Renewable Energy Systems in Smart Grid: Select Proceedings of International Conference on Renewable and Clean Energy (ICRCE) 2022 (Lecture Notes in Electrical Engineering, 938) 9811943591, 9789811943591

Citation preview

Lecture Notes in Electrical Engineering 938

Mohan Lal Kolhe   Editor

Renewable Energy Systems in Smart Grid Select Proceedings of International Conference on Renewable and Clean Energy (ICRCE) 2022

Lecture Notes in Electrical Engineering Volume 938

Series Editors Leopoldo Angrisani, Department of Electrical and Information Technologies Engineering, University of Napoli Federico II, Naples, Italy Marco Arteaga, Departament de Control y Robótica, Universidad Nacional Autónoma de México, Coyoacán, Mexico Bijaya Ketan Panigrahi, Electrical Engineering, Indian Institute of Technology Delhi, New Delhi, Delhi, India Samarjit Chakraborty, Fakultät für Elektrotechnik und Informationstechnik, TU München, Munich, Germany Jiming Chen, Zhejiang University, Hangzhou, Zhejiang, China Shanben Chen, Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, China Tan Kay Chen, Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore Rüdiger Dillmann, Humanoids and Intelligent Systems Laboratory, Karlsruhe Institute for Technology, Karlsruhe, Germany Haibin Duan, Beijing University of Aeronautics and Astronautics, Beijing, China Gianluigi Ferrari, Università di Parma, Parma, Italy Manuel Ferre, Centre for Automation and Robotics CAR (UPM-CSIC), Universidad Politécnica de Madrid, Madrid, Spain Sandra Hirche, Department of Electrical Engineering and Information Science, Technische Universität München, Munich, Germany Faryar Jabbari, Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA, USA Limin Jia, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland Alaa Khamis, German University in Egypt El Tagamoa El Khames, New Cairo City, Egypt Torsten Kroeger, Stanford University, Stanford, CA, USA Yong Li, Hunan University, Changsha, Hunan, China Qilian Liang, Department of Electrical Engineering, University of Texas at Arlington, Arlington, TX, USA Ferran Martín, Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain Tan Cher Ming, College of Engineering, Nanyang Technological University, Singapore, Singapore Wolfgang Minker, Institute of Information Technology, University of Ulm, Ulm, Germany Pradeep Misra, Department of Electrical Engineering, Wright State University, Dayton, OH, USA Sebastian Möller, Quality and Usability Laboratory, TU Berlin, Berlin, Germany Subhas Mukhopadhyay, School of Engineering & Advanced Technology, Massey University, Palmerston North, Manawatu-Wanganui, New Zealand Cun-Zheng Ning, Electrical Engineering, Arizona State University, Tempe, AZ, USA Toyoaki Nishida, Graduate School of Informatics, Kyoto University, Kyoto, Japan Luca Oneto, Department of Informatics, Bioengineering., Robotics, University of Genova, Genova, Genova, Italy Federica Pascucci, Dipartimento di Ingegneria, Università degli Studi “Roma Tre”, Rome, Italy Yong Qin, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China Gan Woon Seng, School of Electrical & Electronic Engineering, Nanyang Technological University, Singapore, Singapore Joachim Speidel, Institute of Telecommunications, Universität Stuttgart, Stuttgart, Germany Germano Veiga, Campus da FEUP, INESC Porto, Porto, Portugal Haitao Wu, Academy of Opto-electronics, Chinese Academy of Sciences, Beijing, China Walter Zamboni, DIEM - Università degli studi di Salerno, Fisciano, Salerno, Italy Junjie James Zhang, Charlotte, NC, USA

The book series Lecture Notes in Electrical Engineering (LNEE) publishes the latest developments in Electrical Engineering - quickly, informally and in high quality. While original research reported in proceedings and monographs has traditionally formed the core of LNEE, we also encourage authors to submit books devoted to supporting student education and professional training in the various fields and applications areas of electrical engineering. The series cover classical and emerging topics concerning: • • • • • • • • • • • •

Communication Engineering, Information Theory and Networks Electronics Engineering and Microelectronics Signal, Image and Speech Processing Wireless and Mobile Communication Circuits and Systems Energy Systems, Power Electronics and Electrical Machines Electro-optical Engineering Instrumentation Engineering Avionics Engineering Control Systems Internet-of-Things and Cybersecurity Biomedical Devices, MEMS and NEMS

For general information about this book series, comments or suggestions, please contact [email protected]. To submit a proposal or request further information, please contact the Publishing Editor in your country: China Jasmine Dou, Editor ([email protected]) India, Japan, Rest of Asia Swati Meherishi, Editorial Director ([email protected]) Southeast Asia, Australia, New Zealand Ramesh Nath Premnath, Editor ([email protected]) USA, Canada: Michael Luby, Senior Editor ([email protected]) All other Countries: Leontina Di Cecco, Senior Editor ([email protected]) ** This series is indexed by EI Compendex and Scopus databases. ** More information about this series at https://link.springer.com/bookseries/7818

Mohan Lal Kolhe Editor

Renewable Energy Systems in Smart Grid Select Proceedings of International Conference on Renewable and Clean Energy (ICRCE) 2022

Editor Mohan Lal Kolhe Department of Engineering and Science University of Agder Grimstad, Norway

ISSN 1876-1100 ISSN 1876-1119 (electronic) Lecture Notes in Electrical Engineering ISBN 978-981-19-4359-1 ISBN 978-981-19-4360-7 (eBook) https://doi.org/10.1007/978-981-19-4360-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Editorial

Smart grid forms an important part of the ‘energy transition’ to integrate renewable and clean energy sources in energy network for making low-carbon energy systems. Substantial amounts of clean energy technologies are going to be integrated in the near future for development of sustainable energy system. The book ‘Renewable Energy Systems in Smart Grid,’ Select Proceedings of International Conference on Renewable and Clean Energy (ICRCE) 2022, special volume of book series Springer Lecture Notes in Electrical Engineering (e-ISSN: 1876-1119) contains peer-reviewed select papers from ICRCE 2022. The objective of this special volume is to present advancements in the smart grid technologies for integration of renewable energy systems for contributing in development of sustainable electrical energy system. The book chapters are exploring cutting-edge solutions and best practices for renewable and clean energy technologies for achieving the UN’s SDG7 to ‘ensure access to affordable, reliable, sustainable and modern energy for all.’ This book presents innovative grid integration technologies for techno-economic operation of renewable and clean energy technologies (e.g., solar photovoltaic, wind energy, electro-chemical energy conversion and integration, etc.). It covers key aspects on energy conversion systems related to renewable energy technologies and their grid integration, techno-economic power dispatching from the distributed environmental-friendly energy sources, electrical energy network operation with increasing penetration of renewable energy sources, energy efficiency, e-mobility, including machine learning applications for intelligent operation of energy systems, etc. The key objective of book is to educate the readers on how renewable energy technologies can be integrated with energy conversion processes for achieving net zero targets in real-world applications. The book will serve as a useful reference for graduate students, academicians, industry professionals and policy makers interested in exploring the potential of energy technologies in development of sustainable energy system. Mohan Lal Kolhe

v

ICRCE 2022 Committee

Conference Advisory Committees Saifur Rahman (IEEE Life Virginia Tech, USA Fellow) Longya Xu (IEEE Fellow) The Ohio State University, USA Ronghai Qu (IEEE Fellow) Huazhong University of Science and Technology, China

Conference General Chair Ryuichi Yokoyama (IEEE Waseda University, Japan Fellow)

Conference Co-chair Mingcong Deng (IEEE Senior Tokyo University of Agriculture and Technology, Member) Japan Lin Meng Ritsumeikan University, Japan

Conference Program Co-chairs Hee-Je Kim Pusan National University, South Korea Jinhua She Tokyo University of Technology, Japan

vii

viii

ICRCE 2022 Committee

Conference Steering Co-chairs Muhammad Arfin Khan Lodhi Texas Tech University, USA Mohan Lal Kolhe University of Agder, Norway

Conference Publicity Chairs Akira NISHIMURA Mie University, Japan Koki Ogura Kyushu Sangyo University, Japan

Conference Technical Committees Chew Tin Lee Shogo Nishikawa Rupp Carriveau Tarlochan Sidhu Janaka Ekanayake Adem Akpınar Diego Bellan Ashwani K. Gupta Ahmed M. El-Nahas Mihaela Popescu Roberto Pfuyo Dinesh Kumar Sharma Wee-Jun Ong Pierluigi Siano Sohrab Mirsaeidi Qiuwei Wu Rania Rushdy Moussa Bahareh Kamranzad Swellam W. Sharshir Pirat Khunkitti Ekkachai Sutheerasak Muhammad Imran Khan Amir M. Fathollahi-Fard Mary Thornbush Mouloud Denaï Jun Zang Mahdi Majidniya

Universiti Teknologi Malaysia, Malaysia Nihon University, Japan University of Windsor, Canada University of Ontario Institute of Technology, Canada University of Peradeniya, Sri Lanka Bursa Uludag University, Turkey Politecnico di Milano, Italy University of Maryland, USA Menoufia University, Egypt University of Craiova, Romania National Technological University of South Lima, Peru IIMT University Meerut (Uttar Pradesh), India Xiamen University Malaysia, Malaysia University of Salerno, Italy Beijing Jiaotong University, China Technical University of Denmark, Denmark British University in Egypt, Egypt Kyoto University, Japan Kafrelsheikh University, Egypt Khon Kaen University, Thailand Burapha University, Thailand Hamad Bin Khalifa University, Qatar Université du Québec, Canada University of Guelph, Canada University of Hertfordshire, UK University of Bath, UK Université de Lorraine, France

ICRCE 2022 Committee

Takuji Matsumoto I-Hsien Liu Fabio Bisegna Ozcan Atlam Wenxian Yang Jalpa Thakkar Smith Eiamsa-ard Chathura Ranasinghe

ix

Central Research Institute of Electric Power Industry, Japan National Cheng Kung University, Taiwan Sapienza University of Rome, Italy Kocaeli University, Turkey Newcastle University, UK UPL University of Sustainable Technology, India Mahanakorn University of Technology, Thailand University of Moratuwa, Sri Lanka

Co-sponsored by

Technical Sponsored by

Supported by

x

ICRCE 2022 Committee

Published by Lecture Notes in Electrical Engineering

Contents

Renewable Energy Engineering and Technology Solar Tower Continuous Saturated Steam Generation . . . . . . . . . . . . . . . . Yimin Deng, Kong Webin, Huili Zhang, Raf Dewil, and Jan Baeyens On the Limitations of Machine Learning (ML) Methodologies in Predicting the Wake Characteristics of Wind Turbines . . . . . . . . . . . . . Mohan Kumar Gajendran, Ijaz Fazil Syed Ahmed Kabir, Shantanu Purohit, and E. Y. K. Ng

3

15

Status and Opportunities of Biogas Power Plants in Thailand and a Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Piyanart Sommani and Anchaleeporn Waritswat Lothongkum

25

Comprehensive Risk Evaluation of International Wind Farm Project in Intertidal Area Under EPC Mode Based on Matter-Element Extension Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wu Fan, Huang Jiuqiang, Peng Dan, Chen Zhi, Wu Shuqian, and He Qing

33

Forthcoming Opportunities for Obtaining Energy from New Renewable Sources in Romania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Violeta-Monica Radu, Iustina Popescu Boaja, Alexandru Anton Ivanov, George Dinca, and Robert Szabo

45

Thermal Engineering and Energy-Saving Technology Evaluation of Stratification in Thermal Energy Storages . . . . . . . . . . . . . . Ioannis Sifnaios, Adam R. Jensen, Simon Furbo, and Jianhua Fan Simulation of Integrated Gasification Combined Cycle (IGCC) and Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xue Sun, Zhen Yang, and Yuanyuan Duan

57

71

xi

xii

Contents

Cooperative Optimization of System Parameters and Heat Exchanger Structure for Geothermal Organic Rankine Cycle . . . . . . . . . . Jian Li, Zhen Yang, Yuanyuan Duan, and Zitao Yu

85

Thermodynamic Evaluation of CCS Waste Heat Recovery by Organic Rankine Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ran Li, Zhen Yang, and Yuan-Yuan Duan

99

Effects of Initial Pressure on the Explosion Temperature Peak of Ethanol-Air Mixture and the Time Difference Between Temperature and Pressure Peak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Xiaoyao Ning, Xuehui Wang, and Jian Wang Environmental and Chemical Engineering Preparation and Photocatalytic Activity of Bi2 WO6 /RGO Composite Photocatalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Ting Lin, Haiyan Fu, Yicheng Wu, Tian Chai, Guoxin Su, and Shuguang Wang Study on Carbon Emission Accounting of Technological Processes in Synergistic Treatment and Disposal of Sludge and Food Waste . . . . . . 127 Changhao Xiao and Jieying Chen Preparation of Cotton Straw Based Multi-pore Biomass Charcoal, Characterization and Electrochemical Properties . . . . . . . . . . . . . . . . . . . . . 139 Jing Tao Dai, Ying Yang, Wen Xuan Zheng, and Li Na Wang Dynamic Gas Control Strategy for Mode Switching in a Proton Exchange Membrane Unitized Regenerative Fuel Cell . . . . . . . . . . . . . . . . 157 Mengdi Guo, Zhonghao Zhang, Zhonghao Yu, Siyue Yao, Diankai Qiu, and Linfa Peng

Renewable Energy Engineering and Technology

Solar Tower Continuous Saturated Steam Generation Yimin Deng, Kong Webin, Huili Zhang, Raf Dewil, and Jan Baeyens

Abstract Concentrated solar energy is an unlimited source of heat, and therefore widely applied in solar electricity generation or in the application of industrial heat. The intermittent nature of solar energy imposes the use of a heat storage when a continuous operation is considered. Solar irradiance is only available during the daytime. Preliminary experiments were carried out in a 30 kW th pilot-scale solar tower located at QinHuangDao (China), a site with moderate annual solar irradiance of about 5 kWh/m2 day, while better locations exceed 10 kWh/m2 day. The present paper assesses the scale-up of the system to an industrial steam generation. The plant layout, process components and economics are evaluated. Solar steam becomes increasingly interesting when the Direct Normal Irradiance (DNI) at the selected location increases. Savings in comparison with a coal-fired boiler can vary between 10 and 50%. Keywords Concentrated solar technology · Steam generation · Plant layout · Economics · Prospects

1 Introduction Concentrated solar energy is gaining increasing interest in solar power generation and in the application of industrial heat. Due to the intermittent nature of solar energy, a heat storage is required when a continuous operation is considered. Y. Deng · R. Dewil · J. Baeyens (B) Process and Environmental Technology Lab, Department of Chemical Engineering, KU Leuven, Sint-Katelijne-Waver, Belgium e-mail: [email protected] K. Webin · H. Zhang College of Life Science and Technology, Beijing University of Chemical Technology, Beijing, China J. Baeyens Beijing Advanced Innovation Centre for Soft Matter Science and Engineering, Beijing University of Chemical Technology, Beijing, China © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_1

3

4

Y. Deng et al.

Preliminary experiments were carried out in a 30 kW th pilot-scale solar tower located at QinHuangDao (CN). The average annual Direct Normal Irradiance (DNI) is only about 5 kWh/m2 day. Earlier reports were published [1, 2] and confirm the very good efficiencies obtained in solar heat capture, storage and use [3–5]. The present paper will investigate the scale-up of the system to an industrial steam generation of 4 ton/h of saturated steam at 7 bar. All essential equipment items will be described and designed. An economic assessment will demonstrate that the generated steam will be available at 10.6 e/ton, which is 10.2% cheaper than if generated by coal or biomass, and 2.8 times cheaper than when generated by a natural gas-based boiler. For locations with higher DNI, solar steam production costs will significantly decrease.

2 Process 2.1 Process Layout The process is a scale-up version of the experimental pilot-scale rig at QinHuangDao (CN), where the process parameters were investigated and the technical viability proven (see Fig. 1). Solar rays reflected by heliostats (1), and concentrated to heat on the central HFB (Horizontal fluidized bed) solar receiver [5] (6) on top of the solar tower (2). To limit the heat loss of the solar receiver, a cavity (7) is constructed and protected by high temperature resistant insulation boards. Powders from the cold storage are transported to the feeding silo (5) on top of the solar receiver by (4). Cold

Fig. 1 Proposed Plant layout

Solar Tower Continuous Saturated Steam Generation

5

Table 1 Monthly average direct solar irradiation energy in Qinhuangdao (kWh/m2 /day) Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

6.06

6.66

6.79

6.62

6.06

4.94

4.03

4.32

5.61

5.51

5.17

5.35

powders are heated by indirect heat transfer in the HFB solar receiver. Hot powders flow into the hot storage silo (8). For steam generation, condensate (from users and stored in a condensate water tank (11)) and make-up water (from a make-up water tank (12) to compensate for blow-down bleed), are pumped to a separated fluidized bed steam generator (9) and/or the back-up fuel steam generator (10). Make-up water (max. ratio 2%) is pre-treated in (13) and demineralization, polyphosphates, antifoam agent, and O2 scavenger are added. Tank (14) feeds the water to the steam generator. Generated steam passes into a liquid-steam separator to ensure that only saturated steam is delivered. The steam is transported to the users. The hot gas exhaust of the HFB solar receiver and fluidized bed steam generator is de-dusted by a high efficiency hot air sintered metal filter [6] and reused as the hot air in both HFB solar receiver and fluidized bed steam generator. Although the horizontal HFB is selected, a vertical heat capture receiver can also be used [7–9].

2.2 Equipment Selection The boiling point of 7 bar saturated steam is 165 °C (Tb ). The temperature of condensate water (Tc ) from the users is supposed to be 15 °C below the boiling point of the saturated steam, and is hence 150 °C. The evaporation latent heat of water at 7 bar is 2066 kJ/kg. The heat load (Q) to produce 4 ton/h saturated steam (7 bar) can be calculated by Eqs. (1) and (2). Q is 2366 kW. g

Q = Fwater (Hm,sensible + △l Hm )

(1)

Hm,sensible = 4.2 × (Tb − Tc )

(2)

Further to the pilot scale results, the energy efficiency from receiver-concentrated solar energy to heat storage in hot powders is 80% (ηm−r ), and the heat transfer efficiency from hot powders to hot water in the fluidized bed steam generator is 95% (η f s ). The heat loss of the total system is considered as 5%.

