):2168–85. [2] Yuksel YE, Ozturk M, Dincer I. Thermodynamic analysis and assessment of a novel integrated geothermal energy-based system for hydrogen production and storage. Int J Hydrogen Energy 2018;43(9):4233–43. [3] Yuksel YE, Ozturk M. Thermodynamic and thermoeconomic analyses of a geothermal energy based integrated system for hydrogen production. Int J Hydrogen Energy 2017;42(4):2530–46. [4] Chen X, Zhou H, Yu Z, Li W, Tang J, Xu C, et al. Thermodynamic and economic assessment of a PEMFC-based micro-CCHP system integrated with geothermal-assisted methanol reforming. Int J Hydrogen Energy 2020;45(1):958–71. [5] Waseem S, Ratlamwala TAH, Salman Y, Bham AA. Geothermal and solar based mutligenerational system: a comparative analysis. Int J Hydrogen Energy 2019;45(9):5636–52. [6] Yuksel YE, Ozturk M, Dincer I. Energetic and exergetic performance evaluations of a geothermal power plant based integrated system for hydrogen production. Int J Hydrogen Energy 2018;43(1):78–90.
223
[7] Khalid F, Dincer I, Rosen MA. Techno-economic assessment of a solar-geothermal multigeneration system for buildings. Int J Hydrogen Energy 2017;42(33):21454–62. [8] Al-Hamed KHM, Dincer I. Investigation of a concentrated solargeothermal integrated system with a combined ejector-absorption refrigeration cycle for a small community. Int J Refrig 2019;106:407–26. [9] Al-Ali M, Dincer I. Energetic and exergetic studies of a multigenerational solar–geothermal system. Appl Therm Eng 2014;71(1):16–23. [10] Mosaffa AH, Zareei A. Proposal and thermoeconomic analysis of geothermal flash binary power plants utilizing different types of organic flash cycle. Geothermics 2018;72:47–63. [11] Wan P, Gong L, Bai Z. Thermodynamic analysis of a geothermalsolar flash-binary hybrid power generation system. Energy Procedia 2019;158:3–8. [12] Nami H, Ranjbar F, Yari M. Thermodynamic assessment of zeroemission power, hydrogen and methanol production using captured CO2 from S-Graz oxy-fuel cycle and renewable hydrogen. Energy Convers Manage 2018;161:53–65. [13] Yuksel YE, Ozturk M. Analysis and performance assessment of a combined geothermal power-based hydrogen production and liquefaction system. Int J Hydrogen Energy 2018;43(22):10268–80. [14] Siddiqui O, Ishaq H, Dincer I. A novel solar and geothermal-based trigeneration system for electricity generation, hydrogen production and cooling. Energy Convers Manage 2019;198:111812. [15] Zhang C, Sun J, Lubell M, Qiu L, Kang K. Design and simulation of a novel hybrid solar-biomass energy supply system in northwest China. J Clean Prod 2019;233:1221–39. [16] Sigurvinsson J, Mansilla C, Arnason B, Bontemps A, Marechal A, Sigfusson TI, Werkoff F. Heat transfer problems for the production of hydrogen from geothermal energy. Energy Convers Manage 2006;47(20):3543–51. [17] Rosen MA, Naterer GF, Chukwu CC, Sadhankar R, Suppiah S. Nuclear-based hydrogen production with a thermochemical copper– chlorine cycle and supercritical water reactor: equipment scale-up and process simulation. Int J Energy Res 2012;36:456–65. [18] Simpson MF, Herrmann SD, Boyle BD. A hybrid thermochemical electrolytic process for hydrogen production based on the reverse deacon reaction. Int J Hydrogen Energy 2006;31:1241–6. [19] International Atomic Energy Agency (IAEA). Advanced applications of water-cooled nuclear power plants. ISBN 978-92-0-105808-9; July 2007. [20] Dincer I, Rosen MA. Exergy: energy, environment and sustainable development. Second Press, New York, USA: Elsevier Science; 2012. [21] Kotas TJ. The exergy method of thermal plant analysis. New York, USA: Butterworth-Heinemann; 1985.
Chapter 14
Multiple flashing in geothermal power plants Ron R. Robertsa, Ibrahim Dincera, and Greg F. Naterera,b a
Clean Energy Research Laboratory, Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON, Canada,
b
Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL, Canada
Nomenclature ex _ Ex h P Q_ s T W_
specific exergy (kJ/kg) exergy rate (kW) specific enthalpy (kJ/kg) pressure (kPa) heat rate (kW) specific entropy (kJ/kgK) temperature (oC) work rate (kW)
Greek letters h
efficiency
Subscripts 0
ambient
14.1 Introduction Geothermal energy is a green power source with minimal to zero greenhouse gas emissions. Using thermal energy inside the Earth, this represents a sustainable power supply. To be a viable energy resource, the thermal gradient should be at least 60°C per 1000 m [1]. In addition, the temperature gradient must occur at a sufficiently shallow depth beneath the surface to make drilling into the resource practical. Many areas in the world have sites that meet these criteria. But most of these locations are in active seismic areas where geological plates meet. The thickness of the Earth’s crust is thinner in these areas, thereby allowing liquid magma beneath it to heat the crust to the desired temperatures. Other areas are in unstable geological regions. Many countries with access to these resources have been rapidly increasing the number and capacity of geothermal systems. Bertani [2] reported that the worldwide installed capacity of geothermal power plants had increased by 20% since 2005, bringing the global capacity to more than 10 GW for the first time, with expectations of a further 70% increase in the next 5 years.
There are many ways to convert geothermal energy into electricity. However, most are based upon a modified Rankine cycle with the Earth acting as the heat source. It is important to note that geothermal sources have qualities varying with respect to their water temperatures and/or enthalpies. In this regard, power plants may be dry steam; single-, double-, or triple-flash steam; or binary plants where one of the power loops is an organic Rankine cycle. Since 2010, more than 40% of the global installed capacity has been single-flash power plants while 20% are double-flash power plants, 27% are dry steam, and 11% are binary plants. The back pressure, hybrid, and triple flash make up the less than 1% remaining [2]. Flash power plants use saturated water or brine as their source fluid. This water is at a high pressure, which allows a reduction in pressure to trigger the generation of steam. The high-quality steam is separated from the liquid in the separator, then it is passed through a turbine where the power is generated, and finally transferred to a condenser. In a singleflash power plant, this process is done once. However, if the pressure is high enough, it is beneficial to reduce the pressure of the separated water to cause it to flash-generate steam again. This is again separated and passed through a turbine. This can be done many times at progressively lower pressures; however, equipment costs limit this typically to at most two or three flashing steps. It is expected that double-flash power plants generate more power output than a single-flash power plant from the same resource. However, this costs significantly more because an additional flashing unit, separator, turbine, condenser, and associated piping are required. The triple-flash plants improve the amount of power produced again, but at a further increase in equipment and cost. Optimization of the flashing pressures to achieve a maximum power output is important because otherwise, a poorly designed unit will not generate as much power as possible. This chapter aims to address the most suitable operational pressures for flashing in single, double, triple, and quadruple-flash units with
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00006-8 Copyright © 2021 Elsevier Inc. All rights reserved.
225
226
PART
II Thermodynamic analysis of geothermal power plants
the ability to extend the method to as many units as desired. Instead of a single case or a small range of cases, a comprehensive family of graphs and models will be presented to cover a range from near-vacuum to 10,000 kPa.
14.2
Geothermal flash power cycles
Note that there are two important criteria in determining the pressure to flash the supply water. At a lower pressure, there is a higher quality of the mixture and, therefore, a massive amount of steam produced for the turbines. However, at a lower pressure, there is less energy in each unit amount of steam produced. The influence of the flashing pressure on both enthalpy and quality for 1000 kPa saturated water is illustrated in Fig. 14.1. In addition, the influence of the flashing pressure on both the exergy and quality for 1000 kPa saturated water is shown in Fig. 14.2. Because it is important to study the importance of varying flashing pressure, there is a need to optimize the power output or maximize the useful output accordingly. For example, this can be studied for the single-flash systems by illustrating the turbine power output rates with respect to the potential flashing pressures, which can be considered most suitable for a potential range of pressures between the condenser pressure and the supply pressure in the respective graphs. A plot shows the maximum with a corresponding optimum flashing pressure. If one considers the options with the flashing units in the form of double, triple, and quadruple options, the matter may become more difficult to evaluate because there is a strong need to estimate the most effective pressures for the flashing process at each potential stage or step accordingly. Several means of determining optimum
flashing pressures for single- and double-flash systems have been previously reported such as Ryley [3], Amiri [4], SelekMurathan [5], and Dagdas [6]. There was also another investigation of triple-flash systems by Chamorro [7]. Each of their approaches was expected to help predict the optimum levels for some of the possible ranges of supply pressures or temperatures for the systems considered. In each case where one can potentially consider determining the optimum pressure levels of the flashing steps, there is a need to study the maximum power output. One can further explore these by performing optimization studies based on the objective function of the maximum power output or the highest efficiency. The flashing pressure can be optimized and studied for the best possible performance. One should remember that the most significant thermodynamic property considered for flashing processes in geothermal power plants is the pressure, which needs to be determined optimally for the practical operation of these geothermal-based power-generating systems. Of course, this needs to be done carefully first for the single-flash systems. Moving from a single- to a double-flash system requires a careful thermodynamic analysis and optimization. Similarly, it is necessary to increase it from a double to a triple and further from a triple to a quadruple. These pressures in each step are obviously treated for thermodynamic analysis, assessment, and optimization. An optimum pressure value of the second step in the double-flashing system is the same as that of a single-flash system with the firststage flashing pressure as its supply pressure. Once the thermodynamic properties for a potentially optimum pressure in the double-flash systems are known, one can evaluate a triple-flash system for obtaining the optimum first-stage
FIG. 14.1 Effects of varying flashing pressure on the enthalpy and quality of steam (1000 kPa source).
Multiple flashing in geothermal power plants Chapter 14
227
FIG. 14.2 Effects of varying flashing pressure on the exergy and quality of steam (1000 kPa source).
flashing pressure by handling the forthcoming second and third stages as a double-flash system with the first-stage flash pressure as its supply pressure. The optimum pressure can be found in one of two ways. First, a direct equation can be used to determine the flash pressure as a function of supply pressure. Next, an indirect equation is developed to determine the pressure drop as a percentage of the supply pressure. In this regard, the flashing pressure is herewith calculated as the supply pressure minus the pressure drop. In general, the system configuration starts with saturated water at the supply pressure. A pressure-reducing valve (V-x) is used to lower the operating pressure, which causes the flashing of some of the saturated water into the steam as required. The steam and liquid water are then separated (F-x) with the steam sent to a turbine (T-x) and the water sent
FIG. 14.3 Illustration of a single-flash system with reinjection.
to either another flashing stage (V-x + 1) or to a pump (P-1) for reinjection back into the geothermal source. The steam sent to the turbine(s) is used to generate power. The exiting low-pressure steam/water mixture (or saturated vapor) from the turbine is then delivered to a condenser (C-x) where it exits as the saturated liquid water. This water from the condenser(s) is sent to a pump (P-2) for reinjection back into the geothermal source. Each of these four types of systems is depicted with their temperature-entropy (T-s) diagrams in the following figures. (a) Single-flash system Here, we first consider a single-flash system and its components for thermodynamic analysis and assessment, which are presented in Fig. 14.3.
228
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.4 T-s diagram for the single-flash system.
In addition, the T-s diagram of the thermodynamic cycle for a pressure value of 10,000 kPa by considering a singleflash unit in the power-generating geothermal system is shown in Fig. 14.4. (b) Double-flash system We now consider a double-flash power-generating system with the respective subunits, which is presented in Fig. 14.5. Fig. 14.6 further illustrates the T-s diagram of the thermodynamic cycle for a pressure value of 10,000 kPa system under the design considered for the double-flash system.
FIG. 14.5 An illustration of the double-flash system with reinjection.
(c) Triple-flash system A triple-flash system and its schematic diagram, along with its components in a geothermal-based power generating plant, are shown in Fig. 14.7. The T-s diagram of the cycle for 10,000 kPa under a triple-flash system is illustrated in Fig. 14.8. (d) Quadruple-flash system In this section, we now consider a geothermal powergenerating system with a quadruple-flash unit, which is shown in Fig. 14.9.
Multiple flashing in geothermal power plants Chapter 14
FIG. 14.6 T-s diagram for the double-flash system.
FIG. 14.7 Schematic of the triple-flash system with reinjection option.
229
230
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.8 T-s diagram for the triple flash system.
FIG. 14.9 Schematic of a quadruple-flash unit with a reinjection option employed in a geothermal-based power-generating system.
Multiple flashing in geothermal power plants Chapter 14
231
FIG. 14.10 T-s diagram of the quadruple-flash unit in a geothermal power-generating system.
The T-s diagram of the thermodynamic cycle for a pressure value of 10,000 kPa under the design of the quadrupleflash system in a geothermal power-generating plant is illustrated in Fig. 14.10.
14.3 Thermodynamic analysis and performance assessment In this analysis, the following assumptions will be made. l
l
l
l
l l
All the systems considered operate under steady-state and steady-flow conditions. The changes in kinetic and potential energies are negligible. The pressure and temperature drops in the pipes and components are insignificant. The throttling processes in the throttling valves are isenthalpic. The separation processes take place adiabatically. All pumps and turbines operate isentropically.
In the thermodynamic analysis and assessment studies, there is a strong need to obtain the state properties correctly and accordingly. The values for thermodynamic properties in this regard are obtained from the Engineering Equations Solver (EES) software. All single-, double-, triple-, and quadruple-flash systems will be analyzed. The following additional assumptions are made [8]. l
The geothermal water employed in the system is obtained at the saturation temperature (quality of x ¼ 0).
l
l
The condensers in the systems operate at the saturation pressure at 40°C (7.381 kPa). The reference state conditions have a temperature of 10°C (T0) and a pressure of 101.3 kPa (P0).
In addition, both exergy and energy efficiencies will be assessed. The specific exergy is defined as [9]: ex ¼ ðh h0 Þ T0 ðs s0 Þ
(14.1)
Note that it is important to make a distinction between geothermal power plants with reinjection and geothermal power plants without reinjection. This will automatically affect the performance of the systems and the exergy values of the streams [10]. For a geothermal power-generating system without reinjection, the power generated by the system is the sum of the total turbine work (W_ turbine ). The energy efficiency is defined as the total turbine power output divided by the input heat flow rate, which is the source energy (Q_ source ); this is written as follows: X W_ turbine energy ¼ (14.2) Q_ source _ source ) yields an exeSimilarly, using the source exergy (Ex rgy efficiency of: X W_ turbine exergy ¼ (14.3) _ source Ex For a geothermal power-generating system with reinjection, there is a need to subtract the sum of the reinjected energy
232
PART
II Thermodynamic analysis of geothermal power plants
(Q_ reinject ) from the incoming geothermal energy. In addition, there is a need to deduct the power consumed by the pumps from the generated turbine power. In conjunction with this, the energy efficiency of the power-generating geothermal system with the reinjection option is defined as follows: X X W_ turbine W_ pump X (14.4) energy ¼ Q_ source Q_ reinject In a similar fashion, one can write the exergy efficiency of the power-generating geothermal system with the reinjection option as follows: X X W_ turbine W_ pump X (14.5) exergy ¼ _ reinject _ source Ex Ex
14.4
Brief discussion of the obtained results
In this section, we discuss the results obtained from the present study investigating the options of switching from single to double, double to triple, and triple to quadruple units for geothermal-based power-generating systems. One needs to know that the amount of work rate obtained from a single turbine system depends strongly upon the inlet pressure, which is assumed to be equivalent to the flashing pressure. This means that increasing the flashing pressure will increase the turbine work rate output. Also, the amount of steam produced in the system will rise to make up for the decrease in the enthalpy of the steam. The energy output then levels off and reaches a peak value. The flashing pressure at
the peak represents potentially an optimum flashing pressure, which is obviously considered the optimum turbine inlet pressure. After reaching such a peak, the amount of work rate generated that can be produced slowly decreases while the flashing pressure continues to rise. At this point, the increase in the amount of steam produced can no longer make up for the continued reduction in the enthalpy of the steam. A plot of various supply pressures yields a family of curves, as shown in Fig. 14.11 for low supply pressures (100–1000 kPa) and in Fig. 14.13 for high supply pressures (1000–10,000 kPa). For a given supply pressure, the optimum flashing pressure can be seen in Fig. 14.12 for pressures between 100 and 1000 kPa and Fig. 14.14 for pressures between 1000 and 10,000 kPa. The equations shown on each of these graphs are valid for supply pressures in kPa and only for the ranges shown. Once the turbine inlet pressure has been determined, the maximum power can be determined from Figs. 14.11 to 14.12. The correlations given in each of these graphs are obtained through careful regression analysis with high correlation coefficients (R2 values) and are valid for turbine inlet pressures in kPa and the ranges shown. The diagrams presented for the double-, triple-, and quadruple-flash systems can be used in a similar manner (Figs. 14.15–14.26). Using these equations for a supply pressure of 3974 kPa (250°C) yields an optimum pressure of 380 kPa, which is a slightly larger pressure drop than the 415 kPa reported by Bodvarson [11]. Using these equations for a 2795 kPa source yields a first-stage flash pressure of 668.5 kPa and a second-stage flash pressure of 97.0 kPa. This is consistent with 666.5 and 96.4 kPa, as reported by Yari [12].
FIG. 14.11 Variation of specific turbine energy versus turbine inlet pressure for a single-flash system (100–1000 kPa supply pressure).
Multiple flashing in geothermal power plants Chapter 14
233
FIG. 14.12 Variation of total specific turbine energy versus turbine inlet pressure for a single-flash system (1000–10,000 kPa supply pressure).
FIG. 14.13 Optimum flash pressure for low-pressure supply under a single-flash system (100–1000 kPa supply pressure).
A curve fit for the optimal turbine pressure vs supply pressure will give accurate values for each range, from 100 to 1000 kPa and 1000 to 10,000 kPa. However, even a sixth-order curve will not properly fit the entire range from 15 to 10,000 kPa as needed for calculation purposes. The error in the single-flash sixth-order curve fit is higher than 5% for much of the range below 300 kPa.
Another approach to developing a model for the optimum supply pressure is to consider both the supply and target in relative terms. Normalize the supply pressure by dividing by the atmospheric reference pressure of 101.325 kPa. This converts the pressure into a dimensionless value. By taking the natural logarithm of this value, the scale can be reduced to a manageable range. The optimum pressure can be
234
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.14 Optimum flash pressure for a high-pressure supply single-flash system (1000–10,000 kPa supply pressure).
FIG. 14.15 Variation of total specific turbine energy versus turbine inlet pressure for a double-flash system (100–1000 kPa supply pressure).
converted to a percentage pressure drop and compared to the available drop by drop % ¼ (supply optimum)/(supply output). A plot of these results for a single-flash system is shown in Fig. 14.27, for a double-flash system in Fig. 14.28, for a triple-flash system in Fig. 14.29, and a quadruple-flash system in Fig. 14.30. When determining the optimum flashing pressures for systems with more than one flashing step, only the first-stage flash was determined parametrically. The
second and any subsequent flashing pressures were determined by equations developed for the preceding systems. The curve shown has an average error of 0.05% over the range of 15–10,000 kPa with a maximum error of 0.14%. This is sufficiently accurate for modeling of a double-flash system. The curve has an average error of 0.02% over the range of 25–10,000 kPa with a maximum error of 0.07%.
Multiple flashing in geothermal power plants Chapter 14
235
FIG. 14.16 Optimum flash pressure for a low-pressure supply, double-flash system (100–1000 kPa supply pressure).
FIG. 14.17 Variation of total specific turbine energy versus turbine inlet pressure for a double-flash system (1000–10,000 kPa supply pressure).
This is sufficiently accurate for use in modeling of the tripleflash system. The curve has an average error of 0.03% over the range of 50–10,000 kPa with a maximum error of 0.09%. This is sufficiently accurate for use in modeling of the quadrupleflash system.
The curve fit has an average error of 0.03% over the range of 100–10,000 kPa with a maximum error of 0.08%. This is sufficiently accurate for modeling of the system and could be used for a quintuple-flash system. The optimal pressures were determined for an isentropic process and were relatively constant when the turbine entropic efficiencies
236
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.18 Optimum flash pressure for high-pressure supply, double-flash system (1000–10,000 kPa supply pressure).
FIG. 14.19 Variation of total specific turbine energy versus turbine inlet pressure for a triple-flash system (100–1000 kPa supply pressure).
were changed. This indicates that the predicted optimum pressure values are robust and can be used over a range of efficiencies. In addition, one can draw a diagram of energy efficiency for the power-generating geothermal system versus the number of turbines employed in the system, as illustrated in Fig. 14.31, with only some small improvements
for the flash geothermal power-generating system with reinjection option. Even when the supply pressures are very high, less than a 1.5% improvement in overall efficiency is seen in Fig. 14.32. Increasing the number of turbines increases the output power. However, an energy analysis alone cannot differentiate between high- and low-quality heat, and as a result, the reinjection heat flow
Multiple flashing in geothermal power plants Chapter 14
237
FIG. 14.20 Optimum flash pressure for low-pressure supply, triple-flash system (100–1000 kPa supply pressure).
FIG. 14.21 Variation of total specific turbine energy versus turbine inlet pressure for a triple-flash system (1000–10,000 kPa supply pressure).
is given more thermal value than it should receive from an exergy perspective [13]. Furthermore, we now illustrate the varying exergy efficiency with the number of turbines employed in the powergenerating system in Figs. 14.33 and 14.34, which clearly shows a more accurate representation. The exergy efficiencies illustrated in these graphs become maximum. Herewith, the actual exergy efficiencies will be lower due to
exergy destruction from irreversible losses in each system component. There is a need to further study switching from a singleflash to a double-flash case for a power-generating geothermal system. In this regard, a graph of varying specific work increases from a single- to a double-flash system reveals that at lower source pressures, there is a larger increase of relative work. However, these values are all fairly close.
238
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.22 Optimum flash pressure for high-pressure supply, triple-flash system (1000–10,000 kPa supply pressure).
FIG. 14.23 Variation of total specific turbine energy versus turbine inlet pressure for a quadruple-flash system (100–1000 kPa supply pressure).
If one considers the supply pressures between the pressure values of 100 and 1000 kPa for practical operations, the increase is approximately 29%. In a similar fashion, the increase for the case where we switch from a double- and triple-flash system becomes approximately 10% over the same pressure range while decreasing to 6% when we increase the number of flashing processes from three to four. It
is estimated that this continues to decrease until the improvement would only be 4% when increasing from a quadrupleflash system to a quintuple-flash system, as shown in Fig. 14.35. We can further continue analyzing and discussing the results. As mentioned earlier, increasing the subject matter pressure will result in a decrease in the relative improvement
Multiple flashing in geothermal power plants Chapter 14
239
FIG. 14.24 Optimum flash pressure for low-pressure supply and quadruple-flash system (100–1000 kPa supply pressure).
FIG. 14.25 Variation of total specific turbine work versus turbine inlet pressure for a quadruple-flash system (1000–10,000 kPa supply pressure).
of the net work. For the pressure range of 1000–10,000 kPa considered for practical geothermal power-generating systems, the improvement from a single-flash to a double-flash system is approximately 25% (with a low point of 23% for 10,000kPa). Similarly, the improvement from a double- to a triple-flash system is only 9%. As the number of flashing
steps increases, the improvements continue to decrease with only a 5% improvement when going to a quadruple-flash system. These values are consistent with the double-flash results of Amiri [4], who reported an increase from 20% to 29%. The triple-flash results of Chamorro [7] reported a 10% improvement (Fig. 14.36).
240
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.26 Optimum flash pressure for high-pressure supply and quadruple-flash system (1000–10,000 kPa supply pressure).
FIG. 14.27 Optimal pressure reduction percentage for the single-flash system.
Multiple flashing in geothermal power plants Chapter 14
FIG. 14.28 Optimal first turbine pressure reduction percentage for the double-flash system.
FIG. 14.29 Optimal first turbine pressure reduction percentage for the triple-flash system.
241
242
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.30 Optimal first turbine pressure reduction percentage for the quadruple-flash system.
FIG. 14.31 Energy efficiency for low-pressure supply (100–1000 kPa supply pressure with reinjection).
Multiple flashing in geothermal power plants Chapter 14
FIG. 14.32 Energy efficiency for high-pressure supply (1000–10,000 kPa supply pressure with reinjection).
FIG. 14.33 Exergy efficiency for low-pressure supply (100–1000 kPa supply pressure with reinjection).
243
244
PART
II Thermodynamic analysis of geothermal power plants
FIG. 14.34 Exergy efficiency for high-pressure supply (1000–10,000 kPa supply pressure with reinjection).
FIG. 14.35 Net work increase with reinjection and lower pressure supply (100–1000 kPa source pressure).
Multiple flashing in geothermal power plants Chapter 14
245
FIG. 14.36 Net work increase with reinjection, high-pressure supply (1000–10,000 kPa source pressure).
14.5 Concluding remarks In this chapter, a new approach was developed to better design the multiflash geothermal power-generating systems for practical applications. In this regard, the current study aimed to predict the most suitable flashing pressure levels for saturated water supply pressures from 100 to 10,000 kPa in geothermal power plants. Such a range is further extended as low as 15 kPa for single-flash units, 25 kPa for double-flash units, 50 kPa for triple-flash units, and 100 kPa for quadruple-flash units in geothermal powergenerating systems. The present study reveals that energy analysis does not provide a qualitative assessment, although it does give a quantitative assessment. However, exergy analysis provides a clear qualitative assessment. An assessment of the exergy efficiency, however, showed an improvement as the number of flashing steps was increased. This improvement became smaller in absolute terms as the number of flashing steps was further increased. Furthermore, an analysis of the relative increase in efficiency showed that a low supply pressure benefits the system more than a high supply pressure. When increasing the number of flashing steps from one to two, the output power increased from 23% at 10,000 kPa to 31% at the 100 kPa supply. With an average improvement of around 28% for single to double flash, the improvement from double to triple was only 10%. Increasing further to a quadruple flash unit would yield a relatively small 5% increase with a projected increase of
between 3% and 4% for a quadruple-flash unit. The use of triple-flash units will require careful evaluation of the cost, given the small increase in efficiency.
References [1] Ameri M, Shamshirgaran SR, Bizhanfard SH, Yousefi MP. The study of key thermodynamic parameters effects on the performance of a flash steam geothermal power plant. In: Proceedings Sustainable Energy and Environment (SEE 2006); 2006. [2] Bertani R. Geothermal power generation in the world 2005–2010 update report. In: Proceedings World Geothermal Congress 2010; 2010. [3] Ryley DJ. An analytical expression in terms of temperature only for optimizing the flash cycle for geothermal power plants. Geothermics 1979;7:9–15. Pergamon Press Ltd. [4] Amiri S, et al. Optimum flashing pressure in single and double flash geothermal power plants. In: Proceeding of 2008 ASME summer heat transfer conference; 2008. [5] Selek-Murathan S, et al. Electricity production from geothermal sources by using double-stage flash system. Energy Sources Part A 2008;30:1884–9. Taylor Francis Group. [6] Dagdas A. Performance analysis and optimization of double-flash geothermal power plants. J Energy Resour Technol 2007;129: 125–33. ASME. [7] Chamorro CR, et al. World geothermal power production status: energy, environmental and economic study of high enthalpy technologies. Energy 2011. https://doi.org/10.1016/j.energy.2011.06.005. Elsevier.
246
PART
II Thermodynamic analysis of geothermal power plants
[8] Cengel YA, Boles MA. Thermodynamics: an engineering approach. New York: McGraw-Hill; 2015. [9] Dincer I, Rosen MA. Exergy: energy, environment and sustainable development. 2nd ed. UK: Elsevier Oxford; 2013. [10] Kanoglu M, Dincer I, Rosen M. Understanding energy and exergy efficiencies for improved energy management in power plants. Energy Policy 2007;35:3967–78. Elsevier.
[11] Bodvarsson G, Eggers DE. The exergy of thermal water. Geothermics 1972;1(3):93–5. [12] Yari M. Exergetic analysis of various type of geothermal power plants. Renew Energy 2010;35:112–21. Elsevier. [13] Naterer GF. Advanced heat transfer. 2nd ed. Boca Raton, FL: CRC Press; 2018.
Chapter 15
Multiobjective particle swarm optimization of geothermal power plants Khurram Kamal, Tahir Abdul Hussain Ratlamwala, Muhammad Afzal Sheikh, Sharjeel Ashraf Ansari, and Muhammad Nouman Saleem Department of Engineering Sciences, National University of Sciences and Technology, Islamabad, Pakistan
Nomenclature m_ h T s c p Q_ PH X_ PH _ in Ex W_ GF W_ ST W_ S W_ HP W_ H hE2S hEx2S hE2PH hEx2PH E_ in
mass flow rate (kg/s) specific enthalpy (kJ/kg) temperature (K) specific entropy (kJ/kg K) specific volume (m3/kg) pressure (kPa) process heat transfer (kW) exergy process heat transfer (kW) input geothermal fluid exergy (kW) geothermal fluid pumping power (kW) power delivered by the steam turbine (kW) net power delivered by the steam turbine and geothermal fluid pump (kW) power of the hexane pump (kW) power delivered by the hexane cycle (kW) the energy efficiency of the simple system the exergy efficiency of the simple system the energy efficiency of the process heater system the exergy efficiency of the process heater system input geothermal fluid power (kW)
15.1 Introduction Due to the advancements in technology, the use of fuels of various types has increased 10-fold. This increase in the use of fuel is a direct effect of the ever-increasing demand for energy. In order to meet this demand, mankind has relied upon fossil fuels such as coal, oil, gas, etc., for ages. This increase in usage has led to a decrease in the reserves of fossil fuels. Even though new fossil fuel reserves are being discovered, they still offer another major problem. Burning of fossil fuels damages the environment by the emission of CO2 (carbon dioxide) and gives rise to the greenhouse effect. CO2 emissions increased by 1.5% in 2018. Other than their harmful nature, fossil fuels are also expensive to acquire. The process of drilling, extracting, refining, and distributing tends to increase the price of fuel. This increase in the price of fuel in the international markets
has been the cause of various energy crises in the past. The energy crises of 1973, 1979, and 1990 were caused either due to a fuel shortage or due to the increase in prices. Wars, political conflicts, strikes by unions, etc., can cause high prices in the market. The 1973 oil crisis was triggered by an oil embargo placed by Middle Eastern countries due to political reasons. This resulted in a 400% increase in oil prices [1]. The aim of achieving a clean and sustainable environment cannot be met as long as the world depends upon traditional fuels. A cleaner substitute for fossil fuels is renewable energy. Renewable energy can eliminate the use of fossil fuels and can be useful in tackling climate change problems. Unlike fossil fuels, which are produced over a cycle of millions of years, renewable energy resources renew themselves in a much shorter period. These resources include all energy that is derived directly or indirectly from the sun as well as from other natural movements and mechanisms of the environment. Geothermal, hydropower, wind, biomass, ocean, solar, etc., are the main sources of renewable energy. These energy sources are sustainable, environmentally friendly, and have the potential to meet energy demand. With the world moving away from conventional methods of electricity generation, the focus has now shifted toward renewable energy resources. It is predicted that in 2020, around 26% of the world’s total electricity will be produced using renewable energy resources [2]. One of the most popular forms of clean energy is geothermal energy. It is an environmentally friendly source of energy that utilizes the heat of the Earth. This heat is extracted from Earth via certain points where it manages to escape from the Earth’s surface. These points of escape usually occur where tectonic plates meet. This heat can be utilized for various purposes such as indoor heating, electricity generation, etc. Radiogenic phenomena taking place within the Earth results in the formation of geothermal energy. Unstable radioactive isotopes of potassium,
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00016-0 Copyright © 2021 Elsevier Inc. All rights reserved.
249
250
PART III Optimization of geothermal power plants
uranium, and thorium undergo decay into a stable form, releasing radioactive radiation within the Earth. The residual heat energy that is produced in the form of radiation is referred to as radiogenic heat [3]. Earth releases 42 1012 W of radiogenic heat from its crust, mantle, and core. The crust accounts for 19% of the total radiogenic heat, whereas the mantle accounts for 76.9% and the core accounts for 4.1% of the total radiogenic heat produced by Earth [4]. Geothermal energy produced within the Earth is classified into seven types based on temperatures ranging from 100°C to more than 300°C [5]. The temperature gradient per km depth plays an important role in the heat energy produced by the geothermal reservoir. In ancient continental regions, a crust temperature gradient of 10°C/km depth is observed while in active volcanic regions, this temperature gradient increases to 100°C/km [6]. Geothermal resources can also be classified on the basis of the exergy of the reservoir and its composition [7]. Geothermal energy can be exploited for space heating and cooling by employing the vapor absorption cycle. Flash steam power plants, dry steam power plants, binary cycle power plants, and hybrid power plants use geothermal resources for power generation [8]. Geothermal steam power plants operate at an energy efficiency of 10%–17% while binary power plants operate at an efficiency of 2.8%–5.5% [9]. In cold areas, a lowtemperature geothermal resource is used for indoor heating by utilizing heat pumps. Flash steam power plants are most commonly used for exploiting these geothermal reserves for electricity production. They are further subdivided into single- and double-flash steam power plants. In a single-flash steam power plant, a high-pressure and high-temperature geothermal resource is extracted from the production well and fed to the separator via a throttling valve. The separator separates the steam at constant pressure and feeds it to the turbine. Steam expands in the turbine, causing it to rotate along with the generator attached to it, thus converting mechanical energy into electrical energy. The steam then enters the condenser, where it is again converted into water and reinjected into the well. In a double-flash steam power plant, two flashing chambers are installed. The hot fluid from the well is flashed once to produce steam that is fed to the high-pressure turbine. The remaining hot fluid undergoes another flashing process and is fed to a lowpressure turbine [10]. An organic fluid can also be utilized as the working fluid by employing a flash-binary cycle. This cycle is suitable for low-temperature geothermal wells. Hot water extracted from the well can be separated into hot fluid and steam via a flashing device. The steam undergoes expansion in the turbine, whereas the hot fluid enters a heat exchanger. This heat exchanger transfers heat to the working fluid of another binary cycle. This cycle can vary from the Kalina
cycle to the organic Rankine cycle. The total output of the plant is then the sum of the total power output from both cycles. This increases the overall power output and hence the efficiency of the plant as the hot fluid that exits the flashing device still contains energy that can be extracted. A geothermal energy reservoir can be used for the production of hydrogen, depending upon the thermal energy content of the reservoir. In high-temperature reservoirs, the heat produced from the reservoir is used in thermochemical and hybrid cycles [11, 12], which produce hydrogen. In low- and medium-temperature reservoirs, electricity is produced at first from a geothermal plant, and this electricity is fed to the electrolyzer for the production of hydrogen. The high-temperature electrolysis of the steam can produce 50%–90% of hydrogen gas [13– 15]. Hydrogen has a high heating value and produces three times more energy than gasoline after combustion. Hydrogen undergoes an electrochemical reaction with oxygen in fuel cells and produces electricity [16], which provides another clean and emission-free energy-producing alternative. Malik et al. [17] designed a renewable base multigeneration system that uses biomass and geothermal resource as the inputs and produces energy by the organic Rankine cycle, a binary flash steam power plant, chilling by the vapor absorption cycle, residential hot water from the separator, and liquified gas. The energy efficiency of the system was 56.5% while its exergy efficiency was 20.3%. AlZahrani [18] designed a system that uses a medium- to a high-temperature geothermal resource as the input and produces electricity by Rankine cycles and hydrogen by electrolysis while providing space heating by the thermal energy of the reservoir. Calise [19] did an exergy and exergoeconomic analysis of the system consisting of an organic Rankine cycle and a distillation unit for producing water, electricity, cooling, heating, and chilled water. A medium- and a high-temperature geothermal resource and a parabolic solar trough collector were used as the input. The exergy efficiency of the system varies between 40% and 50% in winter and 16%–20% in summer. Bicer [20] designed a system that uses geothermal and solar resources as the input and produces electricity, hydrogen, hot water, and cooling and heating of space as the useful outputs. Energy and exergetic analyses show that the corresponding efficiencies were increased by up to 10.8% and 46.3%, respectively.
15.2
System description
Geothermal fluid is pumped from the well and flashed into the chamber via the throttling valve (Fig. 15.1). The steam from the chamber is used to drive the steam turbine. This saturated mixture at the discharge of the steam turbine is utilized for process heating. The high energy content-saturated
Multiobjective particle swarm optimization of geothermal power plants Chapter 15
251
FIG. 15.1 Simple flash geothermal binary power plant.
geothermal liquid is cooled to the ambient conditions by heating hexane in the counter flow heat exchanger. The superheated hexane is fed to the hexane turbine, where it expands and experiences a drop in pressure. The superheated hexane exits the turbine and enters the condenser, where it undergoes a phase change from gas to liquid.
15.2.1
Expansion valve
The expansion valve is an isenthalpic device that reduces the pressure of the geothermal fluid to the flash chamber pressure. This drop in pressure leads to the vaporization of the geothermal stream, which leads to the development of the saturated mixture that is gathered in the chamber.
15.2.2
both phases get separated from each other and come in the equilibrium state. The vapor stream is fed to the turbine from one of the outlets of the separator, whereas from the second outlet, the liquid is discharged and is fed to a heat exchanger. The liquid that comes out of the separator heats the hexane to a temperature that could be used in the binary cycle.
Flash separator
As the fluid passes through the expansion valve, part of it gets converted to the vapor phase. Only the vapor phase is expanded in the turbine; therefore, it is necessary to separate both the components of the mixtures before feeding into the turbine. A flash separator is used to extract the liquid from the mixture. The expanded mixture passes through the separator, where due to the density difference,
15.2.3
Turbine
The turbine is a sort of turbomachine that converts the energy of the flowing stream into the mechanical energy to be used to run the generators for electricity production. In a steam turbine, high-enthalpy steam is fed into the turbine, wherein the steam enthalpy gets converted into the kinetic energy via the nozzle. Then, the rotor blades are driven with the help of the kinetic energy of the fluid. In the system under consideration, two turbines have been used: the first one is driven by the steam, whereas hexane is used to drive the other one. Hexane is heated by exchanging heat with the flashed fluid on its way to the well. Vaporized hexane is then fed through the turbine to generate electric power.
252
PART III Optimization of geothermal power plants
15.2.4
Condenser
This is a heat exchanger that is used to bring the vaporized fluid back into the liquid phase for it to be reused in the cycle. Fluid at the discharge of the turbine is in the vaporized form, and in order to use it again for power generation, it must be reliquified. Usually, it is a shell and tube heat exchanger. A secondary fluid (usually water) is used to extract the heat from the working fluid flowing through the exchanger. The first condenser is used to cool the steam to lower temperatures before reinjecting it into the well while condenser 2 is used to condense the working fluid of the binary cycle, that is, hexane.
15.2.5
Thermodynamic system model
The total energy content of the geothermal fluid drawn from the well is given by the equation below. E_ in ¼ m_ 2 ðh1 h0 Þ
The hexane cycle is an organic Rankine cycle that is used to utilize the energy content available in the saturated liquid discharged from the flash chamber at high pressure and temperature. The net power output from this cycle is the difference in the power output from the hexane turbine and the power input to the hexane pumping unit. The power delivered to the hexane pump and the power obtained from the hexane turbines are given by the relations below.
_ in ¼ m_ 2 ðh1 h0 ðT0 Þðs1 s0 ÞÞ Ex
(15.2)
W_ HT ¼ m_ 8 ðh8 h9 Þ
(15.9)
W_ H ¼ W_ HT W_ HP ¼ m_ 8 ðh8 h9 Þ m_ 8 v10 ðP11 P10 Þ In order to perform a comparative analysis of the system with and without process heating, the First Law and Second Law efficiencies for the two systems were determined. The First Law and Second Law efficiencies of the simple system, which does not have a process heater, are given by the equations below.
(15.4)
The net power output for this steam turbine and geothermal pump is equivalent to the total power output by steam, and it is given by the relation below. W_ S ¼ W_ ST W_ GF ¼ m_ 3 ðh3 h4 Þ m_ 2 v0 ðP2 P0 Þ (15.5) The steam from the discharge of the steam turbine is fed to the process heater. The total heat energy and exergy supplied by this steam are given by the relation below.
ES ¼
W_ S + W_ H E_ in
(15.10)
ExS ¼
W_ S + W_ H _ in Ex
(15.11)
The First Law and Second Law efficiency relations for the system with process heating are given below. EPH ¼
W_ S + W_ H + Q_ PH E_ in
(15.12)
W_ S + W_ H + X_ PH _ in Ex
(15.13)
ExPH ¼
(15.3)
Some portion of this power is consumed in pumping the geothermal fluid from the well. This power utilized is given by the formula below. W_ GF ¼ m_ 2 v0 ðP2 P0 Þ
(15.8)
The net power delivered by the hexane cycle is given by the relation below.
This geothermal fluid enters the flash chamber from where steam is directed to the steam turbine, and saturated liquid is flashed to the hexane heater. The total power delivered by the steam turbine is given by the equation below. W_ ST ¼ m_ 3 ðh3 h4 Þ
W_ HP ¼ m_ 8 v10 ðP11 P10 Þ
(15.1)
Only a limited amount of this energy can be used to obtain the required power. The total useful power that can be obtained from this geothermal fluid is regarded as the exergy content of the geothermal fluid, and its equation is given below.
(15.6)
XProcessHeat ¼ X_ 4 X_ 5 ¼ m_ 3 ððh4 h5 Þ T0 ðs4 s5 ÞÞ (15.7)
Process heater
This is a heat exchanger that utilizes a secondary fluid to absorb the heat energy in the geothermal stream at the discharge of the steam turbine. It brings the vaporized steam back into the liquid phase, which is reinjected into the well.
15.3
Q_ PH ¼ m_ 3 ðh4 h5 Þ
15.4
Multiobjective optimization
A multiobjective optimization problem considers more than one function for optimization as compared to a singleobjective optimization problem that considers only one function for optimization. In contrast to a single-objective optimization that considers only a single dimension to find an optimal solution, multiobjective optimization considers a multidimensional space. Normally, in multiobjective optimization problems, multiple objective functions are converted to a single objective function for which different strategies are employed. However, any multiobjective
Multiobjective particle swarm optimization of geothermal power plants Chapter 15
optimization algorithm performs two tasks. The first task is to find a solution that decides the limits of the Pareto frontier and the second is to find all the solutions that lie on the Pareto frontier. Any multiobjective optimization problem may be defined as the optimization of several objectives subjected to a number of equality and inequality constraints: min =max ∅k ðxÞ, k ¼ 1, 2,…N
(15.14)
subject to ∝ j ðxÞ 0 j ¼ 1,2, …J
(15.15)
bm ðxÞ ¼ 0 m ¼ 0,1, 2, …M
(15.16)
xlow i
< xi < xup i
i ¼ 1, 2, …I
(15.17)
where x is a vector containing process or design parameters. The optimization of a single-objective function involves finding an optimal value, whether minimum or maximum, by comparing different values for a particular objective function. However, in multiobjective optimization, there is a concept of dominance that determines the solution. Solution x1 dominates solution x2 if it fulfills the following conditions: (a) Solution x1 performs no worse than solution x2 for all objectives. (b) For at least one objective solution, x1 performs strictly better than solution x2. The nondominated solution set is defined as a solution set that consists of all solutions that are not dominated by any of the solutions for a given set of solutions. This nondominated set for the whole feasible decision space is called the Pareto-optimal set, and the boundary formed by projecting all the points from the Pareto-optimal set is called the Pareto front. In this chapter, the multiobjective particle swarm optimization (MOPSO) is used as a multiobjective optimization algorithm.