6

Y. Deng et al.

2.3 Heliostats The daytime hours and DNI values in Qinhuangdao are listed in Tables 1 and 2. In the daytime, the first and last one hours are the transition at sunset and sunrise, and the heliostats are defocused during this 2-h period. The initial design is meaning to generate all steam only by solar energy, which means that the HFB solar receiver must collect enough heat in the daytime to also cover the night time. The heat required in the HFB solar receiver is hence calculated by Eq. (3) to include 5% heat loss, and the required heliostat area is calculated by Eq. (4). To account for non-sun and/or night time steam production, the heat to be collected needs to be about 2*Q, i.e. about 4.7 MW th. Q htr = (

Q 24h × 1.05 )× ηfs tdt − 2h

Ah f =

(3)

Q htr ηm−r × D N I

(4)

Because the DNI level and daytime hours during the year change in different months, the required heliostat area is also a function of the time under scrutiny. Figure 2 shows the calculation result of required heliostat area in Qinhuangdao during a year. Table 2 Monthly average daytime in Qinhuangdao (hour) Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

9.68

10.7

11.9

13.2

14.3

14.9

14.7

13.7

12.5

11.1

10

9.4

3.0

Heliostate area(104 m2)

2.5 2.0 1.5 1.0 0.5 0.0

Jan

Feb Mar Apr May Jun

Jul

Aug Sep

Month Fig. 2 Different required heliostat field area for different months

Oct Nov Dec

Solar Tower Continuous Saturated Steam Generation

7

In QinHuangDao, the weather is rather unstable and there can be frequent cloudy or rainy days. So back-up fuel is very important for the concentrated solar utilization plant, especially in the onset months of the year. In the initial design, the required heliostat area is considered as the average value (17,500 m2 ). The cost of all heliostats is hence calculated as 1.4 106 e at a conservative price of 80 e/m2 of heliostat (including sun tracking system, and installed). From a literature survey, the worldwide price of heliostats is about 75–95 $/m2 (NREL [10], Stellio [11], Abengoa [12], Cobra [13]). The area of each heliostat is between 75 and 150 m2 .

2.4 HFB Solar Receiver From the on-sun experimental result, the heat transfer coefficient from the wall to the fluidized bed in the HFB solar receiver is minimum 500 W/m2 °C when using olivine particles of ~60 µm. The temperature difference between the receiver wall and the fluidized bed inside the HFB solar receiver is ~200 °C. According to Eq. (5), the heat transfer area of HFB solar receiver can be calculated, with the logarithmic temperature difference, △Tm,htr calculated according to common heat exchange principles. Ahtr =

h htr

Q htr × △Tm,htr

(5)

The inlet and outlet temperature of powder in HFB solar receiver are 200 °C and 600 °C, respectively. The heat capacity of olivine powder is 1.2 kJ/kg °C. The required flow rate of powders can be calculated by Eq. (6). Q htr = Fhtr × C p,htr × (Tout − Tin )

(6)

The height of each HFB is 0.6 m with 0.5 m of bed operating bed height. The width of HFB is 0.2 m, and the required total length of the HFB can be calculated as 94 m. 12 parallel HFB solar receivers of 7.8 m length each are stacked, hence respecting the recommended receiver height/length ratio of ~1. The total hourly powder flow rate is calculated as 35.3 ton/h, and the powder flow for each HFB solar receiver is 2.94 ton/h. The bulk density of powders in the HFB solar receiver is 1700 kg/m3 , so the flow velocity of powders in each solar receiver is ~0.3 cm/s that gives a residence time of ~44 min [4]. The superficial air velocity of the HFB solar receiver is controlled at 5Umf (0.05 m/s), and the total air flow rate in the HFB solar receiver is calculated as 0.96 m3 /s (3460 am3 /h, or 1080 Nm3 /h). The length and height of the cavity are designed as 1.2 times of sizes of the HFB solar receiver, and the depth of the cavity is 1 m. The feeding silo on top of the HFB solar receiver can contain 15 ton.

8

Y. Deng et al.

2.5 Hot Storage and Cold Storage Hoppers Hot powders produced in the daytime can be stored in a hot storage hopper and are immediately used in the boiler during daytime. The maximum loading capacity of the hopper is 85% of its total volume. The non-sun storage time is considered as 10 h. According to the calculated powder flow rate, the volume of the hot storage can be calculated as ~245 m3 . A compound insulation of 0.3 m thickness is installed to limit heat losses. The volume of the cold storage is the same as that of the hot storage, also insulated with 0.15 m thick combined Al2 O3 -refractory and ceramic fibre.

2.6 Fluidized Bed Steam Generator While heliostats are working, the inlet and outlet temperature of hot powders are 200 °C and 600 °C, respectively. The inlet and outlet temperature of water/steam are 150 °C and 165 °C, respectively. The quantity of steam generated by the steam generator is 4 ton/h and the required collected heat capacity is 2350 kW. With a heat transfer coefficient from hot powders to steam at 500 W/m2 °C according to the on-sun experimental result, Eq. (7) determines the heat transfer area of the tube bundle in the fluidized bed steam generator, for a the fixed logarithmic temperature, △Tm . It is calculated as 27.6 m2 . The heat transfer coefficient can reach 2000 W/m2 °C if finned tubes are used [14]. Afs =

Q fs h f s × △T m

(7)

The inner diameter of the steam pipes in the fluidized bed steam generator is 40 mm with 2 mm wall thickness. The density of 7 bar steam is ~3.67 kg/m3 . According to the steam flow rate in the steam generator (4 ton/h), the number of parallel pipes is 40. According to the heat transfer area of the steam generator, the length of each steam pipe is ~5 m. This will also be the length of the fluidized bed. The steam pipes will be triangularly staggered with a pitch of 120 mm (3*dout ). A bed height of 0.5 m will therefore certainly be acceptable. For a superficial air velocity of 5 Umf (0.05 m/s), the air flow rate of the steam generator fluidized bed is ~0.25 m3 /s (900 am3 /hr or ~580 Nm3 /hr).

2.7 Back-Up Fuel Steam Generator In the event of insufficient hot powder in the storage, coal, electricity, biomass and natural gas can be used as the potential back-up fuels in the system [15–17]. The heat capacity and estimated steam price for these 4 back-up fuels are listed in Table 3.

Solar Tower Continuous Saturated Steam Generation

9

Table 3 Characteristics of different types of backup fuels Backup fuel

Price

Heat capacity

Steam price (e/ton steam)

Coal

82 e/ton

25,000 MJ/ton

~11.2

Electricity

0.08 e/kWh

3600 kJ/kWh

~52.6

Biomass

41 e/ton

18,000 MJ/ton

~12.2

Natural gas

685 e/ton

50,240 MJ/ton

~39.0

For biomass, the pretreatment and dewatering costs are considered as 0.6 and 0.9 of original fuel cost, respectively. For electricity and natural gas, no pretreatment cost applies. The duty of the back-up fuel steam generator should generate 4 t/h of saturated steam, and is calculated as Eq. (8). Q is calculated as Eq. (1). A boiler efficiency, ηbs , of 0.9 is included. Q bs =

Q ηbs

(8)

3 Economic Assessment 3.1 Investment Estimate The investments were complied based upon quotes from Chinese manufactures and given in Table 4. It is clear that the heliostats represent about 44% of the total investment. This will be considerably lower in locations with higher DNI than QHD.

3.2 Production Cost It is understood that a general maintenance will be performed during sun-lean weeks. For the electricity cost, the power of two elevators is 40 kW, the power of all air blowers is 10 kW, the power of all water pumps is 5 kW, and 5 kW is foreseen for other plant items. Four people are needed in the plant to monitor the system and periodically wash the mirrors. Maintenance and repairs are estimated at 2.5% of the investment cost. The production cost are hence summarized in Table 5. The total steam generation capacity is 4 ton/h or 34,000 ton/year. According to the DNI statistics, the ratio of solar steam to the total steam generation is 94.5%, and

10

Y. Deng et al.

Table 4 Capital cost of concentrating solar steam generation plant Items

Investment (103 e)

Note

Heliostats, installed

1400

Including sun tracking system

Solar tower structure

400

50 m, steel structure

HFB solar receiver, with cavity

200

AISI 321 stainless steel

Feeding silo

30

15 ton

Hot storage silo

100

AISI 321 stainless steel, with insulation

Cold storage silo

70

Normal stainless steel

Fluidized bed steam generator

90

Back-up fuel steam generator

150

Make-up water treatment system

50

Powder elevator

80

35 t/h, 50 m height

Powder elevator

40

35 t/h, 20 m height

Insulations

100

Installation

300

Land field cost

-

Transportation

200

Total

3210

Not accounted for

Table 5 Production cost of concentrating solar steam generation plant (per year of 8500 h) Items Electricity Labor Make-up water treatment Depreciation (25 years) Maintenance and repair Total

Cost (103 e/year) 40 100 10 130 81 361

the remaining 5.5% are best generated by coal or biomass. The average production cost is 10.6 e/ton of steam.

3.3 Sensitivity Analysis It is expected that the cost of the heliostats will diminish over the coming years. It is however important to assess the influence of cost changes, as illustrated in Fig. 3. Figure 4 compares the impact of using different heat-sources as back-up heat.

Solar Tower Continuous Saturated Steam Generation

11

15 14

Production cost (€)

13

Heliostat Central tower Electricity Labour

12 11 10 9 8 7

-50%

-25%

0

25%

50%

Price Fluctuation Fig. 3 Sensibility of production costs (e/ton/steam) for variable cost factors

Steam production cost (€/ton)

20 Back-up fuel Coal Electricity Biomass Natural gas

18

16

14

12

10 1.8

2.0

2.2

2.4

2.6

2.8

3.0

Heliostat area (104 m2)

Fig. 4 Influence of the applied back-up fuel

It should moreover be remembered that QinHuangDao is a pessimistic location for this type of solar plants. Other regions in China have a solar irradiance in excess of 9 kWh/m2 -day, implying that the heliostat area and cost can be decreased by a factor of nearly 2. This will reflect itself in an investment cost reduction of the heliostats by a proportional amount. Due to the depreciation rate of 25 years, the influence on the steam production cost reduction will about 2 e/ton—steam.

12

Y. Deng et al.

4 Conclusions Concentrated solar energy is used in solar electricity generation and in the application of industrial heat. Due to the intermittent nature of solar energy, a heat storage is required when a continuous operation is considered. Preliminary experiments were carried out in a 30 kW th pilot-scale solar tower located at QinHuangDao (CN), where the average annual Direct Normal Irradiances (DNI) is only about 5 kWh/m2 day. Excellent efficiencies confirm the potential use of the system toward solar heat capture, storage, and use. The present paper investigated the scale-up of the system to an industrial steam generation of 4 ton/hr of saturated steam at 7 bar. Based upon the design and costing of the different equipment items, an economic assessment will demonstrate that the generated steam will be available at 10.6 e/ton, which is 10.2% cheaper than if generated by coal or biomass, and 2.8 times cheaper than when generated by a natural gas-based boiler. For locations with higher DNI, solar steam production costs will significantly decrease.

References 1. Li S, Kong W, Zhang H, Sabatier F, Ansart R, Flamant G, Baeyens J (2019) The fluidized bed air heat exchanger in a hybrid Brayton-cycle solar power plant, p 140002 2. Baeyens J, Zhang H, Kong W, Dumont P, Flamant G (2019) Solar thermal treatment of nonmetallic minerals: the potential application of the SOLPART technology. In: AIP conference proceedings, p 180002 3. Kang Q, Flamant G, Dewil R, Baeyens J, Zhang HL, Deng YM (2019) Particles in a circulation loop for solar energy capture and storage. Particuology 43:149–156 4. Kong W, Wang B, Baeyens J, Li S, Ke H, Tan T, Zhang H (2018) Solids mixing in a shallow cross-flow bubbling fluidized bed. Chem Eng Sci 187:213–222 5. Zhang H, Kong W, Tan T, Baeyens J (2017) High-efficiency concentrated solar power plants need appropriate materials for high-temperature heat capture, conveying and storage. Energy 139:52–64 6. Li S, Baeyens J, Dewil R, Appels L, Zhang H, Deng Y (2021) Advances in rigid porous high temperature filters. Renew Sustain Energy Rev 139:110713 7. Zhang H, Baeyens J, Cáceres G, Degrève J, Lv Y (2016) Thermal energy storage: recent developments and practical aspects. Prog Energy Combust Sci 53:1–40 8. Zhang H, Kong W, Tan T, Gilles F, Baeyens J (2017) Experiments support an improved model for particle transport in fluidized beds. Sci Rep 7:10178 9. Deng Y, Sabatier F, Dewil R, Flamant G, Le Gal A, Gueguen R, Baeyens J, Li S, Ansart R (2021) Dense upflow fluidized bed (DUFB) solar receivers of high aspect ratio: different fluidization modes through inserting bubble rupture promoters. Chem Eng J 418:129376 10. Zolan A, Hamilton W, Wagner M, Liaqat K (2011) Solar field layout and aimpoint strategy optimization 11. Stellio Heliostat, https://solarimpulse.com/solutions-explorer-fr/stellio-heliostat. Accessed 21 Nov 2021 12. Abengoa, https://www.abengoa.com/web/en/. Accessed 21 Nov 2021 13. Cobra, https://www.grupocobra.com/en/. Accessed 21 Nov 2021 14. Zhang H, Benoit H, Perez-Lopèz I, Flamant G, Tan T, Baeyens J (2017) High-efficiency solar power towers using particle suspensions as heat carrier in the receiver and in the thermal energy storage. Renew Energy 111:438–446

Solar Tower Continuous Saturated Steam Generation

13

15. Baeyens J, Li S, Zhang H, Dewil R, Flamant G, Ansart R, Deng Y (2020) Bio-energy carriers as back-up fuel in hybrid solar power plants. IOP Conf Ser Earth Environ Sci 544:12012 16. Zhang HL, Baeyens J, Degrève J, Cacères G (2013) Concentrated solar power plants: review and design methodology. Renew Sustain Energy Rev 22:466–481 17. Baeyens J (1980) Modelling approach to the effect of equipment scale on fluidized bed heat transfer data. J Powder Bulk Solids Technol 4:1–9

On the Limitations of Machine Learning (ML) Methodologies in Predicting the Wake Characteristics of Wind Turbines Mohan Kumar Gajendran, Ijaz Fazil Syed Ahmed Kabir, Shantanu Purohit, and E. Y. K. Ng Abstract Machine Learning (ML) algorithms have been more prevalent in recent years, and they are being used to tackle complicated issues across a broad range of fields. Wind energy is not an exception, as ML has recently been applied to wind turbine blade design, wake velocity and wake turbulence intensity prediction, and even wind farm optimization. The immense learning ability of ML models enables them to be trained to predict and regress a complex relationship with a high degree of accuracy. However, data for testing ML models often originate from the same rotor simulation used for training, with just slight variations in operating conditions. This research aims to investigate the generalizability of ML-based wake prediction models, i.e., whether ML can correctly predict wake properties using data from a different wind turbine that was not taken for training. This investigation’s observation shows that a generalized ML wake model requires training data from multiple turbines with a wide range of operating conditions. In addition, advanced regularization, complex loss functions, and ML methods that focus on capturing the physics (such as Physics Informed Artificial Neural Networks (PINN) and symbolic regression) can be utilized. Keywords Machine learning · Wind turbines · Wake velocity · Turbulence intensity

1 Introduction Wind energy has grown rapidly and steadily in recent years as one of the most prominent renewable energy resources on the worldwide energy market. It is typical for wind turbines to be clustered together owing to space constraints. The wake M. K. Gajendran School of Computing and Engineering, Civil and Mechanical Engineering Department, University of Missouri-Kansas City, Kansas City, MO 64110, USA I. F. S. A. Kabir · S. Purohit · E. Y. K. Ng (B) School of Mechanical and Aerospace Engineering, College of Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798 Singapore, Singapore e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_2

15

16

M. K. Gajendran et al.

Fig. 1 Wake effect behind a wind turbine under ABL [1]

effect may cause significant power losses to downstream turbines in a wind farm, lowering total power output [1, 2]. Thorough knowledge and precise forecasting of turbine wakes may significantly increase the efficiency of the wind farm and lead to a more efficient turbine layout design [2]. Thus, effective (both accurate and acceptable prediction time) wake prediction is critical for a wind farm’s performance [2]. Figure 1 shows the wake effect behind a wind turbine and its velocity deficit. Straightforward analytical to complicated 3D Navier–Stokes (N–S) models have been generated to depict the wake velocity deficit [2]. Due to their low cost and time consumption, analytical models are still commonly utilized in practice, particularly in industries for wake velocity prediction [2, 3]. Previous research has shown that conventional analytical wake velocity models allow considerable variations compared to sophisticated CFD simulation results [2]. Several analytical wake velocity models would not compensate for the added turbulence intensity in the wake [2, 3]. CFD is now frequently utilized to precisely anticipate complicated flow physics. Complete wind farm CFD simulations using entire rotor geometries are not realistic because the refined mesh used to solve the boundary layer is computationally expensive and time-consuming [1–3]. In recent years, ML models have been created to bridge the gap between analytical models’ simplicity and speed and numerical simulations’ accuracy. These studies utilize specific rotor data from high-fidelity CFD simulations to build a ML model (training and test data sets are created from a same rotor). In general, as compared to analytical wake models, these ML models created on training data have a smaller error percentage. Due to the fact that analytical wake models are already imprecise, ML models with a lower error percentage are hence highly desirable. This paper discusses the significance of generalization in order for these models to be applicable in the future for a variety of wind turbines operating under a practical range of operating conditions. The flow chart in Fig. 2 summarizes the present methodology for generating wake models for wind turbines

On the Limitations of Machine Learning (ML) Methodologies …

17

Fig. 2 Flow chart visualizing existing methodology in developing ML-based wake models and proposed additional evaluation criteria to achieve generalized ML model

using machine learning algorithms, as well as the future work required to generalize the machine learning-based wake models.