15.4.1
Particle swarm optimization
Particle swarm optimization was developed by Eberhart and Kennedy in 1995 as an expansion of the simulation system of animal social behavior. This algorithm simulates the swarm intelligence of birds and fish. Because of its strong maneuverability, simple structure, and easy realization, it has attracted a lot of interest from researchers and scholars. Similar to other optimization algorithms, the population of swarms is exploited in PSO. This swarm consists of potential solutions to the problem that are called particles of the swarm and are modified after each iteration stochastically. For the manipulation of the swarm, the cooperative model instead of the competitive model is used, contrary to other algorithms. For population modification and to favor the particle that is performing most efficiently, the adaptive velocity vector
253
approach is used instead of selection and mutation operators. The advantage of using the adaptive velocity approach is that the adaptive velocity vector changes its position after each iteration. Particles move toward the area of interest with the help of information they have gathered from their own experience and with the experience of their neighboring particles. In order to store the data related to the best position the particle has ever visited in the search space, a separate memory is used. From now onward, PSO for single-objective optimization will be explained. Let there be an m-dimensional search space denoted by A; let the objective function be f: A ! R and M be the number of particles composing the swarm, A ¼ fx1 , x2 , …xm g
(15.18)
The ith point in the search space is represented as xi ¼ fxi,1 , xi, 2 , …xi, m g
(15.19)
And the best position of this ith particle ever visited in the search space is pbest, i ¼ pbesti,1 , pbesti,2 , …pbesti, m (15.20) The velocity vi of the particle xi is also the m-dimensional vector and could be represented as vi ¼ fvi,1 , vi, 2 , …vi, m g
(15.21)
Initially, the particle velocity and best position are populated randomly in the search space in order to start the procedure. Let the particles that exchange information with xi, called neighboring particles, be MZi A, which is a subset of search space A. The best of all the positions attained by any of the particles is called the global best and is denoted as gbest, fðgbest Þ f pbest, l for all xl that belong to MZi If the number of iterations is denoted by t, manipulation of the swarm is done as follows: wvijðtÞ + c1 r1 vijðt + 1Þ ¼
¼ pbestijðtÞ xijðtÞ + c2 r2 gbestðtÞ xijðtÞ (15.22) xijðt + 1Þ ¼ xijðtÞ + vijðt + 1Þ
(15.23)
where i ¼ 1,2, …, M, j ¼ 1,2, …, m, w in the velocity equation is the inertial weight constant. w is the main difference between the classical and other forms of PSO. Its value ranges between 0 and 1; if it is assigned a value of 1, the particle will retain its previous velocity vector and continue to move in the same direction. Due to this, the probability of finding the global best is minimized; hence, it is recommended to keep the value of the inertial weight constant w toward the lowest value (0.1–0.3). c1 and c2 are the two constants known as the cognitive and social
254
PART III Optimization of geothermal power plants
parameters, and r1 and r2 are the two independent variables that assume the values in the range of 0–1. The local best of any particle could be found by using the expression pbestiðt + 1Þ ¼ xiðt + 1Þ , if fðpbesti Þ < fðxi Þ
(15.24)
Similarly, at each iteration, the global best position is also modified according to (15.25) gbest, ðt + 1Þ ¼ pbestiðt + 1Þ , if fðgbest Þ < f pbest, i At the early stages of PSO, the inertial weight constant is not incorporated into the expression, but experiments showed that a swarm explosion could occur if no velocitycontrolling mechanism is included in the model. As in this case, velocity would increase without any bound and causes the divergence of the swarm. In order to avoid this, a boundary limit is applied on the velocity vmax such that if vij >vmax, then vij ¼ vmax. To control the impact of the velocity of the previous iteration on the current one, some other parameters have also been introduced, but the use of vmax is not abandoned. There are two characteristics that any optimization algorithm must have: exploration and exploitation. Exploration refers to visiting new regions in the search space, whereas exploitation refers to carrying out more local searches in a refined way. An updated velocity equation proposed by Clerc and Kennedy is vijðt + 1Þ ¼ a vijðtÞ + c1 r1 pbestijðtÞ xijðtÞ + c2 r2 gbestðtÞ xijðtÞ (15.26) where a is the known constriction factor. The inertial weight version and this version presented above are equivalent to each other. A stability analysis by Clerc and Kennedy shows that a is a function of c1 and c2. By analyzing the algorithm, various models have been derived, and the most promising results are obtained by setting a ¼ 0.729 and c1 ¼c2 ¼ 2.05. Fig. 15.2 shows a flowchart of PSO algorithm..
15.5
Results and discussion
Steam quality is the amount of vapor in a liquid-vapor mixture. 100% quality steam means that there is no liquid in it (Fig. 15.3). The above graph demonstrates the variation of steam quality as a function of the flash chamber pressure. The graph shows that as the flash chamber pressure is increased, the quality of steam goes down. This is because the flash chamber pressure determines the amount of steam in the liquid-vapor mixture. A higher flash chamber pressure means higher temperatures would be required to boil the water. Therefore, as the flash chamber pressure increases, the quality of steam is decreased (Fig. 15.3). The power that can be generated by a turbine staunchly depends upon the rate at which steam is flowing through it.
Higher flow rates of steam result in more work output, as more energy can be extracted from that steam. It can be seen from the graph that the flash chamber pressure and turbine power have an inverse relationship with each other. Increasing the chamber pressure at a constant environmental temperature decreases the amount of steam (vapor) present in the mixture. This reduction in the amount of steam results in decreased power output that can be obtained from the turbine (Fig. 15.4). Hexane turbine power output represents the power generated by the binary cycle turbine when hexane is expanded in it. The hexane cycle is operated by heating hexane with the help of liquid water that is separated from the liquidvapor mixture in a flash chamber. If the flash chamber pressure is increased, the amount of vapor that is separated from the mixture will be less than the amount of liquid. When this high flow rate of liquid water flows through the heat exchanger, a higher flow rate of hexane will be required to exchange heat with it. Now, the power from the turbine depends upon the mass flow rates of the working fluids flowing through it. Higher flow rates of hexane ultimately result in higher power output from hexane turbines (Fig. 15.5). The line graph illustrates the variation of energy efficiency with the input energy for the simple and process heating systems. The energy efficiency for the process heater system is two to four times higher than the simple system. The simple energy efficiency has a minimum value of 12.07% for a power input of 860 kW. However, as the energy input increases, W_ S increases due to which the energy efficiency-simple system increases and reaches the maximum value of 13.77% against a power input of 885.1 kW. Contrarily, the process heat system utilizes the energy content available in the steam turbine discharge; thus, the energy efficiency-process heat has the minimum value of 38.18% for the power input of 860 kW. Furthermore, the energy efficiency-process heat increases with the increase in power input due to the increase in the W_ S and Q_ PH , leading to the maximum efficiency of 53.21% for the power input of 885.1 kW (Fig. 15.6). The graph indicates the exergy efficiency plot against the exergy input for the simple and process heat systems. The exergy efficiency of a simple system shows the linear relation with the exergy input. The simple system has a minimum exergy efficiency of 49.56% for an exergy input of 209 kW. Moreover, as the exergy input increases from 209 to 216.71 kW, the exergy efficiency reaches the maximum value of 56.24%. This increase in efficiency of the simple system is the consequence of the increase in the W_ S . On the other hand, the exergy efficiency of the process heat system shows the quadratic relation with the exergy input to the system. The exergy efficiency-process heat has a minimal value of 80.12% for the exergy input of 209 KW. The exergy efficiency increases with the
Multiobjective particle swarm optimization of geothermal power plants Chapter 15
255
Particle Initializat
Each Particle’s fitness value calculation
Yes
No Whether current fitness value is better than pBest ot not?
Replace pBest with current value
Do not replace pBest with current value
Best particle’s pBest will be assigned to gBest
Each particle’s Velocity calculation
Each particle’s data values will be updated by its velocity value
No
Yes Target or maximum epochs reached
End
FIG. 15.2 Flowchart of PSO algorithm.
increase in the WSteam and reaches the peak value of 82.15% for the exergy input of 214.6 kW. Upon a further increase in exergy input, the drop in hexane power (W_ H ) and process heat exergy (X_ PH ) dominate the increase in the steam power, thus leading to the net drop in exergy efficiency-process heat. Finally, the exergy efficiency-process heat reaches the minimum value of 81.91% for the exergy input of 216.7 kW (Fig. 15.7). In this chapter, results are generated for two multiobjective functions. One is energy efficiency versus the ratio
of energy output to exergy output, and the other is exergy efficiency versus the ratio of energy output to exergy output. These functions are already explained above and are subjected to constraints with maximum and minimum values of variables used in this multiobjective optimization. The variables that are considered in this multiobjective optimi_ in . zation are W_ S , W_ H , Q_ PH , X_ PH , E_ in , Ex Fig. 15.8 shows the Pareto frontier for the multiobjective optimization of system energy efficiency versus the ratio of energy output to exergy output. The points in red show that
256
PART III Optimization of geothermal power plants
FIG. 15.3 Flash chamber pressure (P2) vs steam quality.
FIG. 15.4 Flash chamber pressure (P2) vs steam turbine power output W_ ST .
the particles or solutions lying on the Pareto frontier are the optimal solutions, whereas the particles that are not lying on the Pareto frontier are suboptimal or inefficient solutions to this problem. It is evident from the figure that increasing the ratio of energy output to exergy output results in a decreased system energy efficiency while decreasing this ratio results in an increase in the energy efficiency of the system. The maximum value of system efficiency that is attainable is
56.9% with an energy output to exergy output ratio of 2.5. The minimum energy efficiency of the system that can be achieved is 50.7% with an energy output to exergy output ratio of 3.04. It is clear from the graph that it is not possible to have an optimum of both objectives at the same time, as these are the points that lie on the extremes of the Pareto frontier. However, considering the midpoint of the Pareto frontier may be a good choice to find an
Multiobjective particle swarm optimization of geothermal power plants Chapter 15
257
FIG. 15.5 Flash chamber pressure (P2) vs hexane turbine power output W_ HT .
FIG. 15.6 Energy input vs energy efficiency.
optimal value, but it depends how drastically one objective changes with respect to a slighter change in another. The trendline or fit for the Pareto frontier for energy efficiency versus energy output to exergy output ratio is given as: Eout ¼ 0:0047 2e 0:0596e + 21:81 Exout
(15.27)
However, it should be well understood that the Pareto frontier is composed of all optimal solutions; therefore,
the selection of optimal values is totally dependent on the person who is making the decision. Here, it is worthy to note that the decision variables play a very important role in deciding the optimal values of objectives for a multiobjective optimization problem and providing insight to the decision-maker. Fig. 15.9A–F shows scatter plots of the optimal values of the decision variables of the system. It is clear from the figure that the values for most of the variables are well scattered in a particular optimal range;
258
PART III Optimization of geothermal power plants
FIG. 15.7 Exergy input vs exergy efficiency.
FIG. 15.8 Pareto frontier for energy efficiency vs Eout/Exout.
however, for variables such as Q_ PH , X_ PH , and E_ in , the values in the optimal range lie around a particular value. This may help the decision-maker decide the optimal values for the system. Fig. 15.10 shows the Pareto frontier of the multiobjective optimization of system exergy efficiency versus
the ratio of power output to exergy output with extreme values of 97.1% and 2.388 maximum and 68.8% and 3.048 minimum. The points in red show that the particles or solutions lying on the Pareto frontier are the optimal solutions, whereas the particles that are not lying on the Pareto frontier are suboptimal or inefficient solutions to this
FIG. 15.9 Scattering of variables obtained from Pareto frontier optimization. (A) Turbine work, (B) hexane work, (C) process heat, (D) exergy process heat, (E) energy in, (F) exergy in.
260
PART III Optimization of geothermal power plants
FIG. 15.10 Pareto Frontier for exergy efficiency vs Eout/Exout.
problem. It can be seen from the figure that increasing the ratio of power output to exergy output will result in decreased exergy efficiency. On the other hand, decreasing the ratio results in increased exergy efficiency. The trendline or fit for the Pareto frontier for exergy efficiency versus energy output to exergy output ratio is given as: Eout ¼ 0:000472ex 0:0527ex + 5:7572 Exout
(15.28)
However, again it should be kept in consideration that the Pareto frontier is composed of all the nondominated solutions that are optimal to a multiobjective optimization problem, and the selection of optimal parameters is dependent on the person making the decision. The scatter plots of decision variables are shown in Fig. 15.11A–F. It is evident from the figure that the values are well scattered in a particular optimal range. For W_ S , the population is distributed between 10 and 50 kW. Similarly, the optimal values for W_ H lie between 70 and 90 kW. However, the optimal values for Q_ PH and X_ PH lie about a particular value, although there are some outliers. For E_ in , most of the values lie between 859 and 860 kW with _ in , the some values approaching 861 kW. However, for Ex majority of the values lie around the optimal value of 209.3 kW.
15.6
Closing remarks
The energy and exergy optimization of a geothermal power plant has been done in this chapter. The plant is designed to meet power requirements by generating electrical power as
well as utilizing low-grade heat for process heating purposes. Geothermal plants use hot springs of fluids coming from the wells to produce power, and the properties of such fluids vary with the location of the well. In this chapter, the fluid temperature has been varied from 493 to 503 K while the pressure has been kept constant for the parametric study. The system comprises two cycles: the first is the ordinary Rankine cycle while the second is the organic Rankine cycle that operates at lower temperatures than the ordinary one. By varying the temperature of the geothermal fluid, the values of process heat, energy efficiency, and exergy efficiency were found to vary from 222.9 to 349.1 kW, 12.05% to 13.77%, and 49.56% to 56.24%, respectively. Multiobjective optimization has been done by using the Pareto frontier technique to obtain the optimized values of the operating parameters. The Pareto frontier of system energy versus the power output to exergy output ratio shows the maximum value of attainable system efficiency of 56.9% with a power output to exergy output ratio of 2.5, whereas the minimum energy efficiency of the system that can be achieved is 50.7% with an energy output to exergy output ratio of 3.04. The Pareto frontier of multiobjective optimization of system exergy efficiency vs the ratio of energy output to exergy output shows extreme values of 97.1% (maximum) and 2.388 (minimum) and 68.8% (minimum) and 3.048 (maximum). However, it should be well understood that the Pareto frontier is composed of all optimal solutions; therefore, the selection procedures of optimal values are totally dependent on the person making the decision. Thus, the population distributions of optimized variables are also presented in the chapter.
FIG. 15.11 Scattering of variables obtained from Pareto frontier optimization. (A) Turbine work, (B) hexane work, (C) process heat,(D) exergy process heat, (E) energy in, (F) exergy in.
262
PART III Optimization of geothermal power plants
REFERENCES [1] 1973 Oil crisis—Wikipedia n.d. https://en.wikipedia.org/wiki/1973_ oil_crisis (Accessed 25 June 2020). [2] OECD. Key world energy statistics; 2016. https://doi.org/10.1787/ key_energ_stat-2016-en. [3] Rybach L. Radioactive heat production in rocks and its relation to other petrophysical parameters. Pure Appl Geophys 1976;114:309– 17. https://doi.org/10.1007/BF00878955. [4] Stacey FD, Loper DE. Thermal history of the Earth: a corollary concerning non-linear mantle rheology. Phys Earth Planet In 1988;53:167–74. https://doi.org/10.1016/0031-9201(88)90139-2. [5] Sanyal, S.K. (2005, January). Classification of geothermal systems— a possible scheme. In Proceedings of the 13th workshop on geothermal reservoir engineering, Stanford, California (pp. 85–88). [6] Yip NY, Elimelech M. Thermodynamic and energy efficiency analysis of power generation from natural salinity gradients by pressure retarded osmosis. Environ Sci Technol 2012;46:5230–9. https://doi.org/10.1021/es300060m. [7] Lee KC. Classification of geothermal resources by exergy. Geothermics 2001;30:431–42. https://doi.org/10.1016/S0375-6505(00)00056-0. [8] Bronicki LY. Geothermal power conversion technology. In: Renewable energy systems. New York: Springer; 2013. p. 818–923. https://doi.org/ 10.1007/978-1-4614-5820-3_233. _ Zamfirescu C. Sustainable energy systems and applications. [9] Dinc¸er I, Springer US; 2012. https://doi.org/10.1007/978-0-387-95861-3. [10] Dincer I, Ezzat MF. Renewable energy production. In: Comprehensive energy systems, vols. 3–5. Elsevier Inc.; 2018. p. 126–207. https://doi.org/10.1016/B978-0-12-809597-3.00310-2. [11] Steinfeld A, Brack M, Meier A, Weidenkaff A, Wuillemin D. A solar chemical reactor for co-production of zinc and synthesis gas. Energy 1998;23:803–14. https://doi.org/10.1016/S0360-5442(98)00026-7.
[12] Steinfeld A, Kuhn P, Reller A, Palumbo R, Murray J, Tamaura Y. Solar-processed metals as clean energy carriers and water-splitters. Int J Hydrogen Energy 1998;23:767–74. https://doi.org/10.1016/ s0360-3199(97)00135-3. [13] Fletcher EA, Moen RL. Hydrogen and oxygen from water. Science 1977;197:1050–6. https://doi.org/10.1126/science.197.4308.1050. [14] Steinberg M, Cheng HC. Modern and prospective technologies for hydrogen production from fossil fuels. Int J Hydrogen Energy 1989;14:797–820. https://doi.org/10.1016/0360-3199(89)90018-9. [15] Doenitz W, Schmidberger R, Steinheil E, Streicher R. Hydrogen production by high temperature electrolysis of water vapour. Int J Hydrogen Energy 1980;5:55–63. https://doi.org/10.1016/0360-3199 (80)90114-7. [16] Sharaf OZ, Orhan MF. An overview of fuel cell technology: fundamentals and applications. Renew Sustain Energy Rev 2014;32:810– 53. https://doi.org/10.1016/j.rser.2014.01.012. [17] Malik M, Dincer I, Rosen MA. Development and analysis of a new renewable energy-based multi-generation system. Energy 2015;79:90–9. https://doi.org/10.1016/j.energy.2014.10.057. [18] AlZahrani AA, Dincer I, Naterer GF. Performance evaluation of a geothermal based integrated system for power, hydrogen and heat generation. Int J Hydrogen Energy 2013;38:14505–11. https://doi. org/10.1016/j.ijhydene.2013.09.002. [19] Calise F, D’Accadia MD, MacAluso A, Piacentino A, Vanoli L. Exergetic and exergoeconomic analysis of a novel hybrid solar-geothermal polygeneration system producing energy and water. Energy Convers Manage 2016;115:200–20. https://doi.org/10.1016/j.enconman.2016. 02.029. [20] Bicer Y, Dincer I. Development of a new solar and geothermal based combined system for hydrogen production. Sol Energy 2016;127: 269–84. https://doi.org/10.1016/j.solener.2016.01.031.
Chapter 16
Artificial neural network-based optimization of geothermal power plants a € G€ urcan C¸etina, Osman Ozkaraca , and Ali Kec¸ ebas¸ b a
Department of Information Systems Engineering, Technology Faculty, Mu gla Sıtkı Koc¸man University, Mu gla, Turkey, b Department of Energy Systems
Engineering, Technology Faculty, Mu gla Sıtkı Koc¸man University, Mu gla, Turkey
Nomenclature E˙x h i lr ṁ max min net o P Q_ R s t T x w Ẇ WR
exergy rate (kJ/s or kW) specific enthalpy (kJ/kg) model () learning rate (%) mass flow rate (kg/s) maximum () minimum () transfer function () output value () pressure (kPa) heat transfer rate (kW) given function specific entropy (kJ/kg K) target value () temperature (°C or K) input () weight () work rate (kW) total uncertainty ()
Greek symbols D « c
difference () exergy or Second Law efficiency (%) specific exergy (kJ/kg)
Subscripts dest in i, j, n k out 0
destruction input successive number of elements location output reference state
Abbreviations ANN Con
artificial neural network condenser
EES GEN GPP LVQ MAPE MPE MSE NCG ORC pre-he PU RBFNN Recup RMS R2 SCADA Turb Vap
Engineering Equation Solver generator geothermal power plant learning vector quantization mean absolute percentage error mean percentage error mean square error noncondensable gases organic Rankine cycle preheater pump radial basis function neural network recuperator root mean square correlation coefficient Central Supervisory Control and Data Acquisition turbine vaporizer
16.1
Introduction
Geothermal is a resource of hot water, steam, and gases that contains various chemicals, and it is formed by heat accumulated at various depths of the Earth’s crust. Geothermal energy is a type of energy that is generated from these sources or their ability to do work in direct or indirect ways. It is a renewable, sustainable, inexhaustible, cheap, reliable, environmentally friendly, and domestic primary energy resource. It is an energy resource that does not cause air pollution and minimizes environmental problems when used carefully. The installed capacity of geothermal energy in the world is 14.9 GW, according to data from 2018 [1]. Turkey is located on a geological and geographical location as an active tectonic belt. When countries with a high increase in installed power and energy production between 2010 and 2015 are listed, Turkey took first place with an increase of 610% and 410%, respectively. It is followed by Germany, Kenya, Nicaragua, and New Zealand [1]. Therefore, Turkey is located between the rich countries of the world in terms of geothermal resources.
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00008-1 Copyright © 2021 Elsevier Inc. All rights reserved.
263
264
PART
III Optimization of geothermal power plants
The existing infrastructure of geothermal plants must continuously be modernized with developing technology to consistently benefit from this energy resource. However, the high cost of such systems and the failure to operate them for a short time lead to significant costs and monetary losses. For that reason, the optimal way seems to develop mathematical models of geothermal power plants (GPPs) and make these plants most efficient by considering the outputs of these models. The most challenging issue in this process will be to operate the system most appropriately or efficiently with the available sources. This, in turn, leads to a comprehensive optimization problem. Optimization aims at obtaining the most appropriate solution by providing certain restrictions in line with the given purpose or objectives [2]. Modeling and analysis are essential components of the optimization. The most accurate mathematical model should be created to reflect the actual system data, and the best value should be reached for the objective function(s) determined through this model to obtain successful optimization results. Energy and exergy analyses are well-established performance analysis methods utilized to investigate energy conversion systems. Considering the energy analysis, the amount of consumed energy is calculated regardless of the difference between work and heat. The analysis and design of engineering systems that are based only on the First Law are not sufficient [3]. This is due to factors such as material balances, energy balances, and equilibrium relationships that do not show how a system is used effectively for the given energy sources [3]. Exergy destruction caused by irreversibilities can be calculated with the exergy analysis method. Determining the exergy destruction is a critical thermodynamic feature in the thermal process. Exergy is defined as the maximum amount of work that can be produced by a system as it is brought into equilibrium with environmental conditions. If exergetic calculations are done for a system, it becomes possible to take into account the difference between its energy qualities [4]. Therefore, exergy analysis is a more convenient approach to evaluate systems. The exergy destruction of the system components must be minimized to achieve an optimum efficiency value of a system. GPPs are complex, and their modeling equations represent GPPs as nonlinear systems. They contain many parameters and different states of these parameters. Each parameter can be linearly related to each other, or independent variables can be included in the GPP model. This, in turn, greatly complicates the optimization of the system. The failure or time-consuming feature of the traditional optimization method has led to the emergence of alternative heuristic or metaheuristic analysis methods for modeling purely natural phenomena. Metaheuristic algorithms have attracted particular attention in recent years [5]. The most important of these metaheuristic algorithms, which are
solution methods obtained by adapting certain algorithms to the problem to be solved, are artificial neural network (ANN) algorithms. The application results of ANN optimization algorithms are also significant as they create a lower or upper limit for precise solution methods. Such solution algorithms do not guarantee a definite solution, but they can guarantee a solution that is suitable for the purpose. Application results show that ANN optimization algorithms are problem-dependent algorithms. That is, they may be successful in one problem, but not in another problem [6]. The problems to be solved are so complex that even the heuristic method cannot achieve correct solutions at a reasonable running time [7]. Nevertheless, ANNs have attracted considerable attention as a modeling and estimation technique. Without using any bias, they have a certain capacity to map linear and nonlinear dependencies in the data and to solve complex problems. The advantages of ANN are nonlinearity, flexibility, speed, simplicity, and adaptive learning. ANNs have been successfully applied in many application fields such as health, defense, engineering, economics, meteorology, and education [8]. The application of ANNs to optimization problems goes back to the early 1980s [9]. In parallel, artificial intelligence applications are increasing exponentially in the field of energy. Studies using ANNs and similar methods in energy power plants are mostly focused on topics such as estimation, error detection, and power management. Besides, these studies are carried out in many different energy facilities using algorithmbased methods. For example, Prieto et al. [10] developed an ANN model to estimate the performance of a condenser for a power plant. De et al. [11] developed an ANN model for the cofired combined heat of the reinforced coal biomass and the entire steam cycle of the power plant. Smrekar et al. [12] discussed two ANN models, one for the boiler and one for the turbine, to estimate the power output from a real coal-fired power plant using the fewest controllable parameters as the input. Fast and Palme [13] integrated the ANN model in a computer system for monitoring a combined heat and power plant in Sweden, and the estimation accuracy of the model was very high. Yoru et al. [14] applied the ANN method to a cogeneration system for exergetic evaluation. In this study, the ANN model was trained and tested with the data obtained from the actual operating conditions. An efficient wind energy estimation has recently become very important for power system planning, operation, and control. Tawfek et al. [15] tried to develop a short-term wind energy estimation method using artificial intelligence techniques. Rezaee et al. [16] attempted to estimate the power output of the power plant by considering the speed of the wind, the direction of the wind, and climate factors with the ANFIS system. Ma´rcio et al. [17] compared the performance of computational intelligence techniques to model the estimated monthly potential in hydropower generation.
Artificial neural network-based optimization of geothermal power plants Chapter 16
Two different approaches are used to estimate energy production: a polynomial neural network and conventional ANN. ANN used two different optimization algorithms to train the model: Levenberg-Marquardt and Bayesian regulation. The analysis in the study shows that the models well estimate the values that determine the constant energy of the hydroelectric power plant. This indicates that models developed using ANN can be an important tool for energy planning and decision making. It is difficult to measure the complex relationships between the input and output of a GPP system by analytical methods because of its nonlinear structure with many unknown factors. ANNs have a high learning capacity and are capable of identifying and modeling complex nonlinear relationships between the input and output of a system [18]. Studies conducted with ANNs show that this method can be used for systems involving complex problems such as GPPs. In one study carried out by Arslan [19], the electricity production of the Simav geothermal field using the Kalina cycle was examined. The author suggested the use of ANNs, as the design of such technologies involves complex calculations and requires more competence and a longer time. Arslan and Yetik [20] modeled binary conversion systems similar to the organic Rankine cycle (ORC) in GPPs with an ANN. The power produced in the system and the power of the circulation pump were considered the output in the model. The authors reported that all ANN models were of acceptable accuracy and yielded satisfactory results for the prefeasibility of an ORC-binary GPP. Ruliandi [21] used a feed-forward back-propagation ANN to estimate the performance of a GPP under wide operating conditions. The ANN model was used to estimate the specific steam consumption of the facility using 10 input parameters such as steam inlet parameters, turbinegenerator parameters, condenser parameters, cooling tower parameters, and ambient parameters. The study concluded that suitable and sufficient training data, together with an independent measurement of steam flow, could be used to develop a good performance program that can describe the failure of the facility or vehicles. In case the device reading data show a significant shift from the estimated value in ANN, it is stated especially in the steam sales contract plan that a comprehensive analysis at the facility may be a good sign to prevent losses. Erdogan et al. [22] performed an exergetic optimization of a hybrid solargeothermal power plant based on the Taguchi method. The effects of ambient temperature and solar radiation on the output parameters of a system were first examined using a thermal model. The study showed that turbine inlet and outlet pressures are the dominant factors affecting the performance of the system. Khosravi et al. [23] suggested an ANN method for modeling the geothermal ORC equipped with a solar thermal unit. These models are based on the main design parameters of the geothermal system, that is,
265
solar radiation, well temperature, the mass flow rate of the working fluid, the turbine outlet pressure, the surface area of the solar collector, and the preheater inlet pressure. These developed smart models use the specified input variables to estimate a geothermal ORC’s net power output, energy efficiency, exergy efficiency, and leveled cost of energy. Energy, exergy, and economic analyses are performed for coolers with a low global warming potential. The design of both efficient and low-cost GPPs remains one of the problems faced by energy engineers. It has become extremely important to propose more accurate and systematic approaches to developing energy conversion systems, especially in developing countries. This is due to the harmful effects of these systems on the environment and their complex designs as well as to meet global demands in response to increasing energy needs.
16.2
Artificial neural network
The main reason behind the ability of human beings to produce solutions to problems in natural life is the ability of the biological brain to learn by living or experiencing. An ANN is an information processing technology that is inspired by the way biological nervous systems process information. As shown in Fig. 16.1, a biological neural network is the sum of a large number of nerve cells in our brain. The basic structural unit of the nervous system is the neuron (or nerve cell). Neurons make connections with each other; that is, they communicate in ways similar to electrical circuits, enable the emergence of brain functions, and perform their duties. Biological neural networks have very high performances and are capable of processing complex events. An ANN is the ability of computers to do this. This technology can be defined as a parallel distributed processor consisting of mathematical processing units, each of which has its own memories that are connected to each other by means of variable weight connections [24–26]. The early studies on ANNs began in the 1940s when Donald Olding Heb, neuroscientist Warren McCulloch, and mathematician Walter Pitts adapted artificial neural applications in the field of neurobiology to engineering studies. After these early studies, ANNs have been used in many areas to date because they have produced successful results. For example, ANNs have been widely used in cases where there are multidimensional, noisy, complex, inaccurate, incomplete, and defective sensor data with a high probability of error, and where there are no algorithms but only examples exist. Applications have been developed for many functions such as possible function estimates, classification, association or pattern matching, time series analysis, signal filtering, data compression, pattern recognition, nonlinear signal processing, nonlinear system modeling, optimization, and control applications [27].
266
PART
III Optimization of geothermal power plants
FIG. 16.1 A schematic view of the biological nerve cell.
Thanks to their estimation and learning abilities, ANNs can generate solutions more easily than statistical methods and are more advantageous than conventional methods. Complex problems that cannot be modeled or are very difficult can be easily solved using ANNs. An artificial nerve is simpler than a biological nerve but mimics the functions of the four basic components of a biological nerve. These are dendrite, cell body (or soma), axon, and synapse. The basic unit of an ANN is an artificial nerve called the processing element or node. Therefore, the node in an ANN shown in Fig. 16.2 corresponds to the neuron in a biological neural network. Fig. 16.2 represents the layers of a basic ANN model. According to Fig. 16.2, the symbol x indicates the model inputs. Each of these inputs is multiplied by its weight, w. The results are collected with the threshold value and processed with the activation function to find the cell output. Parameters are used to adjust the effect of inputs on outputs. Weights are multiplied by input values and forwarded. The most critical point in a network is to calculate the optimum weight values by spreading the error according to the training set of the given weights. In order to create an ANN model, the following statements must be determined in advance: (i) the binding topology of the neurons, (ii) the transfer and activation functions used by the processor elements, (iii) the learning method, and (iv) the learning rule and algorithm. The model is designed according to the available data and the type of application you want the network to perform. The success of the developed model is directly related to the correct model architecture. The ANN topology to be applied for the solution of a problem depends primarily on the type of problem. It is crucial to know which network model is more suitable for solving a problem. The choice of ANN topology suitable for the problem depends on the learning algorithm that is considered to be used in the network. Because many ANN structures can be used with a single learning algorithm, the selection of the learning algorithm will also necessarily determine the ANN topology to be used. While determining the ANN structure, multilayer perceptron,
learning vector quantization (LVQ), radial basis function neural networks (RBFNN), and recurrent networks can be selected. The number of layers used in the network and the number of neurons in each layer determine the complexity of the ANN. As the number of layers and the number of neurons in the layers increase, the processing and learning ability of the ANN increases and the convergence time increases. However, the generalization ability of the network decreases, which causes the network to memorize. Using fewer layers than necessary causes the pattern in the data to not be learned adequately. A two- or three-layer network can produce satisfactory results for most problems. In ANNs, one hidden layer can approach any continuous function while two or three hidden layers can approach any arbitrary function. Because they also have more descriptive power, two or three hidden layers tend to adapt more quickly to patterns in training data, so that they learn input-output relationships more quickly [28]. Determining the characteristic features of processor elements is also important in the design of an ANN. The choice of the transfer function and activation function in a neural network cell, as shown in Fig. 16.2, largely depends on the properties, type, and structure of the data. Some transfer functions can be given as Eqs. (16.1)–(16.6) [29], netj ¼
n X
xj wj
(16.1)
xj w j
(16.2)
j¼1
netj ¼
n Y j¼1
netj ¼ max xj wj netj ¼ min xj wj netj ¼
n X
sgn xj wj
(16.3) (16.4) (16.5)
j¼1
netj ¼ netðoldÞ +
n X j¼1
xj wj
(16.6)
Artificial neural network-based optimization of geothermal power plants Chapter 16
267
FIG. 16.2 A simple ANN model.
where, for j ¼ [0, n] range in a system with n input, xj, wj, and netj denote the inputs, the weights, and the transfer function, respectively. The most commonly used transfer function to find the weighted sum is given in Eq. (16.1). The activation function processes the net input to the cell and determines the output that the cell will produce with this input. The most commonly used activation functions are linear, sigmoid, hyperbolic tangent, and logistic functions. The data are subjected to normalization before they are presented to the network. They are used to prevent excessive oscillations in data and to improve system performance. Logarithmic functions can be used for this, but it is generally recommended that the data be scaled to the ranges of either [0, 1] or [1, +1]. Performance functions calculate the cumulative values of the differences between the desired output values and the values produced by the network. It is observed how close the network is to the pattern shown by the training set with these calculated values. And then, the weight values of the connections are changed using this information. Various learning algorithms can be used to achieve relationships between inputs and outputs. The most widely used algorithm is the feed-forward back-propagation learning algorithm [20]. The basic function used to determine the performance of the system in feed-forward ANNs is mean square error (MSE). In addition, mean percentage error (MPE) and mean absolute percentage error (MAPE) are used, as expressed in the following, MSE ¼
n 1X ½ t i oi 2 n i¼1
n 100% X ti oi n i¼1 oi n 100% X ti oi MAPE ¼ o n
MPE ¼
i¼1
i
(16.7)
(16.8)
(16.9)
The purpose of any training algorithm is to minimize global errors such as root mean square (RMS) and correlation coefficient (R2). An important feature of this function can vary across the impact area. The error for hidden layers is determined by the propagation of the error determined for the output layer. The error during learning is called the RMS and is defined as follows: " #0:5 n 1X 2 (16.10) RMS ¼ jti oi j n i¼1 Also, the R2 is:
2XN
3 2 ð t o Þ i i i¼1 5 R2 ¼ 1 4 X N 2 ð o Þ i¼1 i
(16.11)
where t is the target value, o indicates the output value, and i stands for the model. The input and output layers are normalized in the range of (1, +1) or (0, 1). Also, the R2 ranges from 0 to 1. A perfect fit gives an R2 value of 1 while a weak fit gives a value close to 0. The error function is required to be the minimum. For this, the local minimum of these equations should converge in the correct iterative way. Gradient descent, Newton, quasi-Newton, and Levenberg Marquardt optimization algorithms are used to find the local minimum of a function. The gradient descent algorithm is the most used one [30]. This algorithm uses the first derivative of the equation to converge to the local minimum. The Newton method works with the second derivative, and the quasi-Newton method works with an algorithm similar to the second derivative. The change in weights during the learning process is proportional to the learning rate (lr). Generally, this ratio is tried to be taken too big to not to cause any oscillation. The learning rate determines how much the weights will be changed in the next correction step.
268
PART
16.3
III Optimization of geothermal power plants
ANN-based modeling of the system
The optimization study can be performed in three steps to characterize system performance. In the first step, accurate and reliable data are collected from an operating GPP. These data are used as inputs of the analyses. Energy-based and exergy-based thermodynamic analyses are then performed to evaluate the thermodynamic performance of the system and its components. The data obtained from the analysis results, that is, the exergy efficiency and power rate of the GPP, are the input of the ANN-based model. In the second step, 17 parameters are selected from the data collected from the GPP as the optimization parameters. These parameters are the output parameters of the ANN-based model. In this stage, the ANN network is trained for 17 output parameters corresponding to the input data, and to determine the success of the ANN-based calculations for modeling the GPP. In the last stage, values representing the optimal situation in the dataset are selected as optimized values. It should be noted here that the size of the optimized values used in this study is limited to the values expressing the optimal operation status in the dataset, and these values were removed from the dataset during the training of the ANN. These values are added as input data to the ANN in the last stage, and the optimized values of the 17 parameters are estimated. The output data of the system, which is evaluated for the optimized input conditions with ANN, correspond to the best performance of the GPP.
16.3.1
Description of the system
This study focused on an air-cooled binary GPP with an organic Rankine cycle (ORC). The simple schematic design diagram of the GPP is shown in Fig. 16.3. The main components of the GPP consist of generators, turbines, condensers, evaporators (vaporizers, preheaters, recuperator), and pumps. The heat source of the power plant is medium-temperature brine. The brine is extracted from production wells at 165–170°C, 10–14 bar pressure, and 1100– 1700 ton/h of mass flow rate. The brine is then sent to the separator and separated as steam at 165°C, 10 bar, and 30 ton/h and as a liquid at 165°C, 10 bar, and 1600 ton/h. The plant has two different ORCs with high pressure (13 bar and 570 ton/h n-pentane) and low pressure (7 bar and 700 ton/h n-pentane). As seen in Fig. 16.3, the geothermal brine is the heat source of two different ORCs. However, geothermal steam is only sent to the vaporizer on the low-pressure side as an additional heat source. The path followed by the brine liquid is vaporizer I, vaporizer II, preheater II, and preheater I. It is finally taken into the cooling pool at 85°C and 6 bar. It is then pumped into the reinjection well at 70°C by mixing with the condensed liquid, which leaves vaporizer II. There are certain
differences between the two high-pressure and low-pressure ORCs: (i) the fluid pressures (13 and 7 bar) and temperatures (150°C and 120°C) are different, (ii) the recuperator is only used in the high-pressure ORC, and (iii) in addition to geothermal fluid, geothermal steam is supplied to the vaporizer of the low-pressure ORC. In high- and low-pressure ORCs, the pentane is cooled by air-cooled condensing units. Besides, turbines are directly connected to the generator via a single shaft. Here, the speed of the shaft is balanced with injection valves on the turbines. The electricity produced by the generator first meets the electricity need of the main and auxiliary components, such as the fans and pumps of the power plant, and the remaining part is transferred to the network.
16.3.2
Dataset and uncertainty analysis
The actual operational data, that is, temperatures, pressures, and volumetric flow rates, of the current GPP were collected in 2016 by the Central Supervisory Control and Data Acquisition (SCADA) program. The working conditions of the GPP are monitored in this way, and then the plant performance is evaluated for further conditions/projections. In this context, average data are used throughout 2016 for thermodynamic analysis. The annual average data are listed in Table 16.1, according to the state numbers specified in Fig. 16.3. Reference (dead) state temperature and pressure are taken as 18°C and 1 atm, respectively. Errors and uncertainties in data collection can result from device selection, device status, device calibration, environment, observation, and reading and test planning. An uncertainty analysis is required to prove the accuracy and reliability of this data. The uncertainty analysis is performed using the method described by Holman [31]. Certain types of devices and instruments are used to measure the pressures, temperatures, and volumetric flow rates from the GPP. The uncertainties that arise from the measurement of such data are determined by the following expression. " 2 2 2 #0:5 ∂R ∂R ∂R W1 + W2 + ⋯ + Wn WR ¼ ∂x1 ∂x2 ∂xn (16.12) where R and WR denote a given function and total uncertainty, x1, x2,…, xn are independent variables, w1, w2,…, wn are uncertainty in the independent variables. The total uncertainties for the collected, that is, input data such as temperature, pressure, and volumetric flow rate were determined as 0.88% (in °C), 2.04% (in kPa), and 3.25% (in m3/s), respectively. On the other hand, the total uncertainties for the calculated, that is, output data such as exergy rate, power rate, and exergy efficiency are 1.23% (in kW), 1.98% (in kW), and 3.27%, respectively.
Artificial neural network-based optimization of geothermal power plants Chapter 16
Geofluid output
Condensation output
Geofluid steam input Geofluid input 17
6 9
8
269
13
1
1’
Turb 1
Vap 1
pre-he 1
Air Cooler
Recup 15
14
16 Con 1 PU 1
12
4
GEN
2 5
PU 2
19 Con 2
pre-he 2
18
Vap Turb 2 NCG 2
20
21
10 7
Air Cooler
3
n-Pentane Geofluid n-Pentane (High pressure line) (Low pressure line) (Brine) FIG. 16.3 A schematic diagram of a GPP used for the ANN-based optimization process.
16.3.3
Thermodynamic analysis
The First Law of Thermodynamics, that is, energy analysis, is an expression of the law of conservation of energy that easily provides the energy balance of the system. When changes in kinetic and potential energies are ignored in such a plant, the mass and energy balances for the entire system can be given as follows, respectively, X X m_ out ¼ 0 (16.13) m_ in X X Q_ net,in W_ net,out + m_ in hin m_ out hout ¼ 0 (16.14) _ Q_ net,in , W_ net,out , and h are the mass flow rate, net where m, heat input rate, work output rate, and specific enthalpy, respectively. An algorithm that uses mass and energy balances has been developed on the MATLAB program [32] for the thermodynamic modeling of GPP. In addition, the Second Law of Thermodynamics analysis, that is, exergy analysis, is applied to the location, quantity, and sources of irreversibilities for the components that could be evaluated. For each component of the system, the exergy balance can be written as X X X T0 _ _ dest Q k W_ + m_ in cin m_ out cout Ex 1 Tk ¼0
(16.15)
22
GF(Brine) Condensing
GF(Brine) Steam
_ dest and W_ are the exergy rate associated with where Ex _ j and c exergy destruction and power, respectively. Ex denote, respectively, the exergy rate and the specific flow exergy as expressed by _ j ¼ m_ j cj Ex cj ¼ hj h0 T0 sj s0
(16.16) (16.17)
where T0, h0, and s0 are temperature, enthalpy, and entropy in the reference state, respectively. For the system components shown in Fig. 16.3, the mass, energy, and exergy balances given above can be written. The exergy balance and exergy efficiency statements of the components of the system are listed in Table 16.2. To assess exergy analysis, the exergy efficiency for the system and its components can be respectively defined as follows ek ¼
_ k, output Ex _ k,input Ex
(16.18)
where e is exergy or Second Law efficiency; the subscript output refers to net output, product, or desired value; and the subscript input refers to given or used. Details on the exergy-based analysis of GPPs are presented in Chapter 8. Using 2016 data, energy and exergy analyses are carried out with the help of the MATLAB program [32] linked with the Engineering Equation Solver (EES) program.