2 Different ML-Based Wake Velocity Models In our previous work [2], the authors described several conventional analytical wake models, ranging from the very ancient Jensen model to the more contemporary B-P model. The analytical wake velocity models need the following input parameters: Freestream input velocity (U∞ ) in m/s, Distance downstream of the turbine(x) in m, Thrust coefficient (CT ), Local wake radius (rW ) in m, Rotor diameter (D) in m, Hub height (zhub ) in m and Ambient turbulence intensity (Ia ). Table 1 highlights

18

M. K. Gajendran et al.

Table 1 Input parameters for wake velocity deficit and wake turbulence intensity models. [2] Uniform inflow

ABL inflow

U∞ or Uref (ambient uniform velocity) (m/s)

Uhub or Uref (ambient velocity at hub height) (m/s)

x (downstream distance behind the rotor) (m) x (downstream distance behind the rotor) (m) CT (rotor thrust coefficient) (-)

CT (rotor thrust coefficient) (-)

rw (local wake radius) (m)

z−z hub D (for upper half of the wake, z ≥ zhub ) z hub −z (for lower half of the wake z < zhub ) D

z is height above the ground (m) zhub is hub height of the rotor (m) D (Rotor Diameter) (m)

D (Rotor Diameter) (m)

Ia (ambient uniform turbulence intensity) (%) Ihub (ambient turbulence intensity at hub height) (%) zo (roughness length of the terrain)

the general input parameters required depending on whether the inflow is uniform or ABL. Notably, not all parameters are needed by all models; the authors have already enumerated the parameters required by each analytical model in their earlier work [2]. Table 2 summarizes the anticipated output wake characteristics based on the input parameters provided in Table 1. There is presently relatively limited research on wake velocity prediction using ML. Table 3 summarizes the details of how the authors mapped the input and output parameters in their prior work [2] when predicting the wake characteristics. Table 4 summarizes current ML-based wake models, including the wind turbine rotor utilized, the ML method employed, whether direct or indirect rotor modeling was used, and the required input and computed output variables. Table 2 Output parameters predicted. [2] Uniform inflow

ABL inflow

Wake velocity deficit △U(m/s)

Wake velocity deficit △U (m/s)

Turbulence intensity Iwake (%)

Turbulence intensity Iwake (%)

Table 3 Parameters considered for model development. [2] Model

Parameters Considered for mapping

Wake velocity deficit for uniform flow

△U = f(U∞ , x, CT , rW , D, Ia )

Wake turbulence intensity for uniform flow

Iwake = f(U∞ , X, CT , rW , D, Ia )

Wake velocity deficit for ABL inflow (both for upper and lower half)

△U Uhub

= f(Uhub , x, CT , z, zhub , D, Ihub , zo )

Wake turbulence intensity for ABL inflow (both for upper and lower half)

IWake = f(Uhub , X, CT , Z, Zhub , D, Ihub , Zo )

Wake radius for ABL inflow

Wake radius = f(Uhub , X, CT , Zhub , D, Ihub , Zo )

On the Limitations of Machine Learning (ML) Methodologies …

19

Table 4 Recent ML based wake models S. no

Reference

ML algorithm

Wind turbine rotor used

Direct or indirect rotor CFD modeling

Input variables Output variables

1

[2]

Genetic Programming (GP)

NREL Phase VI wind turbine

Direct full All parameters rotor CFD as listed in modeling Table 1 with both uniform and ABL inflows

Wake velocity deficit and wake turbulence intensity

2

[3, 4]

Artificial Neural Network (ANN) model based on the backpropagation (BP) algorithm

Vestas V80 2 MW wind turbine

Indirect (Actuator disk with rotation (ADM-R) and RANS coupling model)

Inflow wind speed turbulence intensity at hub height

Spatial velocity deficit and added turbulence kinetic energy

3

[5]

Support Vector Regression (SVR), eXtreme Gradient Boosting (XGBoost), ANNs

NREL Phase VI wind turbine

Direct full rotor CFD modeling with uniform inflow

All parameters Wake velocity as listed in [5] deficit and wake turbulence intensity

4

[6]

(BP) ANN

Enercon 2.3 MW wind turbine

Indirect (ADM-NR) and CFD coupling model

Incoming wind velocity

Wake velocity

5

[7]

Convolutional Neural Network (CNN) autoencoder model

Vestas V27 wind turbine

Indirect (Actuator Surface (AS) and LES coupling model)

Instantaneous velocity fields

Time-averaged windwise velocity field

Figure 3 illustrates the distinction between traditional programming and machine learning methodologies. It’s worth noting that programs produced using ML are entirely dependent on data. Furthermore, it should be emphasized that these research articles were primarily concerned with establishing the foundation for developing a new wake model based on ML and concluded that more assessment would be conducted in the future.

20

M. K. Gajendran et al.

Fig. 3 Difference between traditional programming and machine learning algorithm

3 Data Generation for ML Algorithm from Direct or Indirect Rotor Modeling More data with varied scenarios is necessary to generalize ML algorithms. For this reason, CFD simulation is commonly employed. Full rotor CFD simulation is accurate but takes days and needs a lot of computational resources. Reduced-order turbine models like Actuator Disk (AD), Actuator Line (AL), and Actuator Surface (AS) are commonly utilized to cut computational costs. Instead of simulating the rotor, these methods calculate the aerodynamic force or pressure drop along the rotor using external analytical rotor aerodynamics tools like BEM and couple with CFD [8]. The fundamental flaw of indirect rotor models is that it generally ignores unsteady and 3D effects like stall delay, which increases lift coefficient [8]. The forces estimated by BEM with no or inaccurate correction models will be erroneous when fed into CFD. Incorrect force values may affect wake velocity prediction. To properly estimate the wake velocity, rotor aerodynamic analysis such as BEM [8, 9] must be upgraded for unsteady and 3D effects as future work.

4 Discussions As previously stated, the application of ML to wake prediction is recent, and most work has focused on the framework. To respect the other research, the authors solely self-evaluated their model [3] for uniform inflow generated from CFD data of NREL Phase VI wind turbine with their CFD results of MEXICO rotor [10] for wind speed 24 m/s in this comparison. Also, unlike other ML models, ours is an empirical model that anybody can use, while other models need at least basic ML understanding. The machine learning model’s predictive performance is demonstrated in Fig. 4 by comparing its predictions for the NREL Phase VI wind turbine operating at 25 m/s at downstream distance 15D to the corresponding CFD results, as well as

On the Limitations of Machine Learning (ML) Methodologies …

21

Fig. 4 Comparison of ML model with CFD results for NREL [2] and MEXICO rotor

its prediction for the MEXICO rotor operating at 24 m/s at downstream distance 15D to the corresponding CFD results. There is a mismatch between ML and CFD findings when using rotors of different designs. This study illustrates the necessity for diverse training data from different turbines to generalize ML-based wake prediction algorithms.

5 Conclusions and Future Works This research extends on prior work [1, 2] by emphasizing the limits of widely used analytical wake prediction models and the need for CFD in forecasting complicated 3D wake characteristics, particularly those associated with various ABLs. Regrettably, CFD is a time- and resource approach. Recent advances in machine learning have enabled the development of algorithms that bridge the gap between basic analytical wake models and accurate CFD models. The purpose of this research is to emphasize the generalization limitations inherent in machine learning. When evaluating data from the same rotor, machine learning models perform better than when testing data from different rotors. For example, as shown in Fig. 1, the authors’ machine learning model using NREL phase VI wind turbine data shows an average inaccuracy of 0.4% for NREL compared to 3.7% for the MEXICO rotor. Validation of data from another rotor is required for future machine learning-based wake models. Consequently, the authors intend to create a new model based on the results of one rotor and evaluate it using the data of different rotors in different ABLs in the near future. The authors intended to conduct CFD assessments of the wind turbines mentioned below under various ABLs and operating circumstances, provided that the authors got geometrical and experimental results from relevant research organizations. The wind turbines are:

22

• • • • • •

M. K. Gajendran et al.

NREL Phase VI wind turbine MEXICO rotor NREL 5 MW wind turbine SANDIA Seri 8 wind turbine Berlin Research turbine AVATAR 10 MW reference wind turbine.

Training data with diverse design and operating ranges can be created by examining a number of wind turbines, which can assist in generalization. If design data for other wind turbines not listed above are available, the authors also aim to consider them. The authors have applied ML-based methods on various complex systems [11– 14] and intend to focus on advanced regularization and ML methods that focus on capturing the physics, such as Physics Informed Artificial Neural Networks (PINNs) and Symbolic Regression with a simple functional form to develop more accurate and realistic ML-based wake models. Another significant difficulty with ABL inflow is that the wake is asymmetrical, and the velocity varies with altitude. As a result, the author intends to construct a three-dimensional wake model that is generalizable and representative of reality.

6 Contributions M.K.G and I.F.S.A: (Equal Contribution) Conceptualization, Data Curation, Formal analysis, Visualization, Investigation, Writing—original draft. S.P: Data Curation, Writing—editing. Ng. E.Y.K: Supervision, Writing—review editing.

References 1. Kabir IFSA, Ng EYK (2019) Effect of different atmospheric boundary layers on the wake characteristics of NREL Phase VI wind turbine. Renew Energy 130:1185–1197 2. Kabir IFSA, Safiyullah F, Ng EYK, Tam VW (2020) New analytical wake models based on artificial intelligence and rivalling the benchmark full-rotor CFD predictions under both uniform and ABL inflows. Energy 193:116761 3. Ti Z, Deng XW, Yang H (2020) Wake modeling of wind turbines using machine learning. Appl Energy 257:114025 4. Ti Z, Deng XW, Zhang M (2021) Artificial Neural Networks based wake model for power prediction of wind farm. Renew Energy 184:405–420 5. Purohit S, Ng EYK, Kabir IFSA (2021) Evaluation of three potential machine learning algorithms for predicting the velocity and turbulence intensity of a wind turbine wake. Renew Energy 184:405–420 6. Luo Z, Luo W, Xie J, Xu J, Wang L (2021) A new three-dimensional wake model for the real wind farm layout optimization. Energy exploration & exploitation, 01445987211056989

On the Limitations of Machine Learning (ML) Methodologies …

23

7. Zhang Z, Santoni C, Herges T, Sotiropoulos F, Khosronejad A (2022) Time-averaged wind turbine wake flow field prediction using autoencoder convolutional neural networks. Energies 15(1):41 8. Kabir IFSA, Ng EYK (2017) Insight into stall delay and computation of 3D sectional aerofoil characteristics of NREL phase VI wind turbine using inverse BEM and improvement in BEM analysis accounting for stall delay effect. Energy 120:518–536 9. Abdulqadir ASW, Ng EYK, Narasimalu S, Bahuguni A (2019) An unsteady stall-delay methodology for floating offshore wind turbines. WSEAS Trans Fluid Mech, ISSN/E-ISSN: 1790-5087/2224-347X, 14(16):142–153 (2019) 10. Purohit S, Kabir IFSA, Ng EYK (2021) On the Accuracy of uRANS and LES-based CFD modeling approaches for rotor and wake aerodynamics of the (New) MEXICO wind turbine rotor Phase-III. Energies 14(16):5198 11. Jahmunah V, Ng EYK, San TR, Acharya UR (2021) Automated detection of coronary artery disease, myocardial infarction and congestive heart failure using GaborCNN model with ECG signals. Comput Biol Med 134:104457 12. Gajendran MK, Khan MZ, Khattak MAK (2021) ECG classification using deep transfer learning. In: 2021 4th international conference on information and computer technologies (ICICT). IEEE, pp 1–5 13. Khan MZ, Gajendran MK, Lee Y, Khan MA (2021) Deep neural architectures for medical image semantic segmentation. IEEE Access 9:83002–83024 14. Mohan Kumar G, Fernando DX, Kumar RM (2013) Design and optimization of de Lavel nozzle to prevent shock induced flow separation. Adv Aerosp Sci Appl 3:119–124

Status and Opportunities of Biogas Power Plants in Thailand and a Case Study Piyanart Sommani and Anchaleeporn Waritswat Lothongkum

Abstract Energy crisis and environmental pollution problems have drawn attention to increase renewable energy resource utilization and replace fossil fuels worldwide. For several decades, Thailand relies on fossil fuels in electricity generation and so on. The Ministry of Energy, Thailand complied the Development of Energy Plans to Thailand Integrated Energy Blueprint (TIEB) in 2020 for sustainable energy development in addition to overcome environmental problems. This case study shows energy recovery from wastes for self-sustain biogas power plants from chicken meat production. The onsite biogas power plant targets at a capacity of 1.5 MW at 24 h for every workday. The chicken manure can yield a biogas output of 128–134 m3 /ton with the rate of electricity per biogas at 2.10 kWh/m3 . Based on the project period for 15 years, the payback period, net present value (NPV), and internal rate of return (IRR) are 8.2 years, about 1,705,000 USD, and 8.86%, respectively. In general, the crucial success factor of biogas power plant installation mainly goes to concrete collaboration among the government and private sectors. Keywords Chicken manure · Biogas power plants · Case study

1 Introduction For decades, the energy demand in Thailand, particularly in terms of the electricity demand, keeps steadily rising. Based on the Energy Balance of Thailand 2019 Report [1], the total amount of electricity consumption for the whole country has been constantly increased from 181,377 GWh in 2015 to 203,714 GWh in 2019 which affects the electricity generation, as shown in Fig. 1a. Figure 1b shows that the amount of electricity consumption in the industrial sector during 2015–2019 has accounted for approximately 40% of the total electricity consumption. In 2015, natural gas and coal were main fuels for the electricity generation (Fig. 1c); however, the decrease P. Sommani (B) Department of Electrical Engineering, School of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand e-mail: [email protected] A. W. Lothongkum Department of Chemical Engineering, School of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_3

25

26

P. Sommani and A. W. Lothongkum

Fig. 1 Plotted from Thailand electricity data from 2015 to 2019 [1]: a electricity consumption and generation, b electricity consumption in each sector, and c proportion of fuel consumption for electricity generation

in non-renewable natural resource consumption such as natural gas, oil, and coal was observed in 2019. Natural gas contributed 59.5% of the total fuel consumption, followed by coal 16.5%, fuel oil and diesel oil 0.1%, as well as the renewable energy sources 23.9%, i.e., agricultural wastes, municipal solid wastes (MSW), biogas, and other energy sources including black liquor and residual gas from production processes. Less fossil fuel utilization for electricity generation has been resulted from active engagement and collaboration among all sectors, i.e., the government agencies, private sectors, and the general publics. This paper addresses current status of biogas power plants in Thailand and their opportunities from the government energy plans, as well as a case study of energy recovery via a biogas power plant of the chicken industry.

2 Thailand’s Energy Plans By considering the energy consumption and resource management, the Ministry of Energy, Thailand developed a long-term Energy Master Plan, namely the Thailand Integrated Energy Blueprint (TIEB) in 2015 [2, 3]. The main offices in charge are the Energy Policy and Planning Office (EPPO), and the Department of Alternative

Status and Opportunities of Biogas Power Plants …

27

Energy Development and Efficiency (DEDE). In brief, TIEB consists of a set of five energy plans which focus on three main objectives, i.e., energy security, economy, and ecology. Currently, Thailand’s energy policies and plans have been revised in accordance with the National Strategy 2018–2037 [4]; and therefore, five energy plans in TIEB have been consecutively amended in 2020 [5] as follows: • • • • •

the Power Development Plan 2018–2037 Revision 1 (PDP2018 Revision 1), the Energy Efficiency Plan 2018–2037 (EEP2018), the Alternative Energy Development Plan 2018–2037 (AEDP2018), the Gas Plan 2018–2037 (Gas Plan 2018), and the Oil Plan 2018–2037 (Oil Plan 2018).

The development of EEP, AEDP, Gas Plan, and Oil Plan shall be in line with the PDP [6]. It is noteworthy that the PDP, EEP and AEDP are key plans that help minimize the adverse environmental impacts. In particular, the AEDP2018 emphasizes on the promotion of renewable energy exploitation, e.g., solar energy, wind energy, small and large hydro power, biomass, biogas, municipal solid wastes, and geothermal energy, for electricity generation or otherwise the reduction of energy dependency on fuel import, in terms of natural gas, oil and, coal [7].

3 Biogas in Thailand and the Power Plants Biogas projects generally benefit both heat and power production from renewable energy sources and environmental impact minimization. Therefore, biogas related projects are widely supported by the EPPO and DEDE [8]. As Thailand is agricultural-based country; and thus, raw materials for biogas production are extensively available, such as agricultural wastes, livestock manure, food industry wastes, organic wastes from communities, or energy crops. In the initial stage, the biogas projects were targeted at household or retail farmer levels. The consulting and supporting services for the supervision of the design, installation, financial funding, and biogas operation were fully managed. Accordingly, several private sectors involved the biogas power generation market by using wastewater from industry, animal manure from livestock, and energy crops. From 2016 to 2019, the annual biogas power generated by wastewater, animal manure, and energy crops continuously increased from 434.9, 475.4, 505.2, and 530 MW, respectively, whereas the annual biogas heat was nearly constant at 593, 634, 634, and 634 ktoe, respectively [7, 9]. According to the PDP2018 Revision 1, by 2037 the total capacities of new commercial and community biogas power plants using wastewater, animal manure, and energy crops at 1,183 MW are expected [6], which encourage the installation of new biogas power plants by private sectors. The common organic sources obtaining from various industries and livestock species to produce biogas are summarized in the AEDP2018 [7]. Table 1 presents the estimated amount of biogas generation from different sources in Mm3 /year. From the Table,

28

P. Sommani and A. W. Lothongkum

Table 1 Estimated biogas generation from the industry and livestock sectors in 2018 [7] Industry

Quantity (No. of Plants)

Estimated amount of wastewater (Mm3 /year)

Estimated biogas generation (Mm3 /year)

Cassava starch

118

203

Palm oil

183

21

577

Ethanol

26

19

571

Rubber Food processing

92

20

149

3

Biogas obtained from industrial wastewater

1,618

53 5 2,824

Livestock

Quantity (Million Heads)

Estimated amount of manure (Mton/year)

Estimated biogas generation (Mm3 /year)

Swine

10

4

188

Beef and dairy cattle

6

8

316

Buffalo

1

8

66

Poultry

456

8

215

Biogas obtained from livestock manure

785

Total biogas obtained from industrial wastewater and livestock manure

3,609

Total biogas used for energy production

2,470

Remaining total potential biogas

1,139

the remaining amount of biogas (1,139 Mm3 /year) has high potential for energy production. Table 2 highlights the number of installed capacity and heat consumption of biogas generated in each region of Thailand, obtaining from the Thailand Alternative Energy Situation 2019 Report [9]. A total installed capacity of 498.5 MW from biogas power plants in Thailand in terms of on-grid system increased 2.6% from 2018, while the heat consumption remained the same as in 2018. More GIS data on biogas status generated by region can be accessed [8]. On-grid biogas power plant map can be accessed from DEDE (https://www.dede.go.th/ewtadmin/ewt/dede_web/ewt_news. php?nid=50267). Table 2 Installed capacity and heat consumption from biogas generated by region in 2019 [9] Region

On-grid installed capacity (MW)

Heat (ktoe)

13.15

118.82

Northeast

106.61

253.4

Central

121.78

232.17

South

256.96

29.93

Total

498.5

634.32

North

Status and Opportunities of Biogas Power Plants …

29

4 A Case Study of a Biogas Power Plant A case study of a biogas power plant in the chicken industry is presented. A chicken meat production company is located at the central part of Thailand. The company consists of three main parts as follows: (1) parental stock farms to produce eggs for hatchery operation; (2) hatchery operation to produce one-day chicken for the company’s broiler farms and contract farms for nurturing the chickens until broilers; and (3) slaughterhouse to produce fresh meat, marinated, and cooked products for domestic and export markets. It has been recognized that a large amount of waste is generated every day in the chicken industry. Types of wastes include chicken manure, feathers, feed, spilled water, process-generated wastewater, bedding, and dead chickens. In general, chicken manure can be used as fertilizer; however, environmental issues should be concerned on the nutrient accumulation in soil, eutrophication effects, and air pollution [10]. The Department of Industrial Works, Ministry of Industry reported that most of the energy consumption in the chicken industry contributed to electricity (80.9%) and heat (19.1%) [11]. Thus, the biogas power plant in this case study shows high potential in energy recovery and waste management [12–14]. In response to environmentally friendly and high business competitiveness, the chicken meat production company decided to produce electricity from biogas at a capacity of 1.5 MW for its own use at the slaughterhouse (24 h for every workday) to attain energy saving and environmental sustainability. The biogas power plant consists of two main systems: biogas production and cleaning system, and gas enginegenerator system, as shown in Fig. 2. Biogas-generated feedstock, i.e., chicken manure is collected at the receiving unit. The bedding materials are separated from feedstock at the bedding removal pond. The biogas production is carried out in three continuous stirred tank reactors (CSTRs) under anaerobic and mesophilic conditions at 25–40 °C. The anaerobic digesters are round-concrete tanks, which are covered by double layer-reinforced PVC membrane containing ultraviolet and precipitationresistant additives. The feedstock is anaerobic digested for about 30–40 days to generate raw biogas about 60%vol-methane concentration. The raw biogas is cleaned by biofiltration technique to remove hydrogen sulfide gas before feeding to the gas

Fig. 2 Simplified schematic diagram of the biogas-based power plant in this study

30

P. Sommani and A. W. Lothongkum

engine-generator system, while the residual digestate is transferred to a fertilizer unit to produce fertilizer. Clean biogas feeding on its demand at the electricity production capacity of 1.5 MW is scheduled. It is found that the chicken manure yields a biogas output of 128–134 m3 /ton, providing a rate of electricity per biogas at 2.10 kWh/m3 . The total investment cost of this case study biogas power plant is about 3.64 million USD. The financial analysis shows that the biogas power plant is feasible at 1.5 MW biogas electricity generation for 24 h of every workday, resulting in the company’s electricity cost saving. Based on the project period of 15 years, the payback period, net present value (NPV), and internal rate of return (IRR) are 8.2 years, about 1,705,000 USD, and 8.86%, respectively.

5 Conclusion The energy policies and plans by Thai government, and the National Strategy promote the status and opportunities of biogas power plants. Energy recovery from a wide variety of reliable renewable energy sources can replace fossil fuels. Waste-to-energy power plants help promote energy security, economy, and ecology. In this study, the biogas power plant of the chicken meat production company is proven commercially viable and enhances the business image. However, to achieve the successful sustainable development, the government shall determine the appropriate supporting policies to stimulate private sector investment. The project constraints associated with the leakages due to biogas pressure and corrosion problems should be further studied. Acknowledgements The authors appreciate the support from the School of Engineering, King Mongkut’s Institute of Technology Ladkrabang. We also thank Mr. Kittipich Monkhlang and the collaboration company for providing the data on biogas power plant.

References 1. Department of Alternative Energy Development and Efficiency (DEDE). In: Energy Balance of Thailand 2019 Report. DEDE, Bangkok (2019) 2. Energy Policy and Planning Office (EPPO). In: Thailand Integrated Energy Blueprint TIEB (2016). Accessed 20 Feb 2021. http://www.eppo.go.th/index.php/en/policy-and-plan/en-tieb/ tieb-pdp 3. Traivivatana S, Wangjiraniran W, Junlakarn S, Wansophark N (2017) Energy Procedia 138:399– 404 4. Royal Thai Government Gazette. In: National Strategy 2018–2037 (Book 135, Part 82 A, 13 Oct 2019) (In Thai) 5. Energy Policy and Planning Office (EPPO). In: News: EPPO Drives Energy Plans and Promotes Community Power Plants (2020). Accessed 20 Feb 2021. http://www.eppo.go.th/index.php/en/ component/k2/item/15538-news-energy190263. (In Thai)

Status and Opportunities of Biogas Power Plants …

31

6. Energy Policy and Planning Office (EPPO). In: Power Development Plan 2018–2037 Revision 1 (PDP2018 Revision 1) (2020). (In Thai) 7. Department of Alternative Energy Development and Efficiency (DEDE). In: Alternative Energy Development Plan 2018–2037 (AEDP2018) (2020). (In Thai) 8. Department of Alternative Energy Development and Efficiency (DEDE). In: Information and Database System for Monitoring and Evaluating Thailand Biogas Production (2014) Accessed 21 Feb 2021. http://biogas.dede.go.th/biogas/web_biogas/. (In Thai) 9. Department of Alternative Energy Development and Efficiency (DEDE). In: Thailand Alternative Energy Situation 2019 Report. DEDE, Bangkok (2019) 10. Jongbloed AW, Lenis NP (1998) J Anim Sci 76:2641–2648 11. Department of Industrial Works (DIW). In: Manual of Slaughterhouse and Chicken Meat Process Control. Chulalongkorn University Press, Bangkok (2009). (In Thai) 12. Deublein D, Steinhauser A (2008) Biogas from Waste and Renewable Resources: An Introduction. Wiley VCH, Weinheim 13. Khanal SK (2008) Anaerobic Biotechnology for Bioenergy Production: Principles and Applications. John Wiley and Sons Inc., Iowa 14. Abbasi T, Tauseef SM, Abbasi SA (2012) Biogas Energy. Springer, New York

Comprehensive Risk Evaluation of International Wind Farm Project in Intertidal Area Under EPC Mode Based on Matter-Element Extension Theory Wu Fan, Huang Jiuqiang, Peng Dan, Chen Zhi, Wu Shuqian, and He Qing Abstract Owing to the change of concept of energy industry development and the corresponding regional preferential policies in recent years, the renewable energy market with huge potential provides an incredible opportunity for Chinese energy companies in developing the international market and seeking the transformation of their strategy. This paper takes the international wind farm project in intertidal area (based on EPC mode) as the research object, and a risk evaluation system containing 4 s-level indicators and 13 third-level indicators has been established from the perspective of EPC contractor to evaluate the comprehensive risk level, where the ANP-Matter Element Extension Model is applied to carry out the comprehensive evaluation. Finally, one wind farm project in Vietnam is employed as the case study to verify the risk indicator system and assessment model established in this paper. And the main significance of this paper lies in that a comprehensive risk evaluation framework for the international wind farm project in intertidal area has been constructed from the EPC contractor’s viewpoint, which can provide a certain support for the decision maker when carrying out similar international wind project to identify the potential risks and take corresponding measurements during the marketing stage or project implementation period. Keywords ANP-matter element extension model · Risk evaluation · International wind farm project · Intertidal area · EPC

1 Introduction With the advancement of the “Belt and Road” strategy raised by the Chinese government in 2015, the Chinese power enterprises have become more and more active on the international stage and the market abroad now is becoming an important field for W. Fan (B) · H. Jiuqiang · P. Dan · C. Zhi · H. Qing PowerChina Huadong Engineering Co., Ltd, Hangzhou, Zhejiang, China e-mail: [email protected] W. Shuqian Hangzhou Industrial Investment Group Co., Ltd, Hangzhou, Zhejiang, China © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_4

33

34

W. Fan et al.

them to seek business transformation and find a new motivity for further development [1]. Compared with the traditional thermal and hydro project, the short construction period, simple working face and lower technical difficulty make the renewable energy project (such as solar power, wind farm and so on) become an important area for the Chinese energy enterprises to develop the international market and increase the market share. So far, the research on characteristics for both offshore wind farm project and onshore wind farm project is relatively sufficient. However, for the wind farm project located in intertidal area, which has properties of both these two types, the potential for corresponding technology and theory study at present stage is still huge. This kind of project refers to the wind farm located close to the coastline where intertidal is a common site geological condition, and the foundation of wind turbine in such area usually will be submerged by sea level during high tide. And such kind of wind farm project shows great economic advantages compared with the offshore wind farm project due to its lower cost, but we still would like to emphasize that the similar difficulties as the offshore wind farm project can never be ignored by both the investor and contractor, such as the poor site conditions, high anti-corrosion level. And for the countries or regions with vast coastlines, the intertidal area wind farm project can effectively save the land resources compared with the onshore wind farms. Meanwhile, the investment costs for such project is expected to be less than the offshore wind farm project. Based on the two above characteristics, the intertidal area wind farm project owes huge investment potential and market scale and is becoming an important direction for the further development of wind power industry. Therefore, this paper employs the intertidal area wind farm project as the research object and makes a comprehensive risk analysis for this kind of wind farm project from the perspective of EPC contractor. Furtherly, a risk evaluation indicator system and the corresponding evaluation model based on ANP- Matter Element Extension theory are also built to carry out the comprehensive risk identity and assessment works through 4 different dimensions (containing economic risk, performance risk, macro risk and force majeure risk), aiming at providing a certain decision support for Chinese power enterprises from the perspective of risk control when carrying out similar international wind farm projects and ensuring the overall risk level of the project under control, which is conductive to the benefit of the project or the enterprises [2].

2 Method 2.1 ANP The Analytic Network Process (ANP) is usually used to deal with the systems of complex interdependence or influence relationships by identifying and analyzing the influences between the elements to improve the reliability and accuracy of the output

Comprehensive Risk Evaluation of International Wind Farm Project …

35

of the system. During the weight determination progress by applying ANP method, the Rule of Nine will usually be applied to determine the relative importance between each indicator to realize the quantitative analysis for qualitative issues [3]. As the core part of using ANP to determine the weight of each indicator, solving the hypermatrix is the main challenge during this progress, which usually will take a long time to seek the results by manual calculation. Therefore, some software usually can be employed to promote the handle of such problem efficiently.

2.2 Matter-Element Extension Model The concept of matter-element analysis aims to handle the complex contradictory system by seeking the expansion possibilities and internal new rules, which is practical for evaluating the international EPC project with complex risks [4, 5]. The mains steps by applying matter-element analysis are as follows: Identify the classical domain, joint domain and the object to be evaluated.  ⎤ ⎡ ⎤ ⎡ N j C1 V j1 N j C1 aj1 bj1   Rj = Nj , Ci , Vji = ⎣ ... ... ⎦ = ⎣ ...  ... ⎦ Cn C jn Cn ajn bjn

(1)

where Rj is the classical domain; Nj is the level of j; c1 ,…, cn means the features of Nj ; vji represents the range of cii , i = 1,2,…,n ⎤ ⎡ ⎤ p c1 < ap1 , bp1 > p c1 vp1 ⎦ Rp = (p, Ci , Vpi ) = ⎣ · · · · · · ⎦ = ⎣ · · · ··· cn vpn cn < apn , bpn > ⎡

(2)

where p means different levels of the object to be evaluated; vpi is the joint domain, the range of cii , i = 1,2,…,n. ⎡

⎤ p0 c1 v1 R0 = ⎣ · · · · · · ⎦ cn vn

(3)

where p0 is the object to be evaluated; vi shows the observation value of indicator ci , i = 1,2,…,n. Weight determination of each indicator. The ANP method is employed in this paper to determine the weight of each index as stated above.

36

W. Fan et al.

Normalization. Divide the value of the object and classical domain by the right endpoint value of joint domain, then the result of normalization can be obtained as following: ⎡

⎤ p0 c1 v1 /bp1 R0  = ⎣ · · · · · · ⎦ cn vn /bpn ⎤ ⎡ Nj c1 < aj1 /bp1 , bj1 /bp1 > ⎦ Rj  = (Nj , Ci , Vji  ) = ⎣ · · · ··· cn < ajn /bpn , bjn /bpn >

(4)

(5)

Obtain the correlation degree. The value of correlation degree can be obtained by applying the following formula: CDj (p0 ) = 1 −

n

ωi Dij

(6)

i=1

D(v, Vji ) = |v − (a + b)/2| − (b − a)/2

(7)

where D shows the distance from the object to each classical domain and CD is the corresponding value of correlation degree. Determine the evaluation level. The final level can be evaluated according to the value of CDj obtained in previous step based on the maximum principle.

3 Comprehensive Risk Indicator System Combining the characteristics of international project under EPC mode and intertidal area wind farm project, a comprehensive risk evaluation index system has been constructed from the perspective of EPC contractor, which includes economic risk, performance risk, macro risk and force majeure risk [6–10]. The detail is shown in Table 1.

4 Case Study One 30 MW wind farm project in Vietnam (hereinafter call “Project A”) is employed as the case to verify the comprehensive risk indicator system and evaluation model of international wind farm project in intertidal area established in this paper. Project A, located in southern coastline of Vietnam, is a typical intertidal area wind farm project

Comprehensive Risk Evaluation of International Wind Farm Project …

37

Table 1 Risk indicator system for international intertidal area wind farm project under EPC mode Second-lever risk indicator

Third-level risk indicator

Economic risk (a)

Profit level (a1) Exchange rate fluctuation (a2) Commercial and Contract risk (a3) Cash flow (a4)

Performance risk (b)

Standard and regulation (b1) Duration risk (b2) Erection and Installation risk (b3)

Macro risk (c)

State relations (c1) Government efficiency (c2) Maturity of local market (c3)

Force majeure risk (d)

Risk of public event (d1) Geological condition risk (d2) Culture difference (d3)

Fig. 1 General layout of Project A located in intertidal area of Vietnam

with the capacity of 30 MW(6*5 MW), and the Chinese contractor provides a package of services for the investor under EPC mode, including engineering, procurement and construction. The general layout of Project A is shown in Fig. 1.

4.1 Weight Determination As mentioned above, the comprehensive risk assessment system established for intertidal wind farm project contains 4 s-level indicators and 13 third-level indicators.

38

W. Fan et al.

And considering the mutule influence between different indicators and the internal complexity, the ANP is applied in this paper for obtaining the weight of each risk indicator and the Super Decision can be used for the calculation of the hypermatrix. Figure 2 shows the internal networks of each indicator and final weight can be seen in Fig. 3.

Fig. 2 Internal network for each risk indicator

Fig. 3 Result of weight by applying ANP

Comprehensive Risk Evaluation of International Wind Farm Project …

39

Table 2 Observed value and the normalization result of each indicator for Project A No

Second-level indicator

Third-level indicator

Observed value

Normalization result

1

Economic risk (a)

a1

5.3

0.530

a2

8.84%

0.850

a3

4.8

0.480

2

3

4

a4

6.5

0.790

Performance risk (b)

b1

3.6

0.360

b2

4.1

0.410

b3

Type-5.0 MW

0.650

Macro risk (c)

c1

Comprehensive strategic partnership

0.100

c2

40.9a

0.591

c3

Mature(≥10 years)

0.100

d1

3.5

0.350

d2

7.2

0.720

d3

2.8

0.280

Force majeure risk (d)

Note source China Lianhe Credit Rating Co., Ltd,《Tracking Rating Report of the Socialist Republic of Vietnam》 a Data

4.2 Normalization of Data According to the comprehensive risk evaluation system established in this paper for intertidal wind farm project, the value or score of each index and the normalization result for Project A can be found in Table 2.