270
PART
III Optimization of geothermal power plants
TABLE 16.1 Some data collected from a GPP. State, j
Fluid type
Tj (° C)
Pj (kPa)
ṁj (kg/ s)
1
Brine
164
1026
444
1
Brinesteam
165
986
5.78
10
NCG
165
1026
2.35
2
Brine
136
725
444
3
Brine
110
583
444
4
Brine
110
583
222
5
Brine
110
583
222
6
Brine
90
497
222
7
Brine
80
451
222
8
Brine
85
497
444
9
Brine
107
583
0.68
10
Brinesteam
107
583
5.30
10
NCG
107
583
2.15
11
n-Pentane
105
1248
158
12
n-Pentane
135
1248
158
13
n-Pentane
80
153
158
14
n-Pentane
60
153
158
15
n-Pentane
30
153
158
16
n-Pentane
35
1248
158
17
n-Pentane
55
1248
158
18
n-Pentane
105
653
164
19
n-Pentane
110
653
164
20
n-Pentane
70
125
164
21
n-Pentane
35
125
164
22
n-Pentane
40
653
164
23
Air
18
101
2018
24
Air
18.4
104
1765
25
Air
18
101
2018
26
Air
18.4
104
1765
0
16.3.4
Modeling
An ANN detects the statistical relationship between a set of input and output parameters. It can learn through training and, therefore, cannot be more accurate than the initial training data. So, selecting the appropriate training data from the existing GPP data is vital for the truthfulness of the estimation of the developed ANN model. The exergy
efficiency of GPP is determined by the thermodynamic analysis mentioned in the previous section. The power generation of GPP is measured from the SCADA system during 2016. These two parameters are the input parameters of the ANN. The output parameters of the ANN are the 17 parameters listed in Table 16.3. These parameters are the pressure and mass flow rate of the liquid and steam brine entering the system, the NCG percentage, the mass flow rates of the n-pentane streams, the outlet pressures of the turbines, the mass flow rates of the air inlets of the condensers, and the number of fans used inside the condensers. These parameters can also be called optimization parameters. In Table 16.3, the min-max constraint intervals of the optimization parameters are given. The study is based on the method of estimating the output parameters by using the ANN, which will ensure the optimal operation of the system using the input parameters, that is, the maximum exergy efficiency and the power output, as listed in Table 16.3. To do so, a feed-forward and multilayered ANN model was first developed in the MATLAB R2015b software. The topology of the feedforward neural network consists of one input layer, three hidden layers, and one output layer. ANN models consisting of two or three hidden layers have more descriptive power, and they tend to adapt more quickly to patterns in training data [28]. The main reason for choosing three hidden layers instead of two hidden layers in this study is that it gave better results during the trials than the two hidden layers model. Each layer contains 2 (for two independent input variables) 6, 10, 14, and 17 (for 17 dependent output values) neurons. In the input layer neurons, the pureline activation function was chosen as the activation function, and Levenberg Marquardt, the fastest learning algorithm with high accuracy, was chosen as the training algorithm [19, 20]. The topology of the used ANN is given in Fig. 16.4. The dataset contains 745 pieces of data, including the total generated power obtained from the pressure, temperature, mass flow rate, NCG, and fan count parameters collected from 16 different points of the system, and the exergy efficiency values calculated with the thermodynamic analysis of the system. However, in the current study, the noisy data with extreme values were removed, and the dataset was reduced to 676 pieces of data. As an example, the graphs showing the variations in P1, P20, power output, and exergy efficiency values in the dataset are given in Fig. 16.5A–D, respectively. In Fig. 16.5C, while the power output for the 309th piece of data is 18,012,038.8 W, the P1 value is 988,329 Pa, the P20 value is 114,504, and the Pa and the exergy efficiency are 14.89. In the data preprocessing stage of the study, 676 pieces of data were normalized in the [1, 1] range. Thus, the values measured at different scales were adjusted to a nondimensional scale to minimize the fluctuations in the dataset.
Artificial neural network-based optimization of geothermal power plants Chapter 16
271
TABLE 16.2 The exergy balances and exergy efficiency equations for the components of GPP. Component Vap-1
Exergy balance
E_ dest, Vap-1 ¼ E_ 1 E_ 2
Exergy efficiency
E_ 12 E_ 11
Vap-2
0 E_ dest, Vap-2 ¼ E_ 2 E_ 3 + E_ 1 E_ 10 E_ 9 E_ 19 E_ 18
pre-he-1
E_ dest, pre-he-1 ¼ E_ 4 E_ 6 E_ 11 E_ 17
pre-he-2
E_ dest, pre-he-2 ¼ E_ 5 E_ 7 E_ 18 E_ 22
Recup
E_ dest, Recup ¼ E_ 17 E_ 16 E_ 13 E_ 14
Turb-1
E_ dest, Turb-1 ¼ E_ 12 E_ 13 W_ Turb-1
Turb-2
E_ dest, Turb-2 ¼ E_ 19 E_ 20 W_ Turb-2
E_ dest, Con-2 ¼ E_ 20 E_ 21 E_ 26 E_ 25
E_ dest, PU-2 ¼ W_ PU2 E_ 22 E_ 21
16.4 Results and discussion In this section, the optimized values of 17 different parameters determined from the GPP, which express the optimal conditions in the dataset and which will give the maximum exergy efficiency and power output of the GPP, are estimated by the ANN method. Thus, the performance of the GPP with reasonable accuracy is improved, and a better physical understanding of the process can also be provided. For this, the ANN tool of the MATLAB software is used. The graphical user interface window of the ANN structure modeled by using MATLAB software is given in Fig. 16.6. As the network, a three-layer feed-forward neural network with sigmoid hidden neurons and linear output neurons is used. The Levenberg-Marquardt back-propagation training algorithm (as specified in [19, 20]) is used to train the network. Herein, first the data with the best exergy and power rate values were removed from the dataset, and 675 pieces of data in total were divided into training (70%), test (15%), and verification data (15%). Then, the ANN network was trained for 17 output parameters corresponding to the input data. The purpose of doing this was to determine to what extent the ANN models the GPP system. In the training of the network, a 0.9 value was chosen as the learning rate (lr). The learning process was carried out with the gradient descent algorithm. As is known, ANN training is based on the trial-and-error method. In order to determine the most appropriate training parameters, the network was trained
_
_
E 17 epre-he-1 ¼ EE_11 E_ 4
6
_ E_ 22 epre-he-2 ¼ EE_18 _ 5 E 7 _
_
E 14 eRecup ¼ EE_ 13 E_ 17
16
_
Con-2
PU-2
2
_ _ eVap-2 ¼ _ _ E 19_0 E 18_ _ E 2 E 3 + E 1 E 10 E 9
eTurb-2 ¼ E_W Turb-2 E_
Con-1
PU-1
1
_ eTurb-1 ¼ E_W Turb-1 _ 12 E 13
E_ dest, Con-1 ¼ E_ 14 E_ 15 E_ 24 E_ 23
E_ dest, PU-1 ¼ W_ PU1 E_ 16 E_ 15
_
_
E 11 eVap-1 ¼ EE_12 E_
19
20
_ E_ 23 eCon-1 ¼ EE_ 24 _ 14 E 15 _
_
E 25 eCon-2 ¼ EE_ 26 E_ 20
21
_ _ ePU-1 ¼ EW16_ E 15 PU1 _
_
ePU-2 ¼ EW22_ E 21 PU2
using the trial-and-error method. This process was continued until the MSE value was minimized. The best validation performance curve is given in Fig. 16.7. According to Fig. 16.7, the training started with an error over 100; the error was reduced to 0.02 after 25 iterations, and the training was completed with a total of 89 epochs. However, in the training process, the best validation performance value was reached in the 83rd epoch, and the error value could not be reduced further in the epochs after this value. Also, according to the graph, the results can be said to be reasonable due to the following reasons: (i) the final MSE is low, (ii) the test and validation dataset errors have similar characteristics, and (iii) no significant overfitting occurred after the 89th epoch. Fig. 16.8 shows performance graphs in terms of gradient, Mu, and validation controls by the epoch number of the training of the ANN model for 473 pieces of data. The gradient curve shows the variation in the gradient to investigate the minimum of the cost function against the number of verification checks. The size of the gradient and the number of validation controls are used to end training. When training reaches its minimum performance, the gradient reaches its lowest value. Moreover, according to the val-fail curve, after the 83rd epoch, the system could not reach a lower MSE value and ended the training. Fig. 16.9 shows the regression values for the 17 parameters obtained in the output layer, depending on the exergy efficiency and power rate values given as input in the study. According to Fig. 16.9, the regression value of the model
272
PART
III Optimization of geothermal power plants
TABLE 16.3 The output (optimization) parameters associated with ANN-based modeling. Optimization parameters
Unit
Constraint range (min–max)
P1
kPa
950–1150
P1,steam
kPa
950–1150
DPVap-1
kPa
0–440
DPVap-2
kPa
0–230
DPpre-he-1
kPa
0–200
DPpre-he-2
kPa
0–200
ṁ1
kg/s
442–446
ṁ1,steam
kg/s
8–10
NCG
%
20–40
P13
kPa
115–185
P20
kPa
90–150
ṁ12
kg/s
159.6–161.6
ṁ19
kg/s
167–172
ṁ23
kg/s
1600–2400
ṁ25
kg/s
1600–2400
Number of fans in level I
piece
21–45
Number of fans in level I
piece
21–45
FIG. 16.4 The used model of ANN for the optimization process.
was found to be 0.98 for training, validation, and test data. This figure indicates a very high accuracy and thus proves the accuracy of the model. After ensuring that the designed ANN model met the desired performance level with training and test data, as a next step, the network was tested for the desired optimization parameters. Fig. 16.10 shows the regression curve obtained by running the feed-forward multilayer ANN model used to estimate the 17 parameters that will give the highest exergy efficiency (34.12%) and power rate (24.60 MW) values from the GPP data. According to Fig. 16.10, the linear curve equation of the obtained results is Output ¼ 1 Target 1.9e +04, and the R-value is 0.99986. The closeness of the R-value to 1 indicates that the model successfully estimated the desired values. The success of a modeling process is expressed by the difference (error) between the outputs produced by the actual system and produced by the ANN model with the same input. Accordingly, the parameters and error rates that provide the desired exergy efficiency and power rate values are given in Table 16.4. According to Table 16.4, for example, the target value of P1, which provides the best exergy and power rate values, was 1,029,810 Pa, and the ANN model estimated this value as 1,032,766 Pa. Accordingly, the model correctly estimated the true value with a positive difference of 0.29%. On the other hand, while the target value for DPpre-he-1 was 5483 Pa, the estimated value was 8065 Pa, with a difference of 47.09%. Although the estimated value here is poor, when the average MPE and MAPE values of the network for 17 parameters were analyzed, it is seen that the values of 0.0496 and 0.0691 were obtained,
Artificial neural network-based optimization of geothermal power plants Chapter 16
11
´ 105
1.5 X: 309 Y: 9.883e+05
1.4 P20 [Pa]
P1 [Pa]
10 9.5
1.3
9
1.2
8.5
1.1
8 0
100
200
(A)
300 400 data
500
0
´ 107
200
300 400 data
500
600
500
600
30
25 Exergy Efficiency [%]
Power Rate [W]
100
(B)
X: 309 Y: 1.801e+07
1.6
1.4
1.2
20 X: 309 Y: 14.89
15
10
5
1
(C)
X: 309 Y: 1.145e+05
1
600
1.8
0.8
´ 105
1.6
10.5
2
273
0 0
100
200
300 400 data
500
0
600
(D)
100
200
300 400 data
FIG. 16.5 Data changes of optimization parameters such as (A) P1, (B) P20, (C) power rate, and (D) exergy efficiency.
respectively. From this, it can be inferred that there is an average of 6.91% difference between target values and estimated values according to the MAPE value that gives the most accurate result. In addition, the error rate between the measurement values and the estimated values was obtained as RMSE ¼ 0.1488. The closeness of the RMSE and MAPE values to 0 shows that the model yields good results. These values prove that the developed ANN model is successful in estimating the 17 parameters used in the system for the best exergy efficiency and power rate values. However, it should be noted here that the optimized values obtained in this study are limited to the estimation of the values in the dataset that express the best situation. The comparison of the exergy analysis and ANN-based optimization results according to the exergy destruction rates of the plant components is presented in Fig. 16.11. As seen in the figure, there is a big change in the exergy
destruction rates of the condensers and pumps while there is a small change in those of the other remaining components. According to the exergy analysis, the exergy destruction rates of the condensers and pumps are the highest. With the use of the ANN-based optimization method, their exergy destruction rates decrease except condenser 1 (Con-1). However, its value in exergy analysis is 1.98 MW for Con-1 while ANN-based optimization increases its value by 17%. According to the results of the exergy analysis, the ANN-based optimization lowers the exergy destruction rates of the pumps, condenser 2, and vaporizers 1 and 2. The primary component to be improved according to the exergy analysis is condenser 2 (Con-2). As shown in Fig. 16.11, the exergy destruction rate of Con-2 for exergy analysis is about 2.49 MW while it decreases to 1.1 MW with a 44% improvement for ANNbased optimization. However, there is a 2.1% decrease in
274
PART
III Optimization of geothermal power plants
FIG. 16.6 ANN tool window in MATLAB.
the exergy destruction rate of 2.46 MW in vaporizer 2 (Vap2) for the ANN-based optimization. This is caused by the increase of the pressure of the brine vapor (P1,steam) and the pressure difference (DPVap-2) in Vap-2, although the brine pressure entering Vap-2 (P2) decreases, as can be understood from Tables 16.1 and 16.4. Thus, the exit pressure (P20) of Turbine 2 (Turb-2) decreases, and its exergy destruction rate increases. The decrease in the exergy destruction rate for Con-2 is caused by the decrease in the exit pressure (P20) of Turb-2 and the pressure difference (DPpre-he-2) of preheater 2 (prehe-2). Thus, the exergy destruction rate of pump 2 (PU-2) decreases. When comparing Tables 16.1 and 16.4, the pressure drops of Turb2 cause a decrease in the mass flow rate (ṁ25) of the cooling air in Con-2. In this way, the electricity consumption of the fan units is reduced. As can be seen in Fig. 16.11, the best improvement of the plant is in pumps. The exergy destruction rates of pumps 1 and 2 for the ANN-based optimization decrease by 88% and 86%, respectively. In contrast, the exergy destruction rates of preheaters 1 and 2 increase by 18% and 14%, respectively. In addition to this, drops in the input pressure of the brine to the plant (P1) and the pressure difference (DPVap-1) of Vap-1 create a small change in the exergy destruction rate of Vap-1. However, an increase in the pressure difference (DPpre-he-1) in prehe-1 increases the working fluid pressure (P13) in the ORC. Therefore, it provides a pump (PU-1) and recuperator (Recup) to work at high pressure. On the other hand, the low output pressure (P13) of Turb-1 causes the low pressure of the Recup and Con-1. The cooling air flow rate (ṁ23) of Con-1 with low pressure also decreases. Due to the high- and low-pressure difference in Recup, the pressure difference between the inlet and outlet of PU-1 decreases. Therefore, as seen in Fig. 16.11, the exergy destruction rate of PU-1 decreases.
16.5
FIG. 16.7 Best validation performance of the ANN.
Closing remarks
In this chapter, information about ANNs has been given, and a sample case study has been conducted. ANNs, which are a technique of artificial intelligence, are highly preferred in terms of reducing cost and resource usage in industrial applications with their strong theoretical foundation and power in solving real problems in different information processing areas. ANN models and algorithms provide much more convenience than statistical models in terms of estimation performance, especially in nonlinear and complex systems. In the case study part of the study, the performance of a feed-forward multilayer ANN model in estimating the optimal parameters that would yield the desired best exergy and power values in the S-GPP system was investigated. In the study, the actual operational data obtained from the SCADA system were used to train the ANN model and
Artificial neural network-based optimization of geothermal power plants Chapter 16
FIG. 16.8 ANN response during epochs.
FIG. 16.9 Regression plots for the training, validation, and test values.
275
276
PART
III Optimization of geothermal power plants
establish numerical relationships between process-related parameters. In the optimization process of the application, the row of data with the highest exergy value was extracted from the existing dataset, and the ANN model was trained and validated without this data. At the last stage, the data, including the highest exergy value, was given to the trained network to test the performance and optimization capability of the network. The regression value of the model was found to be 0.98 for the test data. This value shows very high accuracy. For the estimated 17 parameters, the MPE and MAPE values of the network are 0.0496 and 0.0691, respectively. There is an average difference of 6.91% between target values and estimated values according to MAPE value. Furthermore, the error rate between the measurement values and the estimated values is RMSE ¼ 0.1488. The fact that the RMSE and MAPE values are close to 0 indicates that the model is producing good results. Also, for predicted optimal values, the R-value is 0.99986. The proximity of the R-value to 1 indicates that the model has successfully predicted the desired values. The effects of the 17 parameters optimized for the maximum exergy efficiency of the plant
FIG. 16.10 Regression plot for optimization parameters.
TABLE 16.4 The target and estimated values of the optimization parameters. Parameters
Target values
Estimated values
Unit
MPE
MAPE
P1
1,029,810
1,032,766
Pa
0.0029
0.0029
P1,steam
934,351
925,023
Pa
0.0101
0.0101
DPVap-1
322,965
321,171
Pa
0.0056
0.0056
DPVap-2
127,997
147,479
Pa
0.1522
0.1522
DPpre-he-1
5483
8065
Pa
0.4709
0.4709
DPpre-he-2
28,550
38,548
Pa
0.3502
0.3502
ṁ1
444.17
444.79
kg/s
0.0014
0.0014
ṁ1,steam
6.88
6.35
kg/s
0.0770
0.0770
NCG
34
34
%
0.0000
0.0000
P13
139,776
136,376
Pa
0.0243
0.0243
P20
111,748
107,011
Pa
0.0424
0.0424
ṁ12
159.51
159.53
kg/s
0.0001
0.0001
ṁ19
166.42
166.44
kg/s
0.0001
0.0001
ṁ23
1677
1694
kg/s
0.0101
0.0101
ṁ25
2020
2063
kg/s
0.0213
0.0213
Fans in level I
44.45
44.47
piece
0.0004
0.0004
Fans in level II
35.43
35.20
piece
0.0065
0.0065
0.0496
0.0691
Mean
Artificial neural network-based optimization of geothermal power plants Chapter 16
277
FIG. 16.11 A comparison of the results of the exergy analysis and ANN-based optimization in terms of the exergy destruction rates of the components.
on the exergy destruction rates of the plant components were investigated. Therefore, the ANN-based optimization method clearly shows that condensers and pumps need improvement. As a result, the method in this study showed that the modeled feed-forward multilayered ANN model could be used as an effective method for the optimization of GPP parameters.
References [1] Republic of Turkey Ministry of Energy and Natural Resources. Geothermal. [Online]. Available from: https://www.enerji.gov.tr/en-US/ Pages/Geothermal. [2] Rao SS. Engineering optimization: theory and practice. 4th ed. New Jersey: John Wiley & Sons; 2009. [3] Talbi MM, Agnew B. Exergy analysis: an absorption refrigerator using lithium bromide and water as the working fluids. Appl Therm Eng 2000;20:619–30. [4] Sozen A, Arcaklioglu E. Exergy analysis of an ejector-absorption heat transformer using artificial neural network approach. Appl Therm Eng 2007;27:481–91. [5] Gendreau M, Potvin JY. Handbook of metaheuristics. International series in operations research and management science. 2nd ed. London: Springer; 2010. [6] Blum C, Raidl GR. Hybrid metaheuristics, artificial intelligence: foundations, theory, and algorithms. Switzerland: Springer; 2016. [7] Lu Y. Artificial intelligence: a survey on evolution, models, applications and future trends. J Manage Anal 2019;6:1–29. [8] Alba E. Parallel metaheuristics: a new class of algorithms. New Jersey: Wiley and Sons Inc.; 2005. [9] Kumar J. Application of artificial neural networks to optimization problems in electrical power operation. [Online]. Available from: https://lib.dr.iastate.edu/rtd/16886; 1993. [Accessed December 2019]. [10] Prieto MM, Montanes E, Menendez O. Power plant condenser performance forecasting using a non-fully connected ANN. Energy 2001;26:65–79. [11] De S, Kaiadi M, Fast M, Assadi M. Development of an artificial neural network model for the steam process of a coal biomass cofired combined heat and power (CHP) plant in Sweden. Energy 2007;32:2099–109.
[12] Smrekar J, Pandit D, Fast M, Assadi M, De S. Prediction of power output of a coal-fired power plant by artificial neural network. Neural Comput Appl 2010;19:725–40. [13] Fast M, Palme T. Application of artificial neural networks to the condition monitoring and diagnosis of a combined heat and power plant. Energy 2010;35:1114–20. [14] Yoru Y, Karakoc TH, Hepbasli A. Exergy analysis of a cogeneration system through artificial neural network (ANN) method. Int J Exergy 2010;7:178–92. [15] Mahmoud T, Dong ZY, Ma J. Advanced method for short-term wind power prediction with multiple observation points using extreme learning machines. J Eng 2018;1:29–38. [16] Rezaee MJ, Dadkhah M, Falahinia M. Integrating neuro-fuzzy system and evolutionary optimization algorithms for short-term power generation forecasting. Int J Energy Sect Manage 2019;13(4):828–45. [17] Ma´rcio NGL, Pereira RBR, Costa VA, Silva SJA, Moura RPA, Galdino SA. Artificial neural networks approaches for predicting the potential for hydropower generation: a case study for Amazon region. J Intell Fuzzy Syst 2019;36(6):5757–72. [18] Hashimoto Y. Applications of artificial neural networks and genetic algorithms to agricultural systems. Comput Electron Agric 1997;18:71–2. [19] Arslan O. Power generation from medium temperature geothermal resources: ANN-based optimization of Kalina cycle system-34. Energy 2011;36(5):2528–34. [20] Arslan O, Yetik O. ANN modeling of an ORC-binary geothermal power plant: Simav case study. Energy Sources Part A 2014;36 (4):418–28. [21] Ruliandi D. Geothermal power plant system performance prediction using artificial neural networks. In: IEEE conference on technologies for sustainability 2015, Utah, USA; 2015. [22] Erdogan A, Cakici DM, Colpan CO. Thermodynamic optimisation of a hybrid solar-geothermal power plant using Taguchi method. Int J Exergy 2017;23(1):63–84. [23] Khosravi A, Syri S, Zhao X, Assad MEH. An artificial intelligence approach for thermodynamic modeling of geothermal based-organic Rankine cycle equipped with solar system. Geothermics 2019;80:138–54. [24] Elmas C ¸ . Yapay zeka uygulamaları: Yapay sinir agı, bulanık mantık, genetik algoritma. 3rd ed. Ankara: Sec¸kin yayıncılık; 2016. [25] Hopfield JJ. Artificial neural networks. IEEE Circuits Devices Mag 1988;4(5):3–10.
278
PART
III Optimization of geothermal power plants
[26] Fausett L. Fundamentals of neural networks: architectures, algorithms and applications. New Jersey: Prentice-Hall; 1994. [27] Abiodun OI, Jantan A, Omolara AE, Dada KV, Mohamed NA, Arshadf H. State-of-the-art in artificial neural network applications: a survey. Heliyon 2018;4(11), e00938. [28] Abrahamsen EB, Brastein OM, Lie B. Machine learning in python for weather forecast based on freely available weather data. In: Proceedings of the 59th conference on simulation and modelling; 2017. Oslo, Norway.
€ _ [29] Oztemel E. Yapay sinir agları. 3rd ed. Istanbul: Papatya yayıncılık; 2012. [30] Saleh H. Applied deep learning with PyTorch: demystify neural networks with PyTorch. Birmingham: Pact Publishing; 2019. [31] Holman JP. Experimental methods for engineers. 8th ed. New York: McGraw Hill Inc; 2011. [32] MathWorks. Matlab, http://www.mathworks.com/products/matlab/s. [Accessed 15 January 2020].
Chapter 17
Multiobjective optimization of a geothermal power plant Shoaib Khanmohammadia, Onder Kizilkanb, and Farayi Musharavatic a
Department of Mechanical Engineering, Kermanshah University of Technology, Kermanshah, Iran b Department of Mechanical Engineering, Faculty of
Technology, Isparta University of Applied Sciences, Isparta, Turkey c Department of Mechanical and Industrial Engineering, Qatar University, Doha, Qatar
Nomenclature ex E_ _ Ex h m_ Q_ P s S_ T W_
specific exergy (kJ/kg) energy rate (kW) exergy rate (kW) specific enthalpy (kJ/kg) mass flow rate (kg/s) heat energy rate (kW) pressure (kPa) specific entropy (kJ/kg K) entropy rate (kW/K) temperature (°C) work rate (kW)
Greek letters
hen hex
energy efficiency exergy efficiency
Subscripts Con dest f gen geo in is out P Pre Rec T tot Vap 0
condenser destruction fluid generation geothermal inlet isentropic outlet pump preheater rectifier turbine total vaporizer reference state
17.1 Introduction In recent years, the demand for fossil fuels has grown drastically due to the population increase and industrial
development. This has raised great problems with environmental pollution and significant CO2 emissions from energy plants, which are both related to global warming and the security of the energy supply [1, 2]. Recently, the Paris climate agreement reported that carbon emissions had to be mitigated, and the use of fossil fuels had to be reduced by substitution with renewable energy sources as a significant source of energy needs. Due to this requirement, more research attention has been focused on developing and implementing all kinds of renewable energy sources in an attempt to reduce fossil fuel consumption [3]. The shift from fossil fuel-based energy systems to sustainable and clean energy systems, that is, renewable energy, has become crucial in protecting the environment and developing energy sustainability. Renewable energy sources have a substantial ability to ensure a wide variety of advantages, such as a reduction of greenhouse gases, energy independence, and the growth of regional economies [4]. Nevertheless, the majority of renewable resources, in particular solar and wind energy, rely significantly on weather conditions, resulting in a restriction of their availability and dependability as energy resources. To overcome this issue, geothermal energy is a promising alternative energy resource due to its global availability and higher reliability [5]. In addition, geothermal power systems do not generate significant CO2 emissions during electricity production [6]. Throughout the world, electricity generation is the most commonly used and effective method for the use of geothermal energy. Geothermal power generation technologies in operation mainly involve flash, binary, and flash-binary cycles [1]. In order to generate electricity from geothermal energy, numerous power cycles have been investigated from low-temperature heat sources such as the organic Rankine cycle (ORC) to the supercritical Rankine, Kalina, Goswami, and trilateral flash cycles, among others [7]. Among them, ORC is considered the most promising cycle for power generation using geothermal energy [8].
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00011-1 Copyright © 2021 Elsevier Inc. All rights reserved.
279
280 PART
III Optimization of geothermal power plants
The initial cost of a geothermal power plant is generally higher; however, the operating cost is rather lower than the other power generation plant. Therefore, it is essential to investigate the optimum system design conditions for geothermal fluid at the beginning of the feasibility analyses of the power generation system [4]. Thus, the performance of the geothermal power plant can be improved by optimizing the operation parameters and integrating them with other systems. Among optimization methods, in the energy conversion systems, maximizing the exergy efficiency with minimizing the exergy destructions, and minimizing the total cost rate are more addressed. Since these objectives generally conflict for most optimization procedures, the multiobjective optimization (MOO) method is usually employed for these objective functions [8]. Several studies have been conducted to optimize the operation of a geothermal power plant. Astolfi et al. [9] investigated a geothermal power plant using a novel wet and dry configuration. They optimized the off-design operation of the whole plant from a technoeconomic point of view. Sun et al. [10] investigated double-pressure ORCdriven geothermal energy for power generation. They evaluated the system performance using a mathematical model based on thermodynamic and economic laws. The researchers applied a multiobjective parametric optimization using a genetic algorithm (GA) to determine the best operation conditions. Kolahi et al. [11] proposed a combined flash-binary geothermal system with two configurations, power energy production and water purification. ORC was used for electricity generation by utilizing different zeotropic mixtures as the working fluid. They performed a single-objective particle swarm optimization and then a multiobjective particle swarm optimization for the best operation parameters. Kazemi et al. [12] performed analyses for the simulation and optimization of a geothermal-driven ORC, where an ionic liquid was used as a geothermal fluid. For the optimization variables, evaporation and regeneration temperatures were selected to maximize the exergy efficiency and minimize the investment cost. According to their results, it was found that the specific investment cost was comparatively lower for the ionic liquid as a geothermal fluid. Zhou et al. [13] performed the thermodynamic analysis and optimization of a flash-binary geothermal cycle for power generation using ORC for different mixtures of zeotropic fluids. In addition, the particle swarm optimizer was employed to optimize the net output power as the objective function. Yu et al. [14] performed a multiobjective optimization procedure for a novel geothermal-based integrated system with a Kalina and a transcritical CO2 cycle in order to improve the system performance. Another study on the geothermal-based Kalina cycle was carried out by Ghaebi et al. [15]. They investigated a cascade Kalina cycle power plant driven by geothermal energy. They applied single- and multiobjective
optimization techniques to the proposed system. Li et al. [16] conducted an analysis for a geothermal energy system integrated with ORC and a polymer electrolyte membrane fuel cell with multiobjective genetic algorithm-based optimization in order to determine the optimum design parameters. Aali et al. [17] considered a novel combined flash-binary cycle to power generation from the Sabalan geothermal field in Iran. They carried out single- and multiobjective optimization techniques. It was reported that the proposed flash-binary system has a significantly better performance than previous systems. Furthermore, the above literature survey on the optimization of geothermal power systems can be extended for different cycles such as a multigeneration system including cooling, freshwater, hydrogen, and heat [18]; the binary flashing cycle [19]; the transcritical CO2 cycle [20]; and different ORC configurations [21]. In addition, many optimization algorithms can also be found for the optimization of geothermal systems such as a genetic algorithm [22], a gravitational search algorithm [23], an artificial bee colony [24], a particle swarm optimization [25], neural networks [26], a pattern search algorithm [27], and hybrid optimization techniques [28]. From the recent literature, it can be seen that in order to maximize the energy conversion efficiency, the optimization of geothermal power cycles is essential. However, studies on the multiobjective optimization of actual geothermal power-generation systems are limited. In this study, we aim to investigate the performance of the geothermaldriven organic Rankine cycle located in northwest Iran in terms of the First and Second Laws of Thermodynamics. Parametric analyses are carried out to determine the effects of different system parameters on the cycle performance. Finally, the multiobjective optimization procedure is applied to the geothermal power plant to determine the optimum working conditions.
17.2
Geothermal power cycle
The schematic representation of the geothermal-driven actual Rankine cycle is shown in Fig. 17.1. The geothermal reservoir is located in northwest Iran, and the geothermal water temperature is about 145°C and extracted at 600 kPa with a mass flow rate of 391.9 kg/s. The geothermal power cycle is working with n-pentane, and the cycle consists of two ORCs with four main components: a turbine, vaporizer, condenser, and pump. The working fluid temperature and pressure at the turbine inlet are 123°C and 960 kPa for the first cycle and 92.77°C and 500 kPa for the second cycle. The mass flow rates of n-pentane are 54.64 and 52.42 kg/s for cycle 1 and cycle 2, respectively. In both cycles, the working fluid, after exiting the turbine, enters the regenerator and passes to the condenser where it is cooled below the saturation temperature by means of cooling water from the cooling tower. The mass flow rates
FIG. 17.1 Schematic representation of a geothermal-driven transcritical Rankine cycle.
282 PART
III Optimization of geothermal power plants
l
TABLE 17.1 Operation parameters. Geothermal water
l
Extraction temperature, T15
145°C
Reinjection temperature, T19
86.6°C
Mass flow rate, m_ 15
391.9 kg/s
Extraction pressure, P15
600 kPa
Cycle 1 Pump inlet pressure, P4
90 kPa
Pump outlet pressure, P5
960 kPa
Turbine inlet temperature, T1
123°C
Mass flow rate of n-pentane, m_ 1
54.64 kg/s
Mass flow rate of cooling water, m_ 20
293 kg/s
Cycle 2 Pump inlet pressure, P11
85 kPa
Pump outlet pressure, P12
500 kPa
Turbine inlet temperature, T8
92.77°C
Mass flow rate of n-pentane, m_ 8
52.42 kg/s
Mass flow rate of cooling water, m_ 22
274 kg/s
of the cooling water are 293 and 274 kg/s for the first and second cycles, respectively. Next, the liquid n-pentane is pumped to a higher pressure and then enters the preheater before the vaporizer. In the vaporizer, the working fluid is heated and enters the turbine in the vapor phase. The geothermal fluid is reinjected into the ground after circulating through the cycles. For the performance analysis and multiobjective optimization procedure of the geothermal power plant, the operation parameters are shown in Table 17.1.
17.3
l l l l
The mass balance equation for the steady-state and steadyflow processes can be written as [30] X X m_ in ¼ m_ out (17.1) Here, m_ is the mass flow rate, and the subscript “in” denotes inlet and “out” denotes outlet. The energy balance is expressed as: X X Q_ + m_ in hin ¼ W_ + m_ out hout (17.2) where Q_ is the heat transfer rate, W_ is the work rate, and h is the specific enthalpy. For the exergy analysis, the balance equation is defined as [31] X X _ Q Ex _ W¼ _ f , in _ f ,out + Ex _ dest Ex Ex Ex (17.3) In the above equation, the first and second terms are the _ f is the flow exergy of heat and work, respectively; Ex _ dest is the exergy destruction rate. In the exergy; and Ex above equation, each term is defined as follows: _ dest ¼ T0 S_gen Ex _ Q ¼ Q_ T T0 Ex T
All the processes are in steady-state and steady-flow conditions. The pressure drops throughout the cycle are neglected. Heat gains (losses) from (to) the cycle are neglected. All the heat exchangers are the counter flow type. The turbine and pump operations are assumed to be adiabatic.
(17.4) (17.5)
_ W ¼ W_ Ex
(17.6)
_ f ¼ mex _ f Ex
(17.7)
In Eq. (17.7), exf is the specific flow exergy and can be calculated using the equation below: exf ¼ ðh h0 Þ T0 ðs s0 Þ
(17.8)
The general energy efficiency of the equation is written as en ¼
Thermodynamic assessment
A thermodynamic model is constructed using the Engineering Equation Solver (EES) software [29] to evaluate the energetic and exergetic performances of the geothermal-powered ORC. For the analyses, the following assumptions are made: l
l
The geothermal fluid is assumed to be pure water without any solutes. Pump operations for the cooling tower and geothermal water extraction are neglected. Dead-state temperature and pressure are taken as 22°C and 101 kPa, respectively.
W_ net Q_ in
(17.9)
where Q_ in is the heat energy delivered from the geothermal source and W_ net is the net turbine work rate. The exergetic performance of the power plant that is the ratio of total exergy output to total exergy input can be evaluated using the general exergy efficiency equation defined below: ex ¼
_ out _ dest Ex Ex ¼1 _ in _ in Ex Ex
(17.10)
By applying the above-mentioned general balance equations to individual system elements, the energy capacity, exergy destruction rate, and exergetic efficiency of each system component are given in Table 17.2.
Multiobjective optimization Chapter 17
283
TABLE 17.2 Energy and exergy balance equations for the system components of the geothermal power plant. Component
Energy balance equation
Exergy balance equation
Exergetic efficiency
Turbine
W_ T 1 ¼ m_ 1 ðh1 h2 Þ
_ dest, T 1 ¼ Ex _ f , 1 Ex _ f , 2 W_ T 1 Ex
ex, T 1 ¼ Ex _
Regenerator
Q_ Reg1 ¼ m_ 2 ðh2 h3 Þ ¼ m_ 5 ðh6 h5 Þ
_ dest, Reg1 ¼ Ex _ f , 2 Ex _ f , 3 + Ex _ f , 5 Ex _ f ,6 Ex
Condenser
Q_ Con1 ¼ m_ 3 ðh3 h4 Þ ¼ m_ 21 ðh21 h20 Þ
_ f , 3 Ex _ f , 4 + Ex _ f , 20 Ex _ f , 21 _ dest, Con1 ¼ Ex Ex
Pump
W_ P1 ¼ m_ 4 ðh5 h4 Þ
_ dest, P1 ¼ Ex _ f , 4 Ex _ f , 5 + W_ P1 Ex
Preheater
Q_ Pre1 ¼ m_ 7 ðh7 h6 Þ ¼ m_ 18 ðh18 h19 Þ
_ dest, Pre1 ¼ Ex _ f , 6 Ex _ f , 7 + Ex _ f , 18 Ex _ f , 19 Ex
Ex7 Ex6 ex, Pre1 ¼ Ex _ Ex _
Q_ Vap1 ¼ m_ 1 ðh1 h7 Þ ¼ m_ 15 ðh15 h16 Þ
_ dest, Vap1 ¼ Ex _ f , 7 Ex _ f , 1 + Ex _ f , 15 Ex _ f , 16 Ex
_ 7 _ 1 Ex Ex ex, Vap1 ¼ Ex _ 16 _ 15 Ex
Turbine
W_ T 2 ¼ m_ 8 ðh8 h9 Þ
_ dest, T 2 ¼ Ex _ f , 8 Ex _ f , 9 W_ T 2 Ex
ex, T ¼ Ex _
W_ T 2 _
Regenerator
Q_ Reg2 ¼ m_ 9 ðh9 h10 Þ ¼ m_ 5 ðh13 h12 Þ
_ f , 9 Ex _ f , 10 + Ex _ f , 12 Ex _ f , 13 _ dest, Reg2 ¼ Ex Ex
ex, Reg2 ¼
_ f , 12 _ f , 13 Ex Ex _ f , 10 _ f , 9 Ex Ex
Condenser
Q_ Con2 ¼ m_ 10 ðh10 h11 Þ ¼ m_ 23 ðh23 h22 Þ
_ f , 10 Ex _ f , 11 + Ex _ f , 22 Ex _ f , 23 _ dest, Con2 ¼ Ex Ex
Pump
W_ P2 ¼ m_ 11 ðh11 h12 Þ
_ dest, P2 ¼ Ex _ f , 11 Ex _ f , 12 + W_ P2 Ex
Preheater
Q_ Pre2 ¼ m_ 14 ðh14 h13 Þ ¼ m_ 17 ðh17 h18 Þ
_ dest, Pre2 ¼ Ex _ f , 13 Ex _ f , 14 + Ex _ f , 17 Ex _ f , 18 Ex
Q_ Vap2 ¼ m_ 8 ðh8 h14 Þ ¼ m_ 16 ðh16 h17 Þ
_ dest, Vap2 ¼ Ex _ f , 14 Ex _ f , 8 + Ex _ f , 16 Ex _ f , 17 Ex
Cycle 1
Vaporizer
W_ T 1 _
f , 1 Exf , 2
_ Ex
_ Ex
f ,6 f ,5 ex, Reg1 ¼ Ex _ _ Ex f ,2
f ,3
_ f , 20 _ f , 21 Ex Ex ex, Con1 ¼ Ex _ f , 3 Ex _ f ,4
ex, P1 ¼
_ f ,4 _ f , 5 Ex Ex W_ P1 _
_
18
19
Cycle 2
Vaporizer
17.4 Multiobjective optimization Achieving the best results according to the conditions of a problem is a definition of optimization. Optimization is an integral part of the design process. When a design engineer designs a system, the engineer must consider a variety of necessary conditions. Multiobjective optimization methods are used in many branches of science and engineering and are used when it is essential to establish a trade-off between two or more conflicting goals in order to achieve optimal decisions in the system [32]. The most important feature of such methods is that by using optimization models, more than one candidate solution is provided to system designers and engineers. Each of these responses will display a balance between different objective functions [33].
17.4.1
Pareto optimal solution
There is a concept called the nondominated solution in a multiobjective optimization problem. If a candidate’s solution to a multiobjective optimization problem is called “nondominated,” improving the values generated by one or
f , 8 Exf , 9
_ Ex
_ Ex
f , 23 f , 22 ex, Con2 ¼ Ex _ _ Ex f , 10
f , 11
_ _ Ex Ex ex, P2 ¼ f , 12W_ f , 11 P2 _
_
14 Ex13 ex, Pre2 ¼ Ex _ Ex _ Ex 17
18
_ 14 _ 8 Ex Ex ex, Vap2 ¼ Ex _ 16 Ex _ 17
more objective functions reduces the quality of the values generated by the other target functions. Such responses are called Pareto optimal. Without additional information, all of Pareto’s optimal answers are equally good and considered equal to each other [34, 35].
17.4.2
Decision variables
Each system will be determined by a set of quantities. By changing these quantities, the solution to the problem is changed. Specific quantities that have a fixed value outside the problem are known as characteristics. The optimization process begins with the introduction of a set of variables that defines the system, called design variables. The first step in optimization is to identify system design variables. It should be noted that we make design variables as independent as possible.
17.4.3
Objective functions
A selection criterion is needed to compare the answers to the optimization questions. Such a criterion to which the system
284 PART
III Optimization of geothermal power plants
is optimized and is a function of design variables is called the criterion function, the advantage function, or the objective function. The criterion should be a scalar function whose numerical value can be compared.
17.4.4
Genetic algorithm
A genetic algorithm is a computational search technique for finding approximate solutions to optimize models and search problems. A genetic algorithm is a special type of evolutionary algorithm that uses evolutionary biology techniques such as heredity, mutation biology, and Darwin’s principles of choice to find the optimal formula for predicting or matching the pattern. Genetic algorithms are often a good choice for regression-based prediction techniques. This algorithm was first introduced in 1975 by Holland [36]. In genetic algorithms, each possible solution is represented by a sequence of genes called chromosomes. A selected population of chromosomes is called a community, and each community is called a generation at a specific time. After defining the objective function, the initial community is generated. This population is initially assessed, and each chromosome is assigned a rating based on its value. If the criterion is not satisfied, the problem of the generation cycle is done with the aim of improving the answers. The genetic algorithm uses three types of rules at each stage to produce the next generation [37]. 1. Selection rules: Selects people as parents that lead to the next-generation population. 2. Linking rules: Combines the parent to form the children for the next generation. 3. Mutation laws: Uses random changes in parenting to produce the next generation. Here is a brief description of the usual glossary of genetic algorithms: l l l l l
Gene: Contains a property or a variable. Chromosome or organ: A set of genes or variables. Population: A set of chromosomes. Parents: Next-generation production candidate. Reproduction: The application of a cross-sectional operator and a mutation on the parent and the creation of a new chromosome.