4.3 Comprehensive Risk Evaluation Based on Matter-Element Extension Model Division of risk level. Considering the characteristics of international wind farm project under EPC mode and main topic of this research, the comprehensive risk level of the object to be studied can be divided as following: Extremely low, Low, General, High and Extremely high [11]. Establish the Matter-element extension model. Taking both the characteristic of each indicator and the experts’ experience into consideration, the classic domain of each indicator under different risk levels can be obtained, where Rj means the classical domain under different risk levels, Nj is the corresponding risk level, R’ stands for the object to be evaluated in this paper, the details are as following:

40

W. Fan et al.

⎧ ⎧ ⎧ ⎫ ⎫ ⎫ N1 a1 (0, 0.3) ⎪ N2 a1 (0.3, 0.4) ⎪ N3 a1 (0.4, 0.5) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ a2 (0, 0.1) ⎪ a2 (0.1, 0.2) ⎪ a2 (0.2, 0.3) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.2) (0.2, 0.4) (0.4, 0.6) a a a ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 3 3 3 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.2) (0.2, 0.4) (0.4, 0.6) a a a ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 4 4 4 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.2) (0.2, 0.4) (0.4, 0.6) b b b 1 1 1 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.2) (0.2, 0.4) (0.4, 0.6) b b b ⎨ 2 ⎨ 2 ⎨ 2 ⎬ ⎬ ⎬ R2 = R3 = R1 = b3 (0, 0.2) b3 (0.2, 0.4) b3 (0.4, 0.6) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ c1 (0, 0.2) ⎪ c1 (0.2, 0.4) ⎪ c1 (0.4, 0.6) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.1) (0.1, 0.3) (0.3, 0.4) c c c ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 2 2 2 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.2) (0.2, 0.4) (0.4, 0.6) c c c ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 3 3 3 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0, 0.2) (0.2, 0.4) (0.4, 0.6) d d d 1 1 1 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ d d d (0, 0.2) (0.2, 0.4) (0.4, 0.6) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 2 2 2 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ⎩ ⎩ ⎭ ⎭ ⎭ d3 (0, 0.2) d3 (0.2, 0.4) d3 (0.4, 0.6) ⎧ ⎧ ⎧ ⎫ ⎫ ⎫ N4 a1 (0.5, 0.7) ⎪ N5 a1 (0.7, 1) ⎪ P  a1 (0.530) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ a2 (0.3, 0.6) ⎪ a2 (0.6, 1) ⎪ a2 (0.850) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.6, 0.8) (0.8, 1) (0.480) a a a ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 3 3 3 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.6, 0.8) (0.8, 1) (0.790) a a a ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 4 4 4 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.6, 0.8) (0.8, 1) (0.360) b b b 1 1 1 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.6, 0.8) (0.8, 1) (0.410) b b b ⎨ 2 ⎨ 2 ⎨ 2 ⎬ ⎬ ⎬  R R R4 = = = b3 (0.6, 0.8) b3 (0.8, 1) b3 (0.650) 5 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ c1 (0.6, 0.8) ⎪ ⎪ c1 (0.8, 1) ⎪ ⎪ c1 (0.100) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.4, 0.6) (0.6, 1) (0.591) c c c ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 2 2 2 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.6, 0.8) (0.8, 1) (0.100) c c c ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 3 3 3 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ d1 (0.6, 0.8) ⎪ d1 (0.8, 1) ⎪ d1 (0.350) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ (0.6, 0.8) (0.8, 1) (0.720) d d d ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ 2 2 2 ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ⎩ ⎩ ⎭ ⎭ ⎭ d3 (0.6, 0.8) d3 (0.8, 1) d3 (0.280) Obtain the distance between the object and each classical domain (shown in Table 3). Calculate the correlation degree. On the basis of the results indicated in Table 3 and the weight obtained above, the correlation degree for the risk level of Project A can be calculated as follows: CD1 = 1 −

13

ωi dij = 0.698

i=1

CD2 = 1 −

13

i=1

ωi dij = 0.838

Comprehensive Risk Evaluation of International Wind Farm Project …

41

Table 3 Distance between the object and each classical domain Indicator

d1

d2

d3

d4

d5

a1

0.230

0.130

0.030

0.030

0.170

a2

0.750

0.650

0.550

0.250

0.150

a3

0.280

0.080

0.080

0.120

0.320

a4

0.590

0.390

0.390

0.010

0.010

b1

0.160

0.040

0.040

0.240

0.440

b2

0.210

0.010

0.010

0.190

0.390

b3

0.450

0.250

0.250

0.050

0.150

c1

0.100

0.100

0.100

0.500

0.700

c2

0.491

0.291

0.191

0.009

0.009

c3

0.100

0.100

0.100

0.500

0.700

d1

0.150

0.050

0.050

0.250

0.450

d2

0.520

0.320

0.320

0.080

0.080

d3

0.080

0.080

0.080

0.320

0.520

CD3 = 1 −

13

ωi dij = 0.859

i=1

CD4 = 1 −

13

ωi dij = 0.878

i=1

CD5 = 1 −

13

ωi dij = 0.731

i=1

Determine the comprehensive risk level. From the result obtained above, the corresponding comprehensive risk level of Project A is “High” because the maximum correlation degree is CD4 (0.878).

5 Conclusion Compared with the offshore or onshore wind farm project, intertidal wind farm project can not only save the land resources but also reduce cost for the investor, which makes it become an important global develop tendency of wind power industry when carrying out the corresponding renewable energy planning works, especially for the countries or regions with vat coastlines, such as Vietnam, Malaysia and so on. However, for EPC contractor, who undertakes the engineering, supply and construction of the project at the price close to the onshore wind farm project, shall face

42

W. Fan et al.

almost the same risk of offshore wind farm. In other words, the risks and challenges faced by EPC contractor is improved significantly [12]. From this point of view, a comprehensive risk evaluation indicator system (including 4 s-level indicators and 13 third-level indicators) and the corresponding Matter-element Extension Model have been established to carry out the comprehensive risk analysis for the international intertidal wind farm project under EPC mode. And finally, the Project A in Vietnam is employed to carry out the case study to verify the indicator system and model constructed in this paper. Based on the distribution of each indicator’s weight, it can be seen that the weight of “Commercial & Contract risk”, “Profit level”, “Duration risk” and “ Erection & Installation risk” are higher than other risk indicators, which means these potential risks usually have greater impacts on the comprehensive risk level of international wind farm project in intertidal area, and EPC contractor shall pay extra attention on these risks during the marketing stage or the implementation of project. According to the final risk evaluation result, the level of Project A (the case applied in this research) is “High”, which reminds the EPC contractor that the high level of risk and each potential risks of this project shall be considered carefully and the corresponding measurements in response to these risks shall be applied in advance to avoid the potential damages. The main significance of this paper lies in that based on the evaluation indicator system and model built in this paper, the comprehensive risk assessment for international wind farm project in intertidal area under EPC mode can be carried out from the perspective of EPC contractor, which can provide a certain decision support for related parties during the marketing or implementation of similar projects and realize the risk identification in advance, and it also ensures their own benefit when carrying out related overseas activities.

References 1. Yang X (2020) Research on risk assessment of overseas project based on EPC mode under the background of “Belt and Road Initiative”. Xihua University 2. Ding N (2021) Risk measurement analysis during the management of international projects. China Mark 06:114–115 3. Che L, Feng K, et al (2020). Risk assessment of traffic facilities PPP project based on DEMATEL-ANP. J Civil Eng Manag 37(06):152–157 4. Wen C, Yong S (2006) The scientific significance and future development of extension theory. J Harbin Inst Technol 07:1079–1086 5. Zhiwei W, Shuguang L, Jian X (2021) Research on risk evaluation of rail transit PPP project based on matter-element extension method. Math Pract Theory 51(03):15–25 6. Mytilinou V, Kolios AJ (2019) Techno-economic optimization of offshore wind farms based on life cycle cost analysis on the UK. Renew Energy 132:439–454 7. Chuilin F (2020) Research on risk management of international project under EPC mode. Account Learn 33:61–62 8. Yanrong H (2020) Analysis of risk control and claim during the contract management of offshore wind power project. Technol Econ Guide 28(20):30–31 9. Wu Y, Zhang T (2021) Risk assessment of offshore wave-wind-solar-compressed air energy storage power plant through fuzzy comprehensive evaluation model. Energy 223

Comprehensive Risk Evaluation of International Wind Farm Project …

43

10. Zhiwei F (2020) Research on risk factors of YJ offshore wind power project from the perspective of stakeholders. Beijing Jiaotong University 11. Li J, Wu F, et al (2017) Research on risk evaluation of transnational power networking projects based on the Matter-element extension theory and granular computing. Energies 10:1523 12. Lin X (2020) Risk analysis and measurements for offshore wind farm project. Fujian Build Mater 04:112–113+6

Forthcoming Opportunities for Obtaining Energy from New Renewable Sources in Romania Violeta-Monica Radu , Iustina Popescu Boaja , Alexandru Anton Ivanov , George Dinca , and Robert Szabo

Abstract One of the renewable resources of strategic energy, currently untapped, can be obtained through capturing energy from the natural tendency to cancel the salinity gradients between salt water and fresh water, the so-called “Blue Energy”. From this point of view, Romania presents a potential and untapped major energy resource based on the salinity gradient, having as promoter the discharge of the Danube River into the Black Sea. Globally, research is being conducted at both the laboratory and micropilot levels for certain types of approaches for the techniques used in obtaining energy from the salinity gradient. In this paper are briefly presented several conceptual principles with the applicative technological potential that seeks to capitalize on the energy of the salinity gradient. At the same time, both the performances and the limitations of these techniques are highlighted. Among the phenomena that may be used to capitalize the energy from the salinity gradient following processes are mentioned: ionic and capacitive exchange, osmotic direct mixing, absorption–desorption, based on the difference in vapor pressure. Blue energy may be an important forthcoming opportunity of renewable energy for global electricity demand. Keywords Energy · Renewable sources · Romania

1 Introduction Globally, an increasing energy consumption and the necessity to find new solutions to reduce carbon dioxide emissions are leading to the investigation of new processes and methods for producing renewable energy [1, 2]. The need for renewable and green energy is widely recognized and, therefore, massive efforts are being made globally to develop and capitalize on new sources of green energy, including wind, solar, geothermal, biomass, thermal ocean, waves and tides, more recently evaluating the possibilities of capitalizing the salinity gradient V.-M. Radu (B) · I. P. Boaja · A. A. Ivanov · G. Dinca · R. Szabo Geoecolab Laboratory, Geological Institute of Romania, 1 Caransebes Street, District 1, Bucharest, Romania e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_5

45

46

V.-M. Radu et al.

between freshwater and saltwater, as called sources of “Blue Energy” or “Salinity Gradient Power” (SGP) [3]. So far, green energy sources have not proven to be efficient enough to replace fossil fuels, some of the main problems being both the high costs of installations and the limited efficiency of electricity generation and storage. Salinity gradient energy is an offshore energy source, which can be harnessed by the controlled mixing of two bodies of water with different concentrations of salts, generally freshwater and saltwater, for example, when a river flows into the sea [4, 5]. The main factor in the energy of the salinity gradient is the “difference in chemical potential” between the diluted and concentrated solutions. From this point of view, Romania presents a potential and untapped major energy resource based on the salinity gradient, having as promoter the discharge of the Danube River into the Black Sea (Fig. 1). When saltwater reaches fresh water, the energy of the system decreases compared to the sum of the individual energies before mixing. Thus, a certain amount of free energy is lost due to mixing. The concept of blue energy is to harvest this free energy. A fully integrated and properly functioning internal energy market ensures affordable energy prices, transmits the necessary price signals for green energy investments, ensures energy supply and paves the least expensive path to climate neutrality.

Fig. 1 View of the area where the Danube flows into the Black Sea at Sfantu Gheorghe [6]

Forthcoming Opportunities for Obtaining Energy …

47

2 Methods 2.1 Pressure Retarded Osmosis (PRO) Pressure Retarded Osmosis (PRO) is an osmotically driven membrane process that harnesses the energy of mixing between high and low salinity streams to produce mechanical energy [7]. PRO uses a semi-permeable membrane (Fig. 2) placed between the two streams of different salinity, in which water flows from the less concentrated stream to the more concentrated, in order to balance the concentration gradient between the two solutions [8]. This transport of water from the diluted solution at low pressure to the concentrated solution at high pressure results in a pressurization of the volume of water transported. The transported water can be used to generate electricity in a turbine (Fig. 3). Although the PRO process has been significantly improved with the development of membrane technologies, several obstacles are still hampering the progress [10]. The relatively low power generation of the PRO process at the module scale is one of the issues that limits its (Daily appliances). Although the PRO process has already reached the power density level of 24 W/m2 at the laboratory scale process [11], such an amount of power density is not yet confirmed in the PRO scale process mode,

Fig. 2 Semipermeable membrane [8]

Fig. 3 Cycles of the PRO technique [9]

48

V.-M. Radu et al.

and the power density decreases by 10–15 W/m2 when the process is extended. In addition, another issue is related to the missing PRO-targeting membrane which has not been developed yet. As a result, a PRO hybrid process may be a problem-solving measure. The hybrid reverse osmosis (RO)-PRO process is the most famous form of PRO hybrid processes, having the advantage that it does not require a pre-treatment step for the PRO process and can overcome the PRO autonomous case.

2.2 Reverse Electrodialysis (RED) Reverse electrodialysis (RED) is a promising technology to extract sustainable salinity gradient energy [12, 13]. RED uses ion exchange membranes (either anion exchange membranes or cation exchange membranes), which can selectively allow negatively or positively charged ions to pass (Fig. 4). This membrane based technology, allows the separation of the charge which then leads to the formation of an electrochemical potential. Using a stack of anion exchange or cation exchange membranes that face two different salinity streams, the electric potential can be built and used to generate electricity. The difference in electrical potential between the outer compartments of the membrane stack is the sum of the potential difference on each membrane. The difference in chemical potential determines the transport of ions through membranes from the concentrated solution to the diluted solution. For a sodium chloride solution, the sodium ions penetrate the cation exchange membrane in the direction of the cathode and the chloride in reverse [14]. However, the ion exchange membranes used for the RED system often faces limitations in adapting to a real-world system due to the limited pore size and internal strength.

Fig. 4 Process of the RED technique [15]

Forthcoming Opportunities for Obtaining Energy …

49

Fig. 5 Cycles of the CapMix technique [16]

2.3 Capacitive Mixing (CapMix) The generation of electricity using a CapMix process is based on a cycle of charging and discharging electrodes (Fig. 5). The electrodes in a CapMix device are successively exposed to two solutions that have large differences in salinity, such as natural freshwater and seawater. The CapMix devices have three different approaches, in their creation based on how the electrodes are charged: (1) double layer capacitive expansion (CDLE), (2) capacitive energy extraction based on the Donnan potential (CDP), (3) and entropy mixing battery (MEB). For CDLE, a pair of porous carbon electrodes are charged due to the electric potential produced by changes in the thickness of the electric double layer i.e., the charge layer near the electrode surface that is different from the solution, which changes depending on the concentration of salt from the solution [17]. For the CDP process, a layer of ion exchange polymer (or ion exchange membrane) is placed above the electrodes, which produces a potential (E.g.: the Donnan potential), which can charge the membranes in a high concentration solution and discharge into a solution with low concentration. Both processes are based on conductive materials capable of accumulating electric charges on either the electrode surface (CDLE), either on the membrane (CDE). In both cases there are no chemical reactions. The MEB process uses oxidation/reduction reactions and therefore uses battery electrodes to drive the salt concentration-dependent electrode potentials.

2.4 Importance and Efficiency Salinity gradient energy or blue energy is a promising renewable energy source for the future. The literature estimates a coverage of over 80% of the current global demand for electricity when applied in all river mouths.

50

V.-M. Radu et al.

Due to the different operating principles, for the PRO membranes, the design of the supporting layer and the functional active layers are of major importance, while for the RED membranes, the ionic perm selectivity and ionic conductivity are of great importance [18].

3 Pilot Studies The literature review shows that the most advanced research at pilot level was initiated by Norway in 2009 consisting in an installation based on reverse electrodialysis and low-pressure osmosis with a theoretical capacity of approximately 2 MWh [19]. The Statkraft osmotic power plant at Tofte, Norway (Fig. 6), is the world’s first osmotic power or salinity gradient power generation plant [20]. Mega-ton Water System in Japan (Fig. 7) conducts pilot research on capacitive technologies for obtaining energy from potential gradient [21–23]. In 2014, the Netherlands opened a RED power plant on the Afsluitdijk (Fig. 8) dam with a capacity of 50 KW [24, 25]. The developed technologies have the capacity to generate electricity continuously, regardless of the weather conditions, with very low polluting by-products, which makes them reliable sources of green electricity [26–28]. After analysing these processes, it seems that the implementation is influenced by several factors such as: local constraints, applications (energy production, desalination and purification) for which they are intended and general improvements required of the technology. Factors for assessing the possibility of implementation in developed or developing countries are technical, environmental, political and financial criteria. Like any innovative energy technology, in order to be competitive in the market, it must progress in terms of the level of capacity and competence and the lowest possible operating costs. While salinity gradient energy sources can ensure environmentally friendly, costeffective and renewable energy production, efforts still need to be made for optimal large-scale production.

Fig. 6 Statkraft osmotic power plant at Tofte, Norway [20]

Forthcoming Opportunities for Obtaining Energy …

51

Fig. 7 Outputs of the “Mega-ton Water System” [21]

Fig. 8 REDstack’s 50 kW demo plant on the closure dam Afsluitdijk produces electricity [24]

4 Conclusions The ambitious energy and climate targets for 2030 call for the development of a new electricity market model aimed at increasing energy efficiency, renewable energy production, food security, sustainability, decarbonisation, and innovation. Knowledge of recent key developments and technical progresses in front technologies can help guide the usage of salinity gradient as a sustainable energy source assuring progress and a new strategical approach.

52

V.-M. Radu et al.

The current study presents the main processes through which renewable energy can be obtained from the salinity gradient, energy resulting from the mixture of freshwater and saltwater flows. This study could help create the technical and scientific support needed to obtain new sources of renewable energy, with effects on increasing the quality of life by reducing carbon emissions and protecting our planet.

References 1. Emami Y, Mehrangiz S, Etemadi A, Mostafazadeh A, Darvishi S (2013) A brief review about salinity gradient energy. Int J Smart Grid Clean Energy 2(2):295–300 2. Abbasi-Garravand E, Mulligan CN (2021) Feasibility of pressure-retarded osmosis for electricity generation at low temperatures. Membranes 11(8):556 3. Siria A, Bocquet M-L, Bocquet L (2017) New avenues for the large-scale harvesting of blue energy. Nat Rev Chem Perspect 1(0091):1–9 4. Schaetzle O, Buisman CJN (2015) Salinity gradient energy: current state and new trends. Engineering 1(2):164–166 5. Potisa-ad K, Simasatitkul L, Amornraksa S (2021) Comparison between pressure retarded osmosis model using batch and continuous water supply sources. E3S Web of Conf 302 6. https://earthobservatory.nasa.gov/images/80459/where-the-danube-meets-the-black-sea. Accessed 12 Nov 2021 7. Budde J (2021) A comparison of reverse electrodialysis and pressure retarded osmosis as technologies for salinity gradient power. J Renew Energy Short Rev 72–78 8. https://www.materials.sandvik/cz/aplikace/renewable-energy/ocean-and-marine-energy/sal inity-gradient-power-generation/. Accessed 12 Nov 2021 9. Straub AP, Deshmukh A, Elimelech M (2016) Pressure-retarded osmosis for power generation from salinity gradients: is it viable? Energy Environ 1.https://pubs.rsc.org/en/content/articlela nding/2016/ee/c5ee02985f 10. Tawalbeh M, Al-Othman A, Abdelwahab N, Hai Alamia A, Ghani Olabi A (2021) Recent developments in pressure retarded osmosis for desalination and power generation. Renew Sustain Energy Rev 138:110492 11. Ho Chae S, Ha Kim J (2017) Integration of PRO into desalination processes in pressure retarded osmosis. Renew Energy Gen Recov 129–151 12. Chanda S, Amy Tsai P (2021) Renewable power generation by reverse electrodialysis using an ion exchange membrane. Membranes 11(830) 13. Pintossi D, Chen C-L, Saakes M, Nijmeijer K, Bornema Z (2020) Influence of sulfate on anion exchange membranes in reverse electrodialysis. Clean Water 29 14. Acuña Mora D, de Rijck A (2015) Salinity gradient power in practice. GSDR Brief Blue Energy 15. Tesfaye Besha A, Tilahun Tsehaye M, Aili D, Zhang W, Ashu Tufa R (2020) Design of monovalent ion selective membranes for reducing the impacts of multivalent ions in reverse electrodialysis. Membranes 10(7). https://pubmed.ncbi.nlm.nih.gov/31906203/ 16. https://sites.psu.edu/energyfromwater/technologies/capmix/. Accessed 13 Nov 2021 17. Balzer C, Qing L, Wang Z-G (2021) Preferential ion adsorption in blue energy applications. ACS Sustain Chem Eng 9(28):9230–9239 18. Zhang B, Gao H, Tong X, Liu S, Gan L, Chen Y (2019) Current trends and future developments on (bio-) membranes. Renew Energy Integr Membr Oper 133–152 19. El Bassam N, Eighteen C (2021) Technologies at the experimental stages. Distributed renewable energies for off-grid communities, 2nd edn. Empowering a Sustainable, Competitive, and Secure Twenty-First Century, pp 435–442 20. https://www.power-technology.com/projects/statkraft-osmotic/. Accessed 13 Nov 2021

Forthcoming Opportunities for Obtaining Energy …

53

21. http://www.sawea.org/pdf/2017/19th_techincal_sessions/Innovative_energy.pdf. Accessed 14 Nov 2021 22. Kurihara M, Takeuchi H (2018) SWRO-PRO system in “Mega-ton Water System” for energy reduction and low environmental impact. Water 10(48) 23. Schunke AJ, Hernandez Herrera GA, Padhye L, Berry T-A (2020) Energy recovery in SWRO desalination. Curr Status New Possibilit Front Sutain Cities 2(9) 24. https://www.dutchwatersector.com/news/redstack-nears-first-green-hydrogen-production-atits-salt-fresh-water-blue-energy-plant. Accessed 14 Nov 2021 25. https://www.betterworldsolutions.eu/portfolio/blue-energy-from-holland/. Accessed 15 Nov 2021 26. Shadman M, Silva C, Faller D, Wu Z, de Freitas Assad LP, Landau L, Levi C, Estefen SF (2019) Ocean renewable energy potential, technology, and deployments: a case study of Brazil. Energies 12 27. Hsu W-S, Preet A, Lin T-Y, Lin T-E (2021) Miniaturized salinity gradient energy harvesting devices. Molecules 26(5469) 28. Jalilia Z, Wergeland Krakhellaa K, Etienne Einarsruda K, Stokke Burheimb O (2019) Energy generation and storage by salinity gradient power: a model-based assessment. J Energy Storage 24(100755)

Thermal Engineering and Energy-Saving Technology

Evaluation of Stratification in Thermal Energy Storages Ioannis Sifnaios , Adam R. Jensen , Simon Furbo , and Jianhua Fan

Abstract Thermal stratification in water-based storages can be destroyed by mixing, heat diffusion, and thermal conduction. For this reason, the evaluation of stratification in water-based thermal energy storages is important for assessing their performance. The most promising indicators were identified and assessed based on their suitability for use in practical applications. The selected stratification indicators were calculated for four simulated storage scenarios comprising a fully stratified, a fully mixed, and two realistic storages. It was found that most indicators had severe limitations in their application. For this reason, a new indicator called internal exergy destruction was proposed, which can be used in combination with the overall exergy efficiency for assessing the performance and stratification of thermal energy storages. The main benefit of internal exergy destruction is that it can be used to compare storages with different heat loss coefficients. In addition, it separates the effects of mixing from the heat losses and is easily applied to real-life storages. Keywords Thermal stratification · Heat storage · Exergy analysis

1 Introduction Thermal energy storages (TES) are often used for bridging the time gap between heat generation and heat demand, especially when using non-dispatchable renewable energy sources [1]. Thermal energy is stored in a TES using heating or cooling in order to be used later. The thermal performance of a heating system utilizing a TES is strongly influenced by stratification. Stratification occurs when a temperature gradient in the TES separates fluid at different temperatures. One study found that by creating stratification in a TES with the use of a diffuser, increased the heating I. Sifnaios (B) · A. R. Jensen · S. Furbo · J. Fan DTU Civil and Mechanical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark e-mail: [email protected] I. Sifnaios Sino-Danish College (SDC), University of Chinese Academy of Sciences, Beijing, China © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_6

57

58

I. Sifnaios et al.

system’s coefficient of performance (COP) by 32% compared to having a fully mixed tank [2]. However, achieving a good thermal stratification inside a TES is challenging due to mixing induced by the inlet flow, heat diffusion caused by natural convection in the TES, and downward thermal conduction [3]. The TES geometry, diffuser design, and operation strategy strongly influence the level of stratification. For this reason, it is critical to be able to quantify the degree of stratification inside a TES. Usually, stratification indicators are used to assess stratification in a TES. Expressions have been developed that can be applied to any water-based heat storage that is directly charged/discharged, i.e., does not use a heat exchanger, e.g., in district heating tanks, domestic hot water tanks, pit thermal energy storages, etc. Haller et al. summarized most of the available stratification indicators [4]. The same study employed some of these indicators to characterize a theoretical TES case comprised of one charge, standby, and discharge period. The study pointed out that all of the available methods have some drawbacks, e.g., some of them cannot be used for both charge and discharge, whereas others fail to separate the effects of heat losses from mixing. Overall, the study concluded that the available stratification indicators had limitations in their applications. However, the investigated simulation case was very simplified, including only mixing around the inlet and outlet of the storage and did not include heat losses to the ambient and thermal conduction between the water layers. This study identifies the most promising stratification indicators for assessing the stratification in thermal energy storages for practical applications. The indicators are evaluated on how well they can be used to determine stratification inside a thermal energy storage. Finally, it suggests a new indicator for assessing stratification in TES.

2 Methods First, the stratification indicators used in this study are presented, namely the MIX number, stratification coefficient, exergy efficiency, and overall exergy efficiency, followed by a description of the investigated scenarios.

2.1 MIX Number The MIX number is a dimensionless indicator that quantifies the degree of mixing inside a TES by comparing it to a fully mixed and a fully stratified reference storage [5]. Its range is between zero and one, corresponding to a perfectly stratified and a fully mixed tank, respectively. The MIX number is defined as the ratio of the difference in the moment of energy between a perfectly stratified storage and actual storage to the difference in the moment of energy between a perfectly stratified storage and a fully mixed one:

Evaluation of Stratification in Thermal Energy Storages strati f ied

MI X =

ME

strati f ied

ME

59

− M Eactual f ully−mi xed

− ME

(1)

The moment of energy is calculated for the two theoretical reference cases (stratified and fully mixed) such that they have the same energy content as the actual storage. To calculate the MIX number, the storage is divided into discrete layers (typically corresponding to the number of temperature sensors). The moment of energy for each layer is then calculated by weighing each layer’s energy content with the height from the bottom of the storage to the centroid of the layer. The total moment of energy of the storage is then calculated as the sum of all layers, as seen in Eq. (2). ME =

N 

  ρi · Vi · C p,i · Ti − Tr e f · z i

(2)

n=1

where N is the total number of layers in the storage, ρ i is the water density, V i is the water volume of the layer, C p,i is the specific heat, T i is the water temperature, and zi is the distance from the center of the layer to the bottom of the storage. T ref is the reference temperature, meaning the temperature at which the storage is considered empty.

2.2 Stratification Coefficient The stratification coefficient expresses the degree of stratification based on the massweighted square of the difference of the actual storage temperature to the mean storage temperature [6]: 2  N  m i · Ti − Tavg St = m total n=1

(3)

where Ti is the temperature of each layer, mi is the mass of each layer, Tavg is the average storage temperature, and mtotal is the total mass of the storage.

2.3 Exergy Efficiencies There are several expressions suggested regarding exergy efficiency. In this study, the two expressions presented by Haller et al. [7] and Rosen et al. [8] are used. Haller et al. define the exergy efficiency as the internal exergy loss of an experimental TES relatively to the internal exergy destruction of a fully mixed TES:

60

I. Sifnaios et al.

ηst,ξ = 1 −

ξ destr,ex p ξ destr,mi x

(4)

where the internal exergy destruction is found through the exergy balance of the TES, using Eq. (5): ξ int,destr = ξ f low − ξ stor e − ξ heatloss

(5)

The fully mixed storage is simulated with the same flow rate, inlet temperature, and heat loss coefficient as the experimental storage. For further details, the reader is referred to the paper by Haller et al. [7]. Conversely, Rosen et al. used a general expression for the overall exergy efficiency of a TES, comparing the exergy recovered from the TES to the exergy input of the TES, as defined in Eq. (6). ηoverall =

ξout put ξinput

(6)

It has to be noted that the two expressions are very different in their application, i.e., the former gives information about the precise time when mixing occurs during one storage cycle. In contrast, the latter gives an overall efficiency for one TES cycle.

2.4 Simulated Scenarios In order to demonstrate the performance of the investigated stratification indicators, four idealized storage scenarios were simulated. Mixing was implemented using the methodology recommended by Haller et al. [7]. The investigated scenarios were a fully stratified storage, a fully mixed storage, and two realistic storages. Each case was investigated for two full charge/discharge cycles. The scenarios have been simulated, including and excluding heat losses. The heat loss coefficient used in the simulations was selected such that the “realistic scenario 1” has an energy efficiency of 90%. It has to be noted that the effect of the storage walls was neglected. Each storage was charged using a constant temperature of 90 °C and discharged with a constant inlet temperature of 45 °C. The time step of the simulation was 1 min, such that the charged/discharged flow was equal to the volume of one node. This ensured that numerical diffusion was avoided. An overview of the simulation parameters is presented in Table 1. Note that constant values for ρi and C p,i were used in order to simplify the simulations and focus on mixing effects. In addition, the value of the effective thermal conductivity was set to 2.5 W/(m K) as suggested by Haller et al. [7].

Evaluation of Stratification in Thermal Energy Storages

61

Table 1 Simulation parameters Parameter

Value

Unit

Storage volume

1

m3

Water density

980

kg/m3

Water specific heat capacity

4200

J/(kg K)

Charging temperature

90

°C

Discharging temperature

45

°C

Threshold temperature

10

°C

Number of nodes

60

–

Time step duration

1

min

Flow rate charge/discharge

980

kg/hr

Heat loss coefficient

6

W/(m2 K)

Effective vertical thermal conductivity

2.5

W/(m K)

Storage height

1

m

Fully stratified scenario. There is no mixing between the tank nodes and no vertical heat conduction inside the tank in this scenario. Essentially the flow during charge and discharge is simulated as plug flow. Fully mixed scenario. In the fully mixed scenario, the temperature of the water entering the tank at each time step is instantaneously mixed with the temperature of the rest of the storage. Realistic scenarios. The realistic case mimics the actual conditions inside a storage, hence the naming. In this scenario, as water enters the storage, the nodes’ temperatures close to the inlet are mixed, simulating the inlet jet-mixing phenomenon. Two variations of the realistic scenario were simulated, corresponding to a better performing and a worse performing diffuser. For the case of the better diffuser, denoted from now on as realistic scenario 1, the total tank volume used for imitating the inlet jet mixing was approximately 10% of the tank’s total volume. Similarly, for the case of the worse diffuser, denoted from now on as realistic scenario 2, 20% of the entire tank volume was used. The rest of the simulation conditions were the same for the two realistic scenarios. After mixing, vertical heat conduction was applied between the tank nodes based on the temperature distribution in the storage. Last, heat losses to the ambient were calculated based on the temperature difference between the storage nodes and the ambient temperature (for the cases where heat losses were enabled). An illustration of the investigated storage during charging and discharging is presented in Fig. 1.

62

I. Sifnaios et al.

Fig. 1 Illustration of the tank charging and discharging

2.5 Charging and Discharging Conditions The storage in each simulation starts with charging and is initialized as empty, i.e., having a uniform temperature of 45 °C. The storage immediately switches from charging (operation = 1) to discharging (operation = 0), or vice versa, when the storage is identified as empty or full. The criteria for the storage being full or empty were implemented using a threshold temperature for the top and bottom node, as indicated in Eq. (7).  operation =

1, i f Tbottom ≥ Tcharge − T thr eshold 0, i f Ttop ≤ Tdischarge + T thr eshold

(7)

The temperature profile inside the tank during the two storage cycles is presented in Fig. 2. The top figure shows the temperature profile for the realistic scenario 1 without heat losses, while the bottom figure is the same case but includes heat losses. It can be observed that the presence of heat losses lowers the temperature in the tank, predominantly at the top of the storage, and also slightly increases the required time to charge the storage.

Evaluation of Stratification in Thermal Energy Storages

63

Fig. 2 Temperature profile in the storage for the realistic scenario 1. The top figure does not include heat losses

3 Results The results of the calculated stratification indicators are presented in Figs. 3, 4, 5 and 6. Each indicator is applied to all the possible storage scenarios, i.e., fully mixed (with and without heat losses), realistic 1 and 2 (with and without heat losses), and fully stratified. The fully stratified scenario is only simulated without heat losses since if heat losses did occur, it would no longer be a fully stratified case.

3.1 MIX Number As expected, the MIX number for the fully mixed storage is constantly equal to one, regardless of heat losses, as seen in Fig. 3. Similarly, the MIX number is always equal to zero for the fully stratified storage. For the realistic storages, the MIX number varies throughout the storage cycles. This is partly because the MIX number is strongly affected by the energy content of the storage. Large spikes can be noticed at high (fully charged) and low (fully discharged) energy contents in the MIX number because a fully charged and discharged storage is considered not stratified. By comparing the two realistic storages, it is clear that a worse performing diffuser (case 2) creates more mixing in the TES; thus, it has a higher MIX number. By adding heat losses to the simulation, the MIX number shows the storage to be more stratified during discharging than charging. In fact, the MIX number becomes negative for a few time steps for Case 1, as the heat losses bring the average storage temperature below the reference temperature. While the MIX number does provide some use in comparing similar storages, it is difficult to draw a conclusion regarding stratification.

64

I. Sifnaios et al.

Fig. 3 MIX number for the investigated storages

In addition, it is highly influenced by the heat losses and the choice of the reference temperature.

3.2 Stratification Coefficient For the fully mixed storage, the stratification coefficient was zero since there was a uniform temperature in the storage at all times (Fig. 4). The stratification coefficient was proportional to the energy content for the fully stratified storage and ranged from 0 to approximately 500 K2 . The values were between the two ideal cases for the realistic cases, with a maximum of approximately 400 and 300 K2 for cases 1 and 2, respectively. The results of this indicator are much easier to interpret than the MIX number and were more useful in comparing the two storages. Similar to the MIX number, the stratification coefficient shows a low level of stratification when the storage is almost full or empty. However, this indicator has the benefit of not depending on a reference temperature and, therefore, never becomes negative. Nonetheless, the absolute value of the stratification coefficient has no physical meaning and, like the MIX number, is affected by heat losses. When applying heat

Evaluation of Stratification in Thermal Energy Storages

65

Fig. 4 Stratification coefficient for the investigated storages

losses to the realistic cases, the stratification coefficient shows a smaller degree of stratification during discharging compared to charging. This is because, when charging, the temperature at the top of the storage is constantly 90 °C, while during discharge, the top temperature decreases due to heat losses, as can be seen in Fig. 2. This leads to a lower stratification coefficient during discharge. Overall, this indicator can be used to compare the stratification degree in two storages, but one needs to be cautious when the two storages have different heat loss coefficients.

3.3 Exergy Efficiencies In Fig. 5, the exergy efficiency of the investigated storages is presented. As expected, the fully mixed storage had an exergy efficiency of 0%, while the fully stratified had an efficiency of 100%. Again, the realistic storages had an efficiency between the other two, and applying heat losses reduced the exergy efficiency. The method of Haller et al. [7] gives significantly different results compared to the MIX number and the stratification coefficient.

66

Fig. 5 Exergy efficiency for the investigated storages

Fig. 6 Internal exergy destruction for the investigated storages

I. Sifnaios et al.

Evaluation of Stratification in Thermal Energy Storages

67

However, it is difficult to apply Haller’s exergy efficiency method to real-life scenarios. This method can only be applied for specific, well-defined time periods in the storage operation. For example, it requires a clear distinction between the charge and discharge periods, which in some cases is not possible, e.g., for storages that are used both for short and long-term storage. In addition, it can only be used if the charging and discharging mass flow and inlet temperature are constant during the storage operation. In general, the methods that compare a real-life storage with reference fully mixed or fully stratified cases, i.e., the MIX number and exergy efficiency, can be difficult to use in practice. The overall exergy efficiency, presented in Table 2, is considered the most reliable indicator of stratification performance. This table gives information about the percentage of exergy lost due to mixing and heat losses. For example, the realistic storage 1 has a 10% lower efficiency due to mixing compared to the fully stratified storage, but 18% lower exergy efficiency, including mixing and heat losses. In order to get information about the precise time when mixing occurred in the storage, it is suggested to use the internal exergy destruction as given in Eq. (5). Exergy destruction gives the amount of exergy lost in the storage due to mixing caused by inlet jet mixing and vertical thermal conduction. Since the exergy loss due to heat losses is subtracted from the expression, the internal exergy destruction can be used for comparing the amount of mixing in two or more storages, even if they do not have the same heat loss coefficient. Figure 6 presents the internal exergy destruction for the investigated storages. It can be observed that, apart from the fully stratified storage, the exergy destruction mainly occurs at the start of the charge and discharge period, as this is when the thermocline develops. In the case of the realistic storage 1, which is well stratified, the exergy destruction only occurs at the beginning of charge and discharge and is close to zero during most of the storage operation. However, for a less stratified tank (e.g., for the realistic storage 2), the internal exergy destruction occurs over a longer period, as it takes longer to build up the thermocline. In addition, the internal exergy destruction remains essentially the same regardless of heat losses, allowing the comparison of the level of stratification independent from the heat losses. Table 2 Overall exergy efficiency for investigated storages Overall exergy efficiency

Value (%)

Fully mixed storage

54

Fully mixed with heat losses

51

Realistic storage 1

90

Realistic storage 1 with heat losses

82

Realistic storage 2

85

Realistic storage 2 with heat losses Fully stratified storage

77 100

68

I. Sifnaios et al.

4 Conclusions This paper investigated stratification for four different storage scenarios: a fully mixed, a fully stratified, and two realistic scenarios. Four stratification indicators were assessed from the literature: the MIX number, stratification coefficient, exergy efficiency, and overall exergy efficiency. Apart from the overall exergy efficiency, all the other investigated indicators had significant drawbacks leading to either results that were difficult to interpret or results applicable to specific, well-defined periods in the storage operation. It is suggested to use the overall exergy efficiency and supplement it with the internal exergy destruction for assessing stratification in a storage. The overall exergy efficiency gives a thermodynamically based quantification of the stratification performance of a TES. The internal exergy destruction can then be used to illustrate the specific times at which mixing occurs in the storage. The main benefit of these two methods is that they do not rely on a fully mixed or fully stratified reference storage simulation, which can be difficult or impossible to implement in real-life cases. Applying the recommended methods to real-life storages is a topic of future work. It is believed that they have a great potential for comparing the stratification among storages since they can be applied to all storages regardless of their use, e.g., short term, long term, or combination of the two. Acknowledgements This study was funded by the Danish Energy Agency through EUDP grant no. 64018-0134 and by the Sino-Danish Center for Education and Research (SDC) Ph.D. program.