To begin with, chromosomes must be formed from existing variables, and the genetic algorithm engine must create a heterogeneous primary population of chromosomes. Each chromosome is then tested. The best chromosomes have a better chance of surviving during other periods and being reproduced, and the weaker ones are doomed to extinction. The next step is to create a second generation of the primary population. The most suitable people to cross-pair mutated, and a new generation is created. This process creates a new population of chromosomes that is more suitable than the
previous generation. The whole process for the next generations is repeated until the algorithm termination condition is reached. The termination conditions are: l l l l
The algorithm reaches a fixed number of generations. The right member is found. The algorithm enters uniformity in generations. Manual inspection.
Due to the random nature of the genetic algorithm, the final solution to a problem in multiple performances is usually different but close to each other. This difference in response is due to differences in the internal factors of the genetic algorithm. The whole process of multiobjective optimization and the genetic algorithm can be illustrated in Fig. 17.2.
17.5 17.5.1
Results and discussion Result of energy and exergy analyses
Table 17.3 represents the results of the thermodynamic modeling of the geothermal power plant simulated by EES. In addition, the value of exergy in each point is presented, which is necessary in the exergy analysis of the system. The major parameter in exergy analysis is the exergy destruction rate. The calculation of the exergy destruction rates of each component can give an appropriate vision about the location with high irreversibility. Fig. 17.3 illustrates the share of the exergy destruction rate of each component of the geothermal power plant system. The results indicate that the highest exergy destruction rate belongs to turbine 2 with 1804 kW. Additionally, the total exergy destruction rate of the system is 6601.5 kW.
17.5.2
Parametric study
In the current section, the results of the parametric analysis on the suggested system in some graphs and figures are presented. The six decision variables included are the mass flow rate of the geothermal source (m_ geo ), the temperature of the geothermal fluid (Tgeo), the mass flow rate of cycle I (m_ cycle I ), the mass flow rate of cycle II (m_ cycle II ), the high temperature of cycle I (PH, cycle I), and the high temperature of cycle II (PH, cycle II). Two main outputs—the net output power (W_ net ), and the total exergy destruction rate _ d,tot )—are investigated. It is worth noting that in the (Ex parametric analysis, all parameters are considered to be fixed based on the data presented in Table 17.1, and just a parameter changed in the allowable domain. Fig. 17.4A and B show the effect of the geothermal fluid properties on the system performance. As is seen, increasing the pressure and temperature of the geothermal fluid has a trivial effect on the net output power and First Law
Optimization problem
Evaluatc fitness of each individual Coding of solution
Objective functions
Specific knowledge
Evolutionary operators
Optimum parameters Final optimum solutions (sets of optimum solutions include thermodynamic first law and effectiveness
Is termination criteria reached? No
evaluation End Mutation
Yes
Selection
Recombination Crossover Initial population Decision variables (mgeo, Tgeo, m1, m2, P1 , P2) Objective functions (Wnet, Exd,tot) Genetic algorithm search
Selection of individual
Mutation operator
Crossover operator
FIG. 17.2 Flow chart of multiobjective optimization and genetic algorithm. (Adapted from Khanmohammadi S, Shahsavar A. Energy analysis and multi-objective optimization of a novel exhaust air heat recovery system consisting of an air-based building integrated photovoltaic/thermal system and a thermal wheel. Energy Convers Manage 2018;172:595-610.)
286 PART
III Optimization of geothermal power plants
TABLE 17.3 Thermodynamic characteristics of different points of the studied geothermal power plant system. State
T (°C)
P (kPa)
m_ (kg/s)
h (kJ/kg)
s (kJ/kg K)
ex (kJ/kg))
1
123
960
54.64
519.2
1.397
107.1
2
71.86
90
54.64
447.5
1.451
19.55
3
46.61
90
54.64
400.9
1.311
14.33
4
28.43
90
54.64
5.881
0.02052
0.1422
5
29.77
960
54.64
9.727
0.02858
1.609
6
49.36
960
54.64
56.37
0.1778
4.223
7
81.01
960
54.64
136.1
0.4133
14.38
8
92.77
500
52.42
471.8
1.335
78.1
9
70.71
85
52.42
445.5
1.452
17.38
10
45.81
85
52.42
399.7
1.313
12.36
11
27.18
85
52.42
2.957
0.01083
0.07788
12
27.77
500
52.42
4.694
0.01438
0.7681
13
47.09
500
52.42
50.54
0.162
3.042
14
81.69
500
52.42
137.6
0.4201
13.96
15
145
600
391.9
610.8
1.791
85.8
16
132.5
600
391.9
557.3
1.661
70.66
17
122
600
391.9
512.6
1.549
58.9
18
119.3
600
391.9
501
1.52
55.99
19
86.6
600
391.9
489.9
1.491
53.26
20
21.42
101.3
293
89.96
0.3167
0.002373
21
39.04
101.3
293
163.6
0.5596
1.98
22
21.42
101.3
274
89.96
0.3167
0.002373
23
39.58
101.3
274
165.9
0.5667
2.105
FIG. 17.3 The share of each geothermal power plant component in the exergy destruction rate.
efficiency while they increase the exergy efficiency substantially. In addition, the variations of the mass flow rates of both ORCs are examined in Fig. 17.5A and B. As is seen with increasing the mass flow rate of both cycles, the net output power and exergy destruction have a similar change. It means that increasing the mass flow rate of cycle I increases the net output power of the system, and the exergy destruction rises too. For instance, with increasing the mass flow rate of cycle I from 50 to 60 kg/s, the total exergy destruction rate and the net output power of the system change from 6488 to 6716 kW and 6479 to 5358 kW, respectively. In addition, Fig. 17.6A and B represent the effect of the high pressure of cycle I and cycle II on the main outputs of
Multiobjective optimization Chapter 17
287
FIG. 17.4 The influence of geothermal fluid on the main output of the system (A) ṁgeo,(B) Tgeo.
FIG. 17.5 The influences of the mass flow rate of cycles on the two considered outputs: (A) mass flow rate of cycle I, (B) mass flow rate of cycle II.
the system. As can be observed, the increase in both pressure levels has the same effects on the net output power, the exergy destruction rate, and the energy efficiency.
In this way, the ideal point can be helpful. Based on the definition, the ideal point is a point on the Pareto at which two objective functions have the best possible values. It is evident that the ideal point is not a feasible point on the Pareto plot. So, the nearest point to the ideal point can be selected as the final optimum point, namely point P2. In Fig. 17.7, points P1 and P3 are associated with singleobjective optimization when the total exergy destruction rate and the net output power are the sole objective function, respectively. The characteristics of points P1 to P3 include decision variables, and the objective functions are represented in Table 17.4. As is seen, the ideal point helps designer find a state of the system among all states suggested by multiobjective optimization. In addition to this technique, considering the change of decision variables in the Pareto points can help the designer to select decision variables in such a way that the system has the best performance. In Fig. 17.8, the value of six decision variables for 53 states introduced by Pareto optimization is presented. As is seen, in some graphs the value of the decision variables tends to be in the upper or lower bound of the domain. In this case, it
17.5.3
Multiobjective optimization
In the following section, the results of multiobjective optimization are presented. In the multicriteria optimization, most of the goals are incompatible with each other, so finding an optimal solution at the same time to optimize all functions is almost impossible. However, a set of solutions known as Pareto points can be found that make the best interactions between goals so that they do not improve unless they make other goals worse. Fig. 17.7 shows the Pareto diagram based on the total exergy destruction rate and the net output power. This figure shows the nondominant optimization solutions. It means that there is not any preference between them from a mathematical point of view. However, based on the designer criteria, some points can be selected as the final solution of the optimization problem.
288 PART
III Optimization of geothermal power plants
17.6
Concluding remarks
A multiobjective optimization procedure was applied to a geothermal-driven organic Rankine cycle located in northwest Iran in order to determine the optimum operating parameters. The genetic algorithm method was used for the optimization method. The geothermal plant performance was investigated in terms of the First and Second Laws of Thermodynamics. The analysis leads to the following conclusions: 1. The net power generation of the geothermal power plant is 4994 kW in the reference condition (based on Table 17.1). 2. The highest exergy destruction rate belongs to the turbine of the second cycle with 1804 kW, followed by the vaporizer of the first cycle. The total exergy destruction rate of the system is calculated to be 6601.5kW. 3. The pressure and temperature of the geothermal fluid have a trivial effect on the net power generation and the system’s thermal efficiency. 4. The mass flow rate of n-pentane also has an essential effect on system performance. With increasing the mass flow rate of cycle I from 50 to 60 kg/s, the total exergy destruction rate and the net output power of the system change from 6488 to 6716 kW and 6479 to 5358 kW, respectively. 5. According to the optimization results, a net output power of 5603.38 kW and an exergy destruction rate of 5868.4 kW can be achieved. 6. Based on the multiobjective optimization and the scatter distribution of decision variables, the mass flow rate and high pressure of cycle II tend to be in the higher bounds of these variables in the optimum states suggested by the Pareto.
FIG. 17.6 The effect of the high pressure of the ORCs on the two considered cycles: (A) high pressure of cycle I, (B) high pressure of cycle II.
can be concluded that the objective functions do not have a conflict with changing these decision variables. As the results indicate, the value of the mass flow rate of cycle II is dispersed in the allowable range for this parameter.
Total exergy destruction rate (kW)
6600 P3
6400 6200 P2
6000 5800 5600 5400
P1 Ideal point
5200 5000 5300
5400
5500
5600
5700
5800
Wnet (kW)
FIG. 17.7 Pareto front diagram for the exergy destruction rate and the net power output as two objective functions.
Multiobjective optimization Chapter 17
289
TABLE 17.4 Characteristics of different points represented on the Pareto plot. Decision variable
Objective function
m_ cycle II (kg/s)
PH, cycle I (kPa)
PH, cycle (kPa)
W_ net (kW)
_ d , tot Ex (kW)
m_ geo (kg/s)
Tgeo °C
m_ cycle I (kg/s)
P1
364.69
140.00
59.86
50.08
919.56
591.33
5359.19
5359.83
P2
365.38
140.05
59.95
56.00
966.56
592.26
5603.38
5868.40
P3
377.79
141.94
60.00
60.00
971.56
592.85
5710.39
6438.85
Ideal point
–
–
–
–
–
–
5359.19
6438.85
(A)
(B)
(C)
(D)
(E)
(F)
II
FIG. 17.8 The scatter distribution of decision variables: (A) the mass flow rate of the geofluid, (B) the temperature of the geofluid, (C), the mass flow rate of cycle I, (D) the mass flow rate of cycle II, (E) the high-pressure level of cycle I, and (F) the high-pressure level of cycle II.
290 PART
III Optimization of geothermal power plants
References [1] Cao L, Lou J, Wang J, Dai Y. Exergy analysis and optimization of a combined cooling and power system driven by geothermal energy for ice-making and hydrogen production. Energy Convers Manage 2018;174:886–96. [2] Boyaghchi FA, Nazer S. Assessment and optimization of a new sextuple energy system incorporated with concentrated photovoltaic thermal—geothermal using exergy, economic and environmental concepts. J Clean Prod 2017;164:70–84. [3] Behzadi A, Gholamian E, Ahmadi P, Habibollahzade A, Ashjaee M. Energy, exergy and exergoeconomic (3E) analyses and multiobjective optimization of a solar and geothermal based integrated energy system. Appl Therm Eng 2018;143:1011–22. [4] Lee I, Tester JW, You F. Systems analysis, design, and optimization of geothermal energy systems for power production and polygeneration: state-of-the-art and future challenges. Renew Sustain Energy Rev 2019;109:551–77. [5] Cheng SWY, Kurnia JC, Ghoreishi-Madiseh SA, Sasmito AP. Optimization of geothermal energy extraction from abandoned oil well with a novel well bottom curvature design utilizing Taguchi method. Energy 2019;188:116098. [6] Pan SY, Gao M, Shah KJ, Zheng J, Pei SL, Chiang PC. Establishment of enhanced geothermal energy utilization plans: barriers and strategies. Renew Energy 2019;132:19–32. [7] Darvish K, Ehyaei MA, Atabi F, Rosen MA. Selection of optimum working fluid for organic Rankine cycles by exergy and exergyeconomic analyses. Sustainability 2015;7:15362–83. [8] Gholamiana E, Habibollahzade A, Zare V. Development and multiobjective optimization of geothermal-based organic Rankine cycle integrated with thermoelectric generator and proton exchange membrane electrolyzer for power and hydrogen production. Energy Convers Manage 2018;174:112–25. [9] Astolfi M, La Diega LN, Romano MC, Merlo U, Filippini S, Macchi E. Techno-economic optimization of a geothermal ORC with novel “Emeritus” heat rejection units in hot climates. Renew Energy 2020;147:2810–21. [10] Sun Q, Wang Y, Cheng Z, Wang J, Zhao P, Dai Y. Thermodynamic and economic optimization of a double-pressure organic Rankine cycle driven by low-temperature heat source. Renew Energy 2020;147:2822–32. [11] Kolahi MR, Amidpour M, Yari M. Multi-objective metaheuristic optimization of combined flash-binary geothermal and humidification dehumidification desalination systems. Desalination 2020;490:114456. [12] Kazemi S, Nor MIM, Teoh WH. Thermodynamic and economic investigation of an ionic liquid as a new proposed geothermal fluid in different organic Rankine cycles for energy production. Energy 2020;193:116722. [13] Zhou Y, Li S, Sun L, Zhao S, Talesh SSA. Optimization and thermodynamic performance analysis of a power generation system based on geothermal flash and dual-pressure evaporation organic Rankine cycles using zeotropic mixtures. Energy 2020;194:116785. [14] Yu Z, Su R, Feng C. Thermodynamic analysis and multi-objective optimization of a novel power generation system driven by geothermal energy. Energy 2020;199:117381. [15] Ghaebi H, Namin AS, Rostamzadeh H. Exergoeconomic optimization of a novel cascade Kalina/Kalina cycle using geothermal heat source and LNG cold energy recovery. J Clean Prod 2018;189:279–96.
[16] Li Z, Khanmohammadi S, Khanmohammadi S, Al-Rashed AAAA, Ahmadi P, Afrand M. 3-E analysis and optimization of an organic Rankine flash cycle integrated with a PEM fuel cell and geothermal energy. Int J Hydrogen Energy 2020;45(3):2168–85. [17] Aali A, Pourmahmoud N, Zare V. Exergoeconomic analysis and multi-objective optimization of a novel combined flash-binary cycle for Sabalan geothermal power plant in Iran. Energy Convers Manage 2017;143:377–90. [18] Alirahmi SM, Dabbagh SR, Ahmadi P, Wongwises S. Multi-objective design optimization of a multi-generation energy system based on geothermal and solar energy. Energy Convers Manage 2020;205: 112426. [19] Wang L, Li H, Bu X. Multi-objective optimization of Binary Flashing Cycle (BFC) driven by geothermal energy. Appl Therm Eng 2020;166:114693. [20] Ahmadi MH, Mehrpooya M, Pourfayaz F. Thermodynamic and exergy analysis and optimization of a transcritical CO2 power cycle driven by geothermal energy with liquefied natural gas as its heat sink. Appl Therm Eng 2016;109:640–52. [21] Kazemi N, Samadi F. Thermodynamic, economic and thermoeconomic optimization of a new proposed organic Rankine cycle for energy production from geothermal resources. Energy Convers Manage 2016;121:391–401. [22] Wang J, Wang J, Dai Y, Zhao P. Thermodynamic analysis and optimization of a flash-binary geothermal power generation system. Geothermics 2015;55:69–77. € [23] Ozkaraca O, Kec¸ebas¸ A. Performance analysis and optimization for maximum exergy efficiency of a geothermal power plant using gravitational search algorithm. Energy Convers Manage 2019;185:155– 68. € [24] Ozkaraca O, Kec¸ebas¸ A, Demircan C. Comparative thermodynamic evaluation of a geothermal power plant by using the advanced exergy and artificial bee colony methods. Energy 2018;156:169–80. [25] Ehyaei MA, Ahmadi A, Assad MEH, Rosen MA. Investigation of an integrated system combining an organic Rankine cycle and absorption chiller driven by geothermal energy: energy, exergy, and economic analyses and optimization. J Clean Prod 2020;258:120780. [26] Porkhia S, Salehpour M, Ashraf H, Jamali A. Modeling and prediction of geothermal reservoir temperature behavior using evolutionary design of neural networks. Geothermics 2015;53:320–7. [27] Wu C, Wang SS, Jiang X, Li J. Thermodynamic analysis and performance optimization of transcritical power cycles using CO2-based binary zeotropic mixtures as working fluids for geothermal power plants. Appl Therm Eng 2017;115:292–304. [28] Samin MY, Faramarzi A, Jefferson J, Harireche Q. A hybrid optimisation approach to improve long-term performance of enhanced geothermal system (EGS) reservoirs. Renew Energy 2019;134:379–89. [29] Klein SA. Engineering Equation Solver. (EES). F-chart software; 2020. [30] Cengel YA, Boles MA. Thermodynamics: an engineering approach. New York: McGraw-Hill; 2006. [31] Dincer I, Rosen MA. Exergy: energy, environment and sustainable development. Elsevier Science; 2007. [32] Ahmadi P, Khanmohammadi S, Musharavati F, Afrand F. Development, evaluation, and multi-objective optimization of a multi-effect desalination unit integrated with a gas turbine plant. Appl Therm Eng 2020;176:115414. [33] Dincer I, Rosen MA, Ahmadi P. Optimization of energy systems. 1st ed. Wiley; 2017.
Multiobjective optimization Chapter 17
[34] Khanmohammadi S, Khanmohammadi S, Khorasanizadeh H, Afrand M. Exergy and exergoeconimic analysis and multi-criteria optimisation of 1 MW installed CCHP system (a case study in Kashan University). Int J Exergy 2020;32(1):45–61. [35] Khanmohammadi S, Rahimi Z, Khanmohammadi S, Afrand M. Triple-objective optimization of a double-tube heat exchanger with
291
elliptic cross section in the presence TiO2 nanofluid. J Therm Anal Calorim 2020;140:477–88. [36] Holland JH. Adaptation in natural and artificial systems. Ann Arbor: University of Michigan Press; 1975. [37] Mitchell M. An introduction to genetic algorithms. MIT Press; 1998.
Chapter 18
Optimization of geothermal power plants with MATLAB Washima Tasnina and Lalit Chandra Saikiab a
School of Electrical Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu, India, b Department of Electrical Engineering, National Institute
of Technology, Silchar, Assam, India
Nomenclature f Tyz T Dfw DPDw Hw Dw Rw bw Bw Tggeo Ttgeo Kpw Tpw Prw Ttw Tgw Krw Trw TgST TtST DPtie PI
error y-z
nominal frequency (Hz) synchronizing coefficients (T12, T23, T13) simulation time (s) incremental change in frequency of area w (Hz) incremental load change in area w (pu MW) inertia constant of area w (s) DPDw/Dfw (pu MW/Hz) governor speed regulation parameter of area w (Hz/ pu MW) area frequency response characteristics (AFRC) of area w (¼ Dw + 1/Rw) frequency bias coefficients of area w (1, 2, 3) the steam governor time constant of geothermal power plant (s) the steam turbine time constant of geothermal power plant (s) 1/Dw (Hz/pu MW) 2Hw/ (fwDw) (s) rated power of area w (MW) steam turbine time constant of area w (s) steam governor time constant of area w (s) steam turbine reheat coefficient of area w (s) steam turbine reheat time constant of area w (s) steam governor time constant of the solar thermal plant (s) steam turbine time constant of the solar thermal plant (s) incremental change in tie-line power connecting between area w and area y (pu MW) p
Subscripts w
*
area number (1, 2, 3)
Superscripts optimum value
Abbreviations apf ACE
ACE participation factor area control error
AGC cpf CC CS DISCO DOF DPM FA FO FOCC GENCO GRC GTP I IDD ISO PI PID PIDD PSO SCA ST
18.1
automatic generation control contract participation factor cascade controller cuckoo search distribution company degree of freedom DISCO participation matrix firefly algorithm fractional order fractional-order cascade controller generating company generation rate constraint geothermal plant integral integral double derivative independent system operator proportional integral proportional integral derivative proportional integral double derivative particle swarm optimization sine-cosine algorithm solar-thermal
Introduction
In terms of the generation of electricity, geothermal energy has been quite a reliable resource. It has many desirable attributes such as low emissions and a small carbon footprint. Moreover, it can also provide continuous baseline electrical power, unlike other renewable sources such as solar and wind, which are intermittent in nature. Geothermal energy is available anywhere in the world, but currently it is being drilled up to 4 km due to economic constraints. The deepest hole to date is 12.3 km in Russia [1, 2]. As per the Geothermal Energy Association, in more than 24 countries, 13.3 GW of operating capacity was estimated based on the data from January 2016. By around 2021, the geothermal energy industry is expected to reach about 18.4 GW globally [3]. ThinkGeoEnergy declared the United States to be in first position with 3567 MW in terms of the installed
Thermodynamic Analysis and Optimization of Geothermal Power Plants. https://doi.org/10.1016/B978-0-12-821037-6.00019-6 Copyright © 2021 Elsevier Inc. All rights reserved.