References 1. Li SH, Zhang YX, Li Y, Zhang XS (2014) Experimental study of inlet structure on the discharging performance of a solar water storage tank. Energy Build 70:490–496. https://doi.org/10.1016/j. enbuild.2013.11.086 2. Sifnaios I, Fan J, Olsen L, Madsen C, Furbo S (2019) Optimization of the coefficient of performance of a heat pump with an integrated storage tank—a computational fluid dynamics study. Appl Therm Eng 160. https://doi.org/10.1016/j.applthermaleng.2019.114014 3. Fan J, Furbo S (2012) Thermal stratification in a hot water tank established by heat loss from the tank. Sol Energy 86:3460–3469. https://doi.org/10.1016/j.solener.2012.07.026 4. Haller MY, Cruickshank CA, Streicher W, Harrison SJ, Andersen E, Furbo S (2009) Methods to determine stratification efficiency of thermal energy storage processes—review and theoretical comparison. Sol Energy 83:1847–1860. https://doi.org/10.1016/j.solener.2009.06.019 5. Andersen E, Furbo S, Fan J (2007) Multilayer fabric stratification pipes for solar tanks. Sol Energy 81:1219–1226. https://doi.org/10.1016/j.solener.2007.01.008 6. Wu L, Bannerot RB (1987) An experimental study of the effect of water extraction on thermal stratification in storage. In: Proceedings of the 1987 ASME-JSME-JSES solar energy conference, Honolulu. pp 445–451

Evaluation of Stratification in Thermal Energy Storages

69

7. Haller MY, Yazdanshenas E, Andersen E, Bales C, Streicher W, Furbo S (2010) A method to determine stratification efficiency of thermal energy storage processes independently from storage heat losses. Sol Energy 84:997–1007. https://doi.org/10.1016/j.solener.2010.03.009 8. Rosen MA, Pedinelli N, Dincer I (1999) Energy and exergy analyses of cold thermal storage systems. Int J Energy Res 23:1029–1038. https://doi.org/10.1002/(SICI)1099-114X(199910 10)23:12%3c1029::AID-ER538%3e3.0.CO;2-C

Simulation of Integrated Gasification Combined Cycle (IGCC) and Performance Analysis Xue Sun, Zhen Yang , and Yuanyuan Duan

Abstract As the concern of global warming and energy demand increasing, integrated gasification combined cycle (IGCC) shows its potential as a clean and efficient coal utilization method. In this paper, commercial software Aspen Plus is used for IGCC model simulation, and the effects of key operation parameters on the system have been analyzed. The model consists of air separation unit (ASU), gasification unit, and combined cycle unit. Coal is fed in gasifier with oxygen and steam to produce syngas, and then fed into combined cycle as fuel to produce electricity. Subunits of the system have been validated against the experimental or simulation data in literatures, and results show high consistency. The influence on syngas component, system power and system efficiency of key operation parameters including oxygen/coal ratio, water/coal ratio have been analyzed, and the results can show how to improve the system performance by these parameters. Keywords IGCC · Model · Aspen Plus

1 Introduction According to energy statistics, global energy demand is still increasing [1]. Taking into account the intermittency and instability of renewable energy such as solar and wind energy, the main source of energy in the world is still fossil energy now. Fossil energy has the advantages of abundance, stability, and high utilization efficiency. However, CO2 emissions caused by fossil energy will lead to global warming, so it is very important to find a clean and efficient way to use fossil energy. IGCC (Integrated Gasification Combined Cycle) is proposed as a clean and efficient way of coal utilization, and it is a very promising energy generation technology [2, 3]. IGCC is the technology that combines coal gasification technology and gas-steam combined cycle for efficient power generation. The cascade utilization of energy can be realized through gasification and combined cycle, and the efficiency of the system X. Sun · Z. Yang · Y. Duan (B) Tsinghua University, Beijing 100084, People’s Republic of China e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_7

71

72

X. Sun et al.

can be improved. At the same time, IGCC is also a very potential clean coal utilization method. By adding a CO2 capture device after the syngas generated in the gasifier, the pre-combustion capture of carbon can be achieved, which greatly reduces the difficulty of CO2 emission reduction and power consumption. At present, many studies have been conducted on the IGCC system, mainly focusing on the optimization of sub-components and the improvement of the system. Al-Zareer et al. used Aspen Plus software to establish the kinetics and thermodynamics models of the gasification process. The results showed that the simulation error of the thermodynamic model was slightly larger, but the applicability to different fuels was better [4]. Li et al. simulated coal gasification and used multi-objective optimization and artificial neural network to analyze the influence of oxygen-to-coal ratio and water-to-coal ratio on the gasification process, the results show that the optimal oxygen-to-coal ratio and water-to-coal ratio are 0.89 and 0.05, respectively [5]; Dong et al. simulated M701F gas turbines and analyzed the influence of atmospheric temperature on the performance of gas turbines [6]; Frey et al. integrated the air separation unit and gas turbines unit to improve total system performance [7]; Emun et al. simulated the overall system of IGCC and did some heat exchange pinch analysis [8]. Although the IGCC system sub-components and system simulation have been studied to a certain extent, there has not been a very detailed description and introduction of the overall system model establishment, performance and the influence of the main parameters. In this study, Aspen Plus software was used to establish a complete IGCC system and explained in detail the sub-module simulation process and some key parameters of the gasification island and the power island. Good consistency of established model and previous literature data had been confirmed after comparison. At the same time, this paper simulated and analyzed the effects of oxygen-to-coal ratio and water-to-coal ratio on syngas composition, heating value, gas production rate, cold gas efficiency, the work generated by each sub-module, and the work of the overall system, and also explained the reasons for their changes. In this paper, the detailed simulation process of the IGCC system and the impact of key parameters on the system were comprehensively analyzed and can provide references for future research.

2 System Description and Model Development 2.1 IGCC System Description Figure 1 is a simple IGCC system process flow diagram. The system consists of two parts: a gasification island and a power island, which are composed of subcomponents such as an air separation unit, a gasification unit, and a gas-steam combined cycle unit. The air enters the air separation unit, and oxygen is extracted from the air through separation technology. The oxygen produced by the air separation unit enters the gasifier and are gasified together with coal and water to produce

Simulation of Integrated Gasification Combined Cycle …

73

Fig. 1 Process flow diagram of IGCC system

syngas which is rich in CO and H2 . Syngas is used as fuel to be combusted in the combustion chamber of a gas turbine and produce high-temperature flue gas to drive the gas turbine to produce electricity. The high-temperature exhaust gas of the gas turbine can be used as the heat source of the steam cycle to heat feed water for power generation, ultimately realizing the cascade utilization of energy and improving the system efficiency.

2.2 Model Development The overall model of the IGCC system was established using Aspen Plus software, and the total model diagram is shown in Fig. 2. The model was simulated which combined the air separation unit, gasification unit, gas turbine unit, heat recovery

Fig. 2 Model of IGCC system in Aspen Plus

74

X. Sun et al.

steam generator (HRSG) unit and steam turbine unit. Detailed simulation process and parameters will be introduced later. Gasification Island. The gasification island includes two units: air separation unit and gasification unit. Figure 2 shows the gasification island model established in Aspen Plus. The specific blocks and parameters used in the model are shown in Table 1. Air separation unit is the separation device, which can obtain the oxygen required for gasification from the air. The air (ASU-AIR) enters the compressor AIRCOM to increase the pressure to 8 bar, and then enters the rectification separation and compression, the generated O2 (ASU-O2 ) goes into the gasifier (GASIFER) for coal gasification. In the gasifier, coal will eventually produce CO, H2 and other gases after pyrolysis and gasification. The main reactions in the gasifier are shown in Eq. (1)–(7): C + H2 O → CO + H2

(1)

C + 0.5O2 → CO

(2)

Table 1 Main blocks used in the model and their description and parameters Block ID

Type

Description

Key Parameters

AIRCOM

COMPR

Air compressor

Outlet pressure = 8 bar

B12

COMPR

Oxygen compressor

Outlet pressure = 24 bar

B7

DSTWU

Separate air into O2 and N2

N2 (light key) recovery set 0.99 O2 (heavy key) recovery set 0.01

DECOM

RYIELD

Convert coal to components based on element and mass balance

Operating temperature = 30°C, operating pressure = 24.68 bar; using a calculator based on ultimate analysis

GASIFER

RGIBBS

Gasification process simulation

Operating pressure = 24.78 bar

COMP

COMPR

Air compressor

Pressure ratio = 17 ïiso = 0.9

COMBUSTE

RGIBBS

Combustion process simulation

Operating pressure = 16.7 bar

GT

COMPR

Gas turbine

Pressure ratio = 16 ïiso = 0.9

HRSG

MHEATX

Heat exchange between fuel gas and feed water

Cold steam outlet temperature = 540°C, 280°C

HP

COMPR

High-pressure steam turbine

Pressure ratio = 5 ïiso = 0.9

MP

COMPR

Middle-pressure steam turbine

Outlet pressure = 6 bar ïiso = 0.9

LP

COMPR

Low-pressure steam turbine

Outlet pressure = 0.05 bar ïiso = 0.9

CONDEN

HEATER

Steam condenser

Outlet temperature = 20°C

Simulation of Integrated Gasification Combined Cycle …

75

C + O2 → CO2

(3)

CO + 0.5O2 → CO2

(4)

C + CO2 → 2CO

(5)

H2 + 0.5O2 → H2 O

(6)

CO + H2 O → CO2 + H2

(7)

The current gasifier includes entrained bed gasifier, fluidized bed gasifier, fixed bed gasifier, etc. Because the entrained bed gasifier has the advantages of high efficiency and fuel diversity, this study selects the entrained bed gasifier as the IGCC system gasification unit. Using Aspen Plus to model the entrained bed gasification process, kinetic model and thermodynamic model can be used. Due to the good accuracy of the thermodynamic model and good fuel variability [9], the gasification process was simulated based on the Gibbs free energy minimization approach. The gasification model is composed of the pyrolysis module DECOM and the gasification module GASFIER. The coal enters the DECOM module for pyrolysis, and the simplest components of each constituent element are produced, such as C, H2 , N2 , O2 , S, Cl2 and ash. The output components corresponding to the elements are determined by the ultimate analysis composition of the coal, and the Fortran Calculator is used in the program to calculate. The pyrolysis products are fed into GASFIER and mixed with water and oxygen generated by the air separation unit for gasification. In GASFIER, the Gibbs free energy minimization approach is used to calculate the syngas parameters which contained CO, H2 , CO2 , CH4 , H2 O, etc. Power Island. The combined cycle sub-model includes gas turbine, steam turbine and HRSG [6, 10]. The excess air is compressed by the compressor COMP to reach a high-pressure state, and then enters the combustion chamber COMBUSTE, where burned with the CO/ H2 rich syngas from the gasifier to produce a high-temperature and high-pressure mixture of H2 O and CO2 . Then it is mixed with high-pressure cooling air before the inlet of the turbine (the cooling air volume is 0.17 of the total compressed air volume) and enters the gas turbine GT for expansion and work generation. After generating power, the gas will drop from 1400°C to around 600°C. The 600°C flue gas can be used as the heat source of the high-pressure steam which is only 540°C to realize the cascade utilization of heat. The HRSG and steam cycle unit uses a three-pressure reheat cycle. The 600°C flue gas enters the heat recovery steam generator HRSG and transfers heat to the feed water of the steam cycle, raising the temperature of feed water to the steam turbine demand temperature. After the steam passes through the high-pressure steam turbine HP to generate energy, it enters the HRSG again to absorb heat and raises the temperature to the main steam temperature. The reheat steam then enters the medium-pressure steam turbine MP

76

X. Sun et al.

for expansion and energy generation. The low-pressure main steam and the mediumpressure turbine outlet steam are mixed to enter the low-pressure turbine LP. The exhaust steam enters the condenser CONDEN for cooling, and the setting of the feed water volume is calculated by the heat exchange of the flue gas.

2.3 IGCC Model Validation Gasifier validation. In the IGCC simulation, the gasifier is simulated based on the Gibb free energy minimization method, and the coal used is Illinois #6 coal of which industrial analysis and ultimate analysis are shown in Table 2. In this paper, experimental values of Wen’s gasification [11] were used for validation, and the validation results are shown in Table 3. It can be seen that the molar fraction of the main components including CO\H2 \CO2 of the model and WEN’s experiments are in good agreement, and it confirms that the gasification model can be used to predict the actual gasification performance. Gas turbine validation. The M701F gas turbine is used for the gas turbine model. The validation comparison with Dong [6] in Table 4 shows that the power generated Table 2 Illinois #6 coal information Parameter Proximate

Value

Moisture/wt.%

0.20

Fixed Carbon/wt.%

Ultimate

58.01

Volatile Matter/wt.%

26.46

Ash/wt.%

15.53

C/wt.%

74.05

H/wt.%

6.25

O/wt.%

1.32

N/wt.%

0.71

S/wt.%

1.77

Ash/wt.%

15.53

Heating Value/MJ·kg-1

29.14

Table 3 Illinois #6 coal gasification validation data Element/% (Mole frac. without water)

Wen’s experiment results

Model

CO

57.6

57.39

H2

39.1

38.4

CO2

2.95

2.55

Simulation of Integrated Gasification Combined Cycle …

77

Table 4 M701F gas turbine validation data Parameter

Dong’s results

Model value

Pressure ratio of compressor

17

17

Pressure ratio of gas turbine

16.33

16.33

Gas turbine inlet temperature/o C

1400

1400

Air mass flow/kg·s−1

651

651

586

559.234

270

265.787

Outlet

temperature/o C

Net power/MW

by the gas turbine model is very close to that in literature at the same pressure ratio, fuel volume, and air flow rate. After validation, it is seen that the present model is in good agreement with experimental and literature data, and specific parameter impact analysis will be shown in the next section.

3 Result and Discussion 3.1 Gasification Effect Analysis The syngas produced by the gasifier has a great impact on the entire system. When the composition, heating value, flow rate, etc. of the syngas change, the subsequent combustion air volume, flue gas flow rate, feed water flow rate, and power cycle work will change, so consider the influence of oxygen-to-coal ratio and water-tocoal ratio on the coal gasification process is very important. The main parameters to measure the performance of the gasification process include syngas composition, syngas heating value, gas production rate, cold gas efficiency, etc. The composition of syngas contains CO\ H2 \ CO2 and H2 O. Among them, CO and H2 are the main effective syngas components that will impact the heat value of syngas (LHV syngas ). Y g (gas production rate) is the gas production volume per kg coal, and the CGE (cold gas efficiency) can be calculated [12]: CGE =

L H Vsyngas × Yg × 100% L H Vcoal

(8)

where LHV syngas and LHV coal represent the heating value of syngas and coal respectively; Y g represents the gas production volume per kg coal. Figure 3 shows the influence of the oxygen-to-coal ratio on syngas component mole fraction as the water-to-coal at 0.2. It shows the mole fraction of CO in syngas decreases when the oxygen-to-coal ratio increases. The main reason is as the oxygento-coal ratio increases, the oxygen content in the gasifier increases, which makes it

78

X. Sun et al.

Fig. 3 Effect of oxygen/coal on the mole fraction of the syngas component

easier to convert CO into CO2 . At the same time, it can be found that the CO2 content continues to rise with the increase of the oxygen-to-coal ratio. The H2 in the syngas first increases and then decreases, and the highest point occurs when the oxygen-tocoal ratio is around 0.82. In the beginning, the oxygen-to-coal ratio increases which promotes the gasification process and converts coal to H2 continuously. When the oxygen-to-coal ratio rises to around 0.82, as the oxygen increases, the amount of H2 converted into H2 O gradually increases and decreases the H2 mole fraction. The trend of H2 O is consistent with H2 . The total molar content of the effective components CO and H2 of syngas increases after a little decreasing and then reaches a peak before keeping decreasing with the increase of the oxygen-to-coal ratio, mainly because the H2 first increases and then decreases while CO decreases continuously. Figure 4 shows the trend of the syngas calorific value, gas production rate and cold gas efficiency as the oxygen-to-coal ratio increase. It can be seen from Fig. 4 that the calorific value of the syngas has a peak around 0.82, which is mainly affected by the changes in the composition of CO and H2 in the syngas. The gas production rate first increased sharply with the increase of the oxygen-to-coal ratio, and then slowly increased. In the early stage, with the rapid increase of hydrogen, the volume

Fig. 4 Effect of oxygen/coal on the syngas LHV, Y g , CGE

Simulation of Integrated Gasification Combined Cycle …

79

of gas per unit mass increased faster, and in the later stage, due to the continuous increase of the oxygen-to-coal ratio, the amount of oxygen in the system continues to increase, which leads to an increase in the gas production rate. The cold gas efficiency comprehensively considers the gas production rate and the calorific value of the syngas. And it increases first and then decreases with a maximum of 0.91 as oxygen-to-coal ratio at 0.82. Figures 5 and 6 show the trend of the components, heating value, gasifier gas production rate and cold gas efficiency of the syngas when the oxygen-to-coal ratio is 0.8 and the water-to-coal ratio changes from 0.2 to 1.2. It can be seen from Fig. 5 that as the water-to-coal ratio increases, the CO in the syngas decreases rapidly, while the H2 decreases slowly, and both CO2 and H2 O show an upward trend. This is because as the water-to-coal ratio increases, the H2 O entering the gasifier increases, and the temperature in the gasifier decreases, resulting in low-level gasification and reduced H2 and CO production. On the other hand, the H2 O content continues to increase, which intensifies the water–gas shift reaction in the gasifier, which makes CO and H2 O convert to CO2 . After comprehensive effecting, CO drops very quickly, which ultimately leads to a rapid decrease in the amount of effective components. It

Fig. 5 Effect of water/coal on the mole fraction of the syngas component

Fig. 6 Effect of water/coal on the syngas LHV, Y g , CGE

80

X. Sun et al.

Fig. 7 Effect of oxygen/coal and water/coal on CGE

can be found in Fig. 6 that with the continuous increase of the water-to-coal ratio, the heating value of the syngas is rapidly decreasing, the gas production rate is continuously increasing, and the comprehensive cold gas efficiency is decreasing. The water-to-coal ratio increases and the effective component content decreases, resulting in a decrease in calorific value. However, due to the continuous increase in the amount of water in the gasifier, the volume of the final product becomes larger and the gas production rate continues to increase. The combination of the two ultimately leads to cold gas decreasing. Figure 7 shows the cold gas efficiency change in the comprehensive consideration of water-to-coal ratio and oxygen-to-coal ratio. It can be found from the figure that as the oxygen-to-coal ratio increases, the cold gas efficiency first increases and then decreases, and there is a maximum value of CGE. As the water-to-coal ratio increases, the cold gas efficiency continues to decrease, along with the maximum cold gas efficiency point moving to the right. This is because as the water-to-coal ratio increases, the temperature in the gasifier decreases and the content of the effective components of the syngas decreases. At this time, more oxygen is needed to increase the gasifier temperature, and ultimately increases the effective syngas component, so that the cold gas efficiency reaches maximum.