293
294 PART III Optimization of geothermal power plants
capacity, and the Philippines had 1868 MW online capacity as of January 2017. Geothermal energy has been utilized in several fields for quite a long time. It is well known that it can be harnessed both as a source of electricity generation with the help of renewable energy along with being directly used for heating and cooling purposes. Automatic generation control (AGC) is another primitive technique playing an important role in the electricity generation of a power system because it always tries to keep a balance between the total power generated, the actual load demanded, and the accompanying losses. Any hindrances in this balance will automatically deviate the frequency as well as the tie-line powers from the schedule values [4]. The vertically integrated utility is the traditional structure of the power market, and it is named so because a single utility controls the power flow. The monopoly prevails entirely as generation is followed by transmission and distribution. Nevertheless, as the Federal Energy Regulatory Commission has put up the Notice of Proposed Rulemaking, this vertical arrangement has eventually been converted into a horizontal one. This horizontal structure comprises generating companies (often called GENCOs) and distribution companies (DISCOs), together with the transmission companies and an independent system operator (ISO). This scenario promoted a competitive spirit compared to the conventional arrangement and is termed the deregulated environment. In such a scenario, some rules need to be followed; the GENCOs have the liberty to decide whether they want to participate in AGC. Likewise, freedom is also given to the DISCOs for selecting any GENCO, whether it be from its own area or not, for matching the demanded load. ISO does the important job of supervising all the market players and their power transactions so that stable and smooth operation of the entire power system is ensured. Some of the other existing transactions are the poolco, bilateral, and area regulation contracts [5–7]. The area control error maintains the incremental power output (denoted as ACE in short). Again, the contribution of each GENCO is decided by the apf (which is the ACE participation factor). Another important element for deregulation studies is the DISCO participation matrix, which is formed by the contract participation factors (known as cpfs). The DISCO participation matrix is often abbreviated as DPM. Each of these terms and their related concepts are elaborately discussed in [8], and their practical implementations in the price-based scenario are distinctly described in [9]. In AGC, Donde et al. [9] are considered pioneering for analyzing and applying the deregulation concepts considering the simplest of the two-area thermal GENCOs. But as time passed, hydro, gas, diesel, and even nuclear were incorporated both singly or in combination, giving rise to
multisource systems. One such system has been presented by Mohanty et al. [10]. Although the trend of integrating other sources with the thermal units was prevalent for quite a long time, eventually it was felt that some of these were nonrenewable in nature, and some were damaging the Earth. Hence, for safety and survival as well as keeping in mind the energy requirements of the future, the focus was shifted to renewable resources, mainly due to the fact that there is much less possibility regarding their depletion because they are abundantly available. Even if they are depleted, easy replenishment is attainable. However, as of now, studies considering renewable resources—and those under the deregulated scenario—are quite limited and thus there are great prospects in terms of future research. Solar and wind energy are the most dominant ones among all the renewables, and their modeling and applications are well explained in [11]. Yatin et al. [12] discussed the modeling of the solar-thermal (ST) plant, and Das et al. [13] designed the wind simulator model. Although an emerging renewable source, geothermal is also not so far behind in terms of electricity generation. Heimisson in [14] extensively emphasized its control strategies. This was something to be investigated about the different sources required for the modeling of the power system. The other essential components shall be discussed sequentially. For controlling the frequency in quite a refined way, additional controllers are necessary that must be sturdy enough to ensure negligible ACE. Some commonly used conventional controllers such as integral (I) [8], proportional integral (PI), and proportional integral derivative (PID) [7] as well as some of their specific combinations such as integral double derivative (IDD) [10] and proportional integral double derivative (PIDD) [10] are being utilized in AGC. The degree-of-freedom controller concept was also brought into the picture by developing the 2DOF-PID [15] and the 2DOF-IDD [16]. Likewise, the fractional-order (FO) [17] and cascade controllers (CC) are also being analyzed together with their combination as a fractional-order cascade controller (FOCC), which has been developed quite recently. Now, a complimentary part of the controller is sought to be the optimization technique, which is essentially required for setting the controller gains to the optimum values. The traditional technique for setting the controller gains was by trial and error, which was quite a lengthy and hectic process. Hence, optimization techniques began to be used. In the AGC field, some of those are the firefly algorithm (FA) [18], the cuckoo search (CS) algorithm [16], particle swarm optimization (PSO) [19] and the gray wolf optimizer [12]. In this chapter, the convergence characteristics of several algorithms such as FA, CS, and PSO have been found and compared with the sine-cosine algorithm (SCA) [20]. Later, for the studies, the gains of the
Optimization of geothermal power plants with MATLAB Chapter 18
controllers and the respective essential parameters are optimized using the SCA itself. Thus, geothermal, solar thermal, and wind, along with the conventional thermal unit system, have been taken into the investigation in this aspect in the entire study. Broadly distinguishing between three unequal areas conventional and three unequal areas deregulated system, but first a simple nonequal two-area system with multiple sources in the conventional scenario is being considered for easy analysis. Hence, the objectives can be compiled as 1. To incorporate wind, solar-thermal, and a geothermal plant (GTP) together with thermal units with the generation rate constraint (GRC) in several areas, respectively, for the considered unequal area conventional and deregulated systems. 2. To optimize the gains of controllers I, PI, PID, and FOPI-FOPD using SCA for the considered system and to find the best controller. 3. To study the effect of the renewable sources considered in the model such as GTP, ST, and wind on the operation of the entire system, considering the already found finest controller. 4. To carry out the thermoeconomic analysis of GTP.
18.2 System description Combining the concept of both the conventional as well the deregulated environment, basically, three main systems are taken into consideration if seen broadly. First, a nonequal two-area system incorporating renewables and thermal power systems with a capacity ratio of Area 1/Area 2 ¼ 1/ 4. Second, a nonequal three-area conventional system possessing an area capacity in the ratio of 1/4/8 is considered. As said earlier, geothermal, solar-thermal, and wind are connected in Areas 1, 2, and 3, respectively, along with thermal being common in each area. Finally, the unequal three-area system of the second case, possessing an area capacity ratio of 1/4/8, is reconstructed under the deregulated scenario. Figs. 18.1–18.3 respectively depict the schematic model for each of the three considered systems separately. A GRC of 3%/min is considered for thermal plants. The parameters for the thermal plant are taken from [7] and those for solar and wind from [12, 13], respectively, as shown in Table 18.1. With the per-unit values being the same on their respective bases, a12 ¼ P1/P2 is considered for a two-area system and a12 ¼ P1/P2, a23 ¼ P2/P3, a13 ¼ P1/P3 for a three-area system, where P is the rated power of a particular area. The areas are interconnected, and their details are provided in Table 18.1. Further, Bw ¼ bw and Rw ¼ 4% droop [1]. The I, PI, and PID are considered secondary controllers for comparison. For the conventional systems, the operation
295
is analyzed, considering a 1% step load perturbation. The parameters of the controller are optimized with the help of SCA. Integral squared error (ISE), as represented in Eq. (18.1), is the cost function used for the entire optimization process with w, y, z as the area numbers from 1 to 3, and y 6¼ z. ðt n 2 o ðD fw Þ2 + D Ptie error yz dt (18.1) J¼ 0
18.2.1
Geothermal plant
As GTP is the prime focus, the modeling of its parameters needs to be in a quite detailed manner. From various literature and resources, it has been found that the operation of a geothermal plant is not very different from a conventional thermal plant without the reheating application, as it utilizes the steam beneath the Earth’s surface for power generation, which is already under intense pressure and temperature. The temperature of the fluid varies from around 140–250°C, and the pressure is more than 7 bar; to be specific, it is 7.4 bar at 180°C [14]. A GTP schematic is modeled likewise and portrayed in Fig. 18.4. The modeling details of the essential components are provided in the following subsections.
18.2.1.1 Modeling of the governor for a geothermal power plant D Pref and D o are the two inputs to the governor, and the combined effect of the increase in D Pref and the decrease in D o will finally lead to an increase in D Pg as [1], 1 DPg ¼ DPref Do MW R
(18.2)
Dxi is the change in position of the actuator valve, and it escalates with the rise in DPg and reduces on lowering the valve output, DPv. Hence, Dxi ¼ DPg DPv MW But,
(18.3)
ð DPv ¼ kH Dxi dt
(18.4)
kH, which is a constant, is dependent on the structure of the cylinder, orifice, and the pressure of the fluid inside the cylinder. Applying the Laplace transform to Eqs. (18.2) and (18.3) results in DPv 1 ¼ GG ðsÞ ¼ DPg Tggeo s + 1
(18.5)
where Tggeo ¼ k1H is the time constant of the governor.
FIG. 18.1 Two-area conventional schematic model.
Governor1
Area-1 – –
FOPI
+–
apf11
+ –
FOPD apf12
B1
+ – 1 R1
1 Tgt1.s+1
– –
FOPI
+–
apf21 FOPD
+ –
B2 1
1
Tggeo.s+1
Ttgeo.s+1
Governor2
Turbine2
Turbine3
1 Ttt2.s+1
1 Tgt2.s+1
KST Delay T Insolation ST.s+1 ST
– –
FOPI
+–
+ –
FOPD apf32
B3
+ – 1 R3
FIG. 18.2 Three-area conventional schematic model.
Geothermal plant
+ +
– +
KP1
2*pi*T12
+
TP1.s+1
s
–
PS1
a12
Reheat2
Kr2.Tr2.s+1 Tr2.s+1
´
GRC2
1 TtST.s+1
Governor4
Turbine4
Reheat3
1 Tgt3.s+1
1 Ttt3.s+1
Kp2
Tp1.s+1 s+1
Hydraullic pitch actuator
Tr3.s+1 Kp3 s+1
+ +
+ –
Tie12
Tie23 2*pi*T23
–
s
+
KP2
+
TP2.s+1
–
PS2
Tie13 2*pi*T13 s
+ + GRC3
Thermal plant
Kpc
Blade Datafit pitch response characteristics
Wind plant
a13 – + + –
KP3 TP3.s+1 PS3
SLP3
– +
a23
SLP2
Solar thermal plant
Kr3.Tr3.s+1
Tp2.s+1
Thermal plant
1 TgST.s+1
Turbine5
Governor5
apf31
Thermal plant
GRC1
Tr1.s+1
1
R2
Area-3
1
SLP1
+ –
apf22
Reheat1 Kr1.Tr1.s+1
Ttt1.s+1
Governor3
Area-2
Turbine1
+ –
DISC 02 cpf11
++
cpf21
++
cpf31
++
cpf41
++
cpf51
++
cpf61
++
DISC 04
cpf12
cpf13
cpf22
cpf23
cpf32
cpf33
cpf42
cpf43
cpf52
cpf53
cpf62
DISC 01
++ ++ ++
cpf15
cpf24
cpf25
cpf34
cpf35
cpf54
++
cpf63
cpf64
++
DISC 03
cpf16
++
++
DISC 05
cpf36
++
cpf55
+ + + + + + + + + + + +
cpf26
++
cpf45
cpf44
++
DISC 06
cpf14
cpf46 cpf56
++
cpf65
cpf66
++
+ + + +
Area-1 – –
FOPI
+–
FOPD
+ + + + –
apf2 B1
+
+ + + + –
apf1
1 R1
2*pi*T13 s
–
1 Tggeo.s+1
1 Ttgeo.s+1
Governor1
Turbine1
+
Geothermal plant
1 Tg1.s+1
1 Tt1.s+1
Kr1.Tr1.s+1
Governor2
Turbine2
Reheat1
–
+ + GRC
Tt1.s+1
Thermal plant
–
2*pi*T12 s
PS1 KP1
+ + + +
a12 –
– –
FOPI
+–
+ + + + – Insolation
apf3
FOPD
apf4 B2
1 R2
Area-3 – –
FOPI
+–
+ + + + –
apf5
FOPD
apf6 B3
+ + + + –
1 R3
FIG. 18.3 Three-area deregulated schematic model.
+ + + + –
KST TST.s+1
´ Delay
1 TgST.s+1
1 TtST.s+1
Governor3
Turbine3
Solar thermal plant –
ST
1 Tg2.s+1
1 Tt2.s+1
Kr2.Tr2.s+1
Governor4
Turbine4
Reheat2
Tr2.s+1
+ + GRC
Thermal plant
–
+
Tp1.s+1
TP2.s+1
s+1
Hydraullic pitch actuator
KP3 s+1 Datafit pitch response
1 Tg3.s+1
1 Tt3.s+1
Kr3.Tr3.s+1
Governor5
Turbine5
Reheat3
Tr3.s+1
Kpc Blade characteristics
GRC
Wind plant
Thermal plant
2*pi*T23 s a23
PS2 KP2 TP2.s+1
SLP
+ KP2
–
+ –
+ + –
SLP
PS3 KP3 TP3.s+1
+ + + +
–
+
Area-2
–
+
TP1.s+1
SLP
+
a13
–
+
+ –
+ –
+ –
Optimization of geothermal power plants with MATLAB Chapter 18
Applying the Laplace transform to Eq. (18.9),
TABLE 18.1 System parameters. Nominal values of model
f-60 Hz; system loading-50% Kpw ¼ 120 Hz/puMW; Tpw ¼ 20s; Twy ¼ 0.086 puMW/rad Dw ¼ 8.33 103 puMW/Hz; Bw ¼ bw ¼ 0.425 puMW/Hz Rw ¼ 2.4 puHz/MW, Hw ¼ 5 s a12 ¼ Pr1/Pr2 ¼ 1000/4000 ¼ 1/4 a23 ¼ Pr2/Pr3 ¼ 4000/8000 ¼ 1/2 a13 ¼ Pr1/Pr3 ¼ 1000/8000 ¼ 1/8
Thermal
Tgw ¼ 0.08 s; Krw ¼ 0.5; Trw ¼ 10s; Ttw ¼ 0.3 s
Wind
TP2 ¼ 0.041 s; KP2 ¼ 1.25; TP1 ¼ 0.6 s; KP3 ¼ 1.4; KPC ¼ 0.8
Solar-thermal
TST ¼ 1.8 s; KST ¼ 1.8; TgST ¼ 1 s; TtST ¼ 3 s
18.2.1.2 Modeling of the turbine for a geothermal power plant As we know, the continuity equation is given by, m_ in m_ out ¼ V
dr dW ¼ dt dt
(18.6)
indicating m_ as the steam flow rate in kg/s; V as the volume in m3; r as the steam density in kg/m3; and W as the steam weight in kg. Now, assuming that the pressure inside is proportional to the output flow rate as m_ out ¼
m_ o P P0
(18.8)
Now m_ in m_ out ¼ V
∂r dP ∂r P0 d m_ out dm_ out ¼V ¼ Ttgeo (18.9) ∂P dt ∂P m_ o dt dt
∂r P0 where Ttgeo ¼ V ∂P m_ o is the time constant of the turbine.
FIG. 18.4 Schematic model of GTP.
m_ in m_ out ¼ Ttgeo sm_ out
(18.10)
1 m_ out ¼ G T ðsÞ ¼ m_ in Ttgeo s + 1
(18.11)
Hence,
Hence, it can be observed that the basic governing equations for the governor and turbine modeling of a geothermal power plant are derived as GG ðsÞ ¼ Tggeo1s + 1 and GT ðsÞ ¼ Ttgeo1s + 1 with Tggeo as the time constant of the governor with its values around 0.1 s. Ttgeo is the time constant for the turbine possessing values in the range of 0.1–0.5 s [4]. During the entire study, the exact values for Tggeo and Ttgeo are optimized using the SCA technique. As nothing in the practical world is ideal, the same goes for the turbine. Hence, losses are an inevitable part of the process in the turbine. Upon entering the turbine, due to its expansion, the pressure and density of the steam reduce. Thus, the isentropic efficiency of the turbine is calculated, which is the ratio between the actual work done to the maximum work possible and is denoted by s in Eq. (18.16). During the operation cycle, the working fluid undergoes various phase changes such as from liquid to gas and also sometimes a mixture of both at some point in time. Hence, basic thermodynamics laws are applicable for GTP operation, the conservation of mass and energy to be specific. The quality of the steam of the gas phase in the mixture of the two can be given by Eq. (18.12) [1].
(18.7)
where m_ o is the flow rate, P0 is the rated pressure, and P is the steam pressure inside the container, in kPa. Considering the process to be isothermal, dr dP ∂r ¼ dt dt ∂P
299
x¼
m_ g m_ g + m_ l
(18.12)
m_ g and m_ l denote the flow rate of the gas and liquid, respectively. Several thermodynamic processes are going on within a geothermal plant, but they are generally considered to be ideal, and the losses are neglected. For maintaining the balance, the mass flow into any component must always be equal to the mass flow out, m_ in ¼ m_ out
(18.13)
Again, applying the first law of thermodynamics, the energy in is equal to the energy out, and the total energy in the
300 PART III Optimization of geothermal power plants
_ where h in (J/kg) is the specific working fluid is mh, enthalpy. Thus, m_ in hin ¼ m_ out hout + W_ + Q_
(18.14)
W_ is the work done per unit time in (W) and Q_ denotes the heat load in (W). The ratio between the output work done to the heat flow into the process is termed the thermal efficiency th as, th ¼
W_ Q_ Q_ out ¼ in _ Q in Q_ in
(18.15)
Similarly, the isentropic efficiency is defined as s ¼
W_ m_ in ðhin hout Þ ¼ _ W s m_ in ðhin hs,out Þ
(18.16)
W_ s is the power output of the turbine and hs, out is the outlet specific enthalpy [14, 21].
18.2.2
Solar thermal energy
Solar energy can be considered as the most utilized renewable energy, and it arose as a substitute for conventional energy resources. Solar thermal, solar photovoltaics, and dish Stirling technology are some of the different applications of solar energy. In this chapter, we are going to focus on solar thermal technology. In this technique, electric power is produced by concentrating the solar radiation to heat the working fluid, which in due time produces steam to drive the heat engine or the generator. Other than being intermittent in nature, solar bears some other disadvantages, including that it is not cost-effective as the per kW price is comparatively high. Similarly, the efficiency of the system is quite low [12]. One of the major advantages is that it can be integrated into the thermal energy storage systems that enable it to continue operation during cloudy weather or maybe during the night [19, 22–24].
18.2.3
Wind energy
Wind is another intermittent renewable technology holding several attributes in addition to eliminating the carbon dioxide emissions from fossil fuels and being indigenous in nature. A substantial amount of land is required to set up a wind farm; in addition, they also face criticism because of the tall structures that create hazards for aircraft. The speed of the wind at a certain instant is the deciding factor for wind turbine generator output. Several nonlinearities such as in structural design as well as rotor and blade configurations are present in the wind turbine system and are elaborately discussed in [13, 25, 26]. The wind model considered in this chapter is a constant output wind turbine and is referred from [13, 25, 26].
18.3
Optimization technique
Optimization has always been a multidisciplinary topic, and hence it has been defined and used differently by various researchers in their respective disciplines. However, speaking in general terms, optimization is a technology that ensures that certain predetermined objectives are achieved by utilizing the resources available in the system in the most efficient way possible. Optimization techniques thus help in arriving at real-time solutions for the challenges faced in the real world by exploration and exploitation and reaching for optimal solutions. The process of decision making, along with the quality of the decision, is taken into account. It all started in the 18th century with Newton and Lagrange, and most of the important developments in this area of optimization were seen after this phase. The two most significant components for an optimization process are the modeling and analysis of the optimization. Deriving a mathematical expression for the problem faced in the real world can be termed modeling, whereas the analysis part includes several stages for achieving the perfect result [27]. During the starting phases of the development of optimization technology, modeling was the prime focus for researchers, and the works primarily dealt with mathematical modeling. As the days have passed, gradually studies on optimization have begun to be carried out, emphasizing the analysis part; several algorithms have also been developed to date. Basically, these algorithms are referred to as optimization algorithms, and they are broadly classified into deterministic and stochastic algorithms. As the deterministic ones ensure a significant sequence of actions, the objective function and the design variables may possess certain similar values or follow the same route for the solution. But unlike the stochastic algorithms are based on randomness. Moreover, some hybrid form may also be developed, combining the essence of both the deterministic and stochastic methods [28]. A detailed classification of the optimization algorithm is demonstrated in Fig. 18.5 [27].
18.3.1
Sine-cosine algorithm
The SCA is a population-based algorithm that obtains the best solution with the help of a mathematical model that is designed based on sine and cosine functions. If Xi denotes a search agent, P is the best position of the solution or the destination to be reached, and Xti the current solution position in the tth iteration, the position updating function is governed by these two sets of equations as follows [18]. ) ( t Xi + r1 sin ðr2 Þ r3 Pti Xit if r4 < 0:5 t+1 Xi ¼ if r4 0:5 Xit + r1 cos ðr2 Þ r3 Pti Xiit (18.17)
Optimization of geothermal power plants with MATLAB Chapter 18
301
FIG. 18.5 Classification of optimization techniques.
where r1 is the direction of movement, r2 depicts the extent to which there must be movement either toward or outwards P; r3 is a parameter that allocates an arbitrary weight to the destination for emphasizing (r3 > 1) or deemphasizing (r3 < 1) the impact of the final position P for defining the distance, and r4 is for switching from the sine to cosine functions and vice versa. Any random number within the interval [0, 2p] can be assigned to r2 for performing the exploration, irrespective of the location being inside or outside. Hence, the ranges for exploration are considered as (1,2] and [2,1), and that for exploitation is considered between [1,1]. For balancing between exploration and exploitation, the sine and cosine functions in Eq. (18.17) can be adaptively varied using Eq. (18.18) with a being a constant, and t and T being the present and maximum iterations, respectively. a r1 ¼ a t (18.18) T The detailed parameter tuning method is followed in the next section, and the operating flowchart is also depicted in Fig. 18.6. The entire version of the algorithm with practical working examples is explained in [20].
Step 4: The value of a corresponding to the minimum ISE is selected and fixed. Step 5: Now keeping a fixed as selected in step 4 and i as the earlier value, n is changed to different values, say 5, 10, 15, 20. Step 6: The system is simulated for the ISE value in each case, and the value of n corresponding to the minimum value of ISE is chosen. Step 7: The third parameter, i is now changed, keeping intact a and n. In each change, the system is simulated again and again for the ISE value. Step 8: Finally, the value of i for the minimum ISE is also attained. Thus, the tuned parameters obtained are a ¼ 2, n ¼ 10, i ¼ 50.
START Initialize the search agents in the population Xi (i=1,2...m) Initialize maximum number of iteration, Tmax
while t