3.2 IGCC System Effect Analysis Figure 8 shows the relationship between the power generation and power consumption of each sub-module of the system per kg coal (power consumption of air separation, power generation of gas cycle and steam cycle) and the relationship of net output power per kg coal with the oxygen-to-coal ratio when the water-to-coal ratio is 0.2. As the oxygen-to-coal ratio increases, the amount of oxygen input increases,

Simulation of Integrated Gasification Combined Cycle …

81

Fig. 8 Effect of oxygen/coal on sub-module power generation or consumption

Fig. 9 Effect of water/coal on sub-module power generation or consumption

which will increase the air inlet volume of the air separation part and then increase the power consumption of the air separation compressor. As mentioned in the previous section, the increase in the oxygen-to-coal ratio will first increase and then decrease the effective components of the syngas, leading to changes in the air volume in the combustion chamber of the gas turbine. On the other hand, the increase in the oxygento-coal ratio will increase the gas production rate, which will affect the gas turbine inlet flow. At the same time, the gas turbine inlet temperature first increases and then decreases with the increase of the oxygen-to-coal ratio. These all comprehensively cause the work generated by the gas turbine to increase first and then decrease, subsequently affect the heat exchange of the HRSG and the work generated by the steam cycle. Integrating air separation, gas cycle, and steam cycle, the total work generated by the system shows a slight downward trend.

82

X. Sun et al.

Fig. 10 Effect of oxygen/coal and water/coal on IGCC system power generation

Figure 9 shows the change of the system work per kg coal when the oxygen-to-coal ratio is 0.8, with the water-to-coal ratio between 0.2 and 1.2. As the water-to-coal ratio increases, the power consumption of the air separation unit remains consistent, while the power consumption of gas and steam cycles decreases. This is mainly because as the water-to-coal ratio increases, the effective components of the syngas decrease and the gasification temperature decreases, then the inlet and outlet temperatures of the gas turbine decrease, resulting in a decrease in the power of the gas-steam combined cycles, and a decrease in the net output of the system. Figure 10 is a graph of system work generation with changes in comprehensive consideration of water-to-coal ratio and oxygen-to-coal ratio. It can be found that the net output work per kg coal of the system decreases as the oxygen-to-coal ratio increases, and at the same time as the water-to-coal ratio increases. This indicates that when the temperature and flow rate of the various components of the system meet the requirements, the oxygen-to-coal ratio and the water-to-coal ratio can be appropriately reduced to increase the performance of the system.

4 Conclusion In this paper, a simple but complete IGCC system was developed in Aspen Plus. Detailed simulation process and specific parameters of sub-module including gasification island and power island were introduced. The gasification unit and gas turbine unit had been validated by experimental and literature data. The characteristic of syngas from the gasifier is the junction of the total system so the analysis of the influence of key gasification parameters such as oxygen-to-coal ratio and water-to-coal ratio on syngas and the total system was finished. It was found that the cold gasification efficiency of the gasifier will increase first and then decrease with the increasing

Simulation of Integrated Gasification Combined Cycle …

83

of oxygen-to-coal ratio, and will decrease continuously as the water-to-coal ratio increases. Analysis of the sub-module power generation and consumption and total system power generation in different oxygen-to-coal ratios and water-to-coal ratios was finished. The net power generated per kg coal of IGCC system decreases as the oxygen-to-coal ratio increases, and at the same time as the water-to-coal ratio increases. Acknowledgements This work is supported by National Science and Technology Major Project (J2019-I-0009-0009).

References 1. Global Energy Review 2021 (2021) IEA, https://www.iea.org/reports/global-energy-review2021 2. Xia C, Ye B, Jiang J, Shu Y (2020) Prospect of near-zero-emission IGCC power plants to decarbonize coal-fired power generation in China: implications from the GreenGen project. J Clean Prod 271:122615 3. Yan L, Cao Y, Wang Z, He B (2020) On a novel carbon-negative IGCC system with cascade CO2 combined cycle. Energy Convers Manage 221:113202 4. Al-Zareer M, Dincer I, Rosen MA (2016) Effects of various gasification parameters and operating conditions on syngas and hydrogen production. Chem Eng Res Des 115:1–18 5. Li R, Yang Z, Duan Y (2021) Modeling, prediction and multi-objective optimization of the coal gasification system. In: E3S web of conferences 2021, vol 242. EDP Sciences 6. Dong J, Wang SL, Chen HP, Zhang XL (2005) Performance analysis of M701F gas turbine. Electr Power Sci Eng 4:12–14 7. Emun F, Gadalla M, Majozi T, Boer D (2010) Integrated gasification combined cycle (IGCC) process simulation and optimization. Comput Chem Eng 34(3):331–338 8. Frey HC, Zhu Y (2006) Improved system integration for integrated gasification combined cycle (IGCC) systems. Environ Sci Technol 40(5):1693–1699 9. Al-Zareer M, Dincer I, Rosen MA (2018) Influence of selected gasification parameters on syngas composition from biomass gasification. J Energy Res Technol 140(4):041803 10. Ma S (2008) Simulation and analysis on IGCC system based on Aspen Plus. North China Electric Power University, Beijing 11. Wen CY, Chaung TZ (1979) Entrainment coal gasification modeling. Ind Eng Chem Process Des Dev 18(4):684–695 12. Niu M, Xie J, Liang S (2021) Simulation of a new biomass integrated gasification combined cycle (BIGCC) power generation system using Aspen Plus: performance analysis and energetic assessment. Int J Hydrogen Energy 46(43):22356–22367

Cooperative Optimization of System Parameters and Heat Exchanger Structure for Geothermal Organic Rankine Cycle Jian Li, Zhen Yang, Yuanyuan Duan, and Zitao Yu

Abstract Organic Rankine cycle (ORC) is a promising geothermal power generation technology. While, the high purchased cost of heat exchangers is a key obstacle for its promotion. To further enhance the thermo-economic performance and competitive advantage of geothermal ORC systems, the cooperative optimization of system parameters and heat exchanger structure is proposed in this work. The common shell-and-tube heat exchanger is focused. The influences of the inner diameter and wall thickness of heat exchange tube on the optimum design and thermo-economic performance of ORC system are analyzed. The performance improvement effects of cooperative optimization way are evaluated. The results show that the inner diameter and wall thickness have non-negligible influences on the optimum design and thermo-economic performance of geothermal ORC system. A larger thermoeconomic benefit can be obtained by optimizing the inner diameter and wall thickness at low geothermal temperature and large geothermal flow rate. While, the thermoeconomic benefits only differ slightly for various organic fluids. The cooperative optimization is confirmed as an effective way to substantially enhance the thermoeconomic performance of geothermal ORC system, and the minimum specific investment cost can decrease by 12.5–15.6% compared with the traditional system-level optimization. Keywords Geothermal energy · Organic Rankine cycle (ORC) · Heat exchanger · Thermo-economic performance · Cooperative optimization

J. Li · Z. Yang · Y. Duan (B) Key Laboratory for Thermal Science and Power Engineering of MOE, Tsinghua University, Beijing 100084, People’s Republic of China e-mail: [email protected] J. Li · Z. Yu (B) State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou 310027, People’s Republic of China e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_8

85

86

J. Li et al.

1 Introduction Geothermal energy is abundant and widely distributed, which is more stable and less subject to weather fluctuations than wind energy and solar energy. Its total reserve is equivalent to 4.03 × 108 EJ/year in 10 km of the crust surface layer [1]. Geothermal power generation is a promising application direction since lots of geothermal fields are far from the residential and commercial areas, and direct heat utilization is difficult. Organic Rankine cycle (ORC) is a promising geothermal power generation technology with advantages of high efficiency, simplicity, stability, flexibility, long service life, and wide installed capacity range [2–4]. Improving the performance of ORC system can promote geothermal development, which has been an important research issue recently. Heat exchangers are the core equipment to realize the heat exchange in ORC systems, including evaporator and condenser. The performance of heat exchangers substantially impacts the optimization and performance of ORC system, because they generally have the largest exergy loss and high purchased cost ratio. The exergy loss ratio can exceed 70% [5], and the purchased cost ratio is 46–77% [6, 7]. Furthermore, the geothermal fluid is often highly corrosive and easy to scale. Stainless steel is generally used as the heat exchanger material to ensure a long service life of ORC system and decrease the failure rate. While, the low thermal conductivity and high cost of stainless steel greatly increase the purchased cost of heat exchanger, which brings a worse economic performance to the geothermal ORC system [8]. The results of Li et al. [8] indicated that the development cost of geothermal resources would be 11.7% higher than industrial waste heat at the same conditions, just because of corrosivity. The high purchased cost of heat exchangers becomes a key obstacle for promoting geothermal ORC system. The structure of heat exchanger is an important factor determining its heat transfer performance and purchased cost. Optimizing the heat exchanger structure is an attractive way to reduce the purchased cost. To optimize the heat exchanger structure, it is necessary to know its operating condition beforehand, which is determined by the operating parameters of ORC system. However, the optimal selections of operating parameters and thermo-economic performance of ORC system are also significantly affected by the heat exchanger structure. Therefore, for the geothermal ORC system, it is necessary to carry out the cooperative optimization of system parameters and heat exchanger structure to achieve optimal system performance. There are many studies on the optimization design of geothermal ORC system and heat exchanger structure. However, the optimizations at system and component levels are often independent of each other. In general, the system-level optimization will fix the heat exchanger structure beforehand or even simplify the heat exchanger to a fixed performance index (e.g., average heat transfer coefficient, pinch point temperature difference, etc.). The optimization of heat exchanger structure is restricted to specific operating conditions but usually not the optimal operating conditions of ORC system. The independent optimization at the system and component levels is difficult to obtain a real globally optimal solution, but only obtain a sub-optimal solution in most cases.

Cooperative Optimization of System Parameters …

87

That will limit the thermo-economic performance improvement of geothermal ORC system, and even seriously underestimate the development potential of geothermal resources to obtain a misleading conclusion. To fully exploit the performance improvement potential of geothermal ORC system, a cooperative optimization of system parameters and heat exchanger structure is proposed. The influences of heat exchanger structure on the optimum design and thermo-economic performance of ORC system are analyzed. The performance improvement effects of cooperative optimization way are evaluated. This work can provide important guidance for geothermal ORC systems to achieve higher efficiency and lower cost.

2 Methods 2.1 System and Heat Exchanger As presented in Fig. 1, the typical subcritical ORC system is selected. The system is composed of the centrifugal feed pump, radial-flow turbine, and shell-and-tube heat exchangers with countercurrent flow. The shell-and-tube heat exchanger has advantages of easy maintenance, good reliability, large capacity, and low cost, and is widely adopted in geothermal ORC systems [9, 10]. The organic fluid flows in tubes, whereas the cooling water and geothermal fluid flow on the shell-side. The inner diameter and wall thickness of heat exchange tube are key structural parameters affecting the heat transfer performance and purchased cost, selected as the studied structural parameters. The selection schemes of the inner diameter and wall thickness are shown in Table 1, and they can guarantee safety in operating pressure ranges [11].

(b)

Temperature, T

1 6 2 2s 5

3 4

Tcool,in

Cooling water

Entropy, s

Fig. 1 Diagrams of typical subcritical ORC: a System layout; b Cycle process

Tcool,out Tcool,pp

88

J. Li et al.

Table 1 Selection schemes of the inner diameter and wall thickness Structural parameter

Value

Inner diameter, d i /mm

5, 10, 20

Wall thickness, δ w /mm

0.5, 1.0, 1.5, 2.0, 2.5, 3.0

Table 2 Properties of organic fluids Organic fluid

Critical pressure, pc /MPa

Critical temperature, T c /°C

R1234yf

3.38

94.7

R1234ze(E)

3.63

109.36

R600a

3.63

134.66

R245fa

3.65

153.86

2.2 Working Conditions The geothermal water is 100–150 °C with 3–10 kg·s−1 . The cooling water is 20 °C, and the temperature increase in condensing process is 5 °C. Four common organic fluids are selected, listed in Table 2. The system is in a steady state, and the pressure drops in connecting pipes and heat exchangers and the heat loss of fluids are ignored [12].

2.3 Performance Models The thermodynamic models are based on the first law of thermodynamics and the pinch point analysis method. The calculation models of power outputs and consumptions are shown in Table 3. The efficiency of turbine (ηT ), efficiency of feed pump (ηP ), and head of cooling water pump (H) are selected as 0.8, 0.75, and 10 m, respectively. In heat exchangers, the inlet velocities of liquid and vapor phases are 1 m·s−1 and 8 m·s−1 , respectively [11]. The material is stainless steel. The method and detailed heat transfer models to calculate heat exchange areas are similar to those in Ref. [8]. Table 3 Calculation models of power outputs and consumptions Power

Calculation model

Output of turbine, W T /kW

WT = m˙ O (h 1 − h 2 ) = m˙ O (h 1 − h 2s )ηT

Consumption of feed pump, W P /kW

WP = m˙ O (h 5 − h 4 ) = m˙ O (h 5s − h 4 )/ηP

Consumption of cooling system, W cool /kW

Wcool = m˙ cool g H , m˙ cool =

Net output of system, W net /kW

Wnet = WT − WP − Wcool

m˙ O (h 3 −h 4 ) h cool,pp −h cool,in

Cooperative Optimization of System Parameters …

89

The economic models are from the prediction method proposed by Turton et al. [13]. In an ORC system, the purchased equipment cost (PEC) can be assessed as:   P EC=C 0p FBM =C 0p B1 +B2 FM F p

(1)

 2 log10 C 0p = K 1 + K 2 log10 Y + K 3 log10 Y

(2)

 2 log10 F p = C1 + C2 log10 p + C3 log10 p

(3)

where Y represents the heat transfer area of heat exchanger or the power of turbine and feed pump, and the values of parameters can refer to Ref. [8]. Considering inflation, the PEC of component will be amended as: P EC2019 = P EC2001 (C E PC I2019 /C E PC I2001 )

(4)

where CEPCI 2019 is 607.5, and CEPCI 2001 is 397. Considering other additional costs, the total investment cost (TIC) of system is [14]: T I Csys = 1.18

n 

P ECi = 1.18P ECtotal

(5)

i=1

The thermo-economic index, specific investment cost (SIC), is calculated as [15]: S I C = T I Csys /Wnet

(6)

2.4 Optimization In this work, the system parameters and heat exchanger structure are cooperatively optimized to achieve the minimum SIC for the ORC system. The optimized system parameters include the evaporating pressure (pe ) and evaporator outlet temperature (T 1 ). The optimized heat exchanger structural parameters are the inner diameter and wall thickness of heat exchange tubes. The pinch point temperature difference (PPTD) of heat exchanger is also optimized, and its selectable range is 5–15 °C [13]. The selectable range of pe is from pcond + 100 kPa (100 kPa higher than condensing pressure) to 0.9pc . The lower limit of T 1 avoids droplets in the turbine [3], and the upper limit is equal to the geothermal temperature minus PPTD. The properties of fluids are from the REFPROP 10.0. The optimization processes and validation can refer to our previous work [8], and they are not described here to avoid repetition.

90

J. Li et al.

3 Results and discussion 3.1 Influence Rules of Inner Diameter and Wall Thickness The qualitative influences of inner diameter and wall thickness on the minimum SIC of ORC system are similar for different geothermal conditions and organic fluids, and Fig. 2 is taken as an example. Given the inner diameter, the smaller the wall thickness, the lower the SIC of ORC system is. Reducing the wall thickness can increase the overall heat transfer coefficient and reduce the heat exchange area, and also help to reduce the material consumption at the same heat exchange area, and thereby obtaining a lower purchased cost. The influences of wall thickness on the SIC will become more significant at a lower inner diameter. As the wall thickness decreases from 3 to 0.5 mm, the SIC decreases by 14.5% at the inner diameter of 5 mm, much larger than that at the inner diameter of 20 mm (7.4%). Results prove that the wall thickness significantly impacts the thermo-economic performance of ORC system, and its reasonable selection is vital in the system design. Especially for the performance comparisons between ORC technology and other power technologies (such as Kalina cycle and Stirling cycle), the selection of inner diameter may change the final comparison results, thus directly determining the application potential of ORC technology. Reducing the inner diameter is conducive to improving the thermo-economic performance when the wall thickness is low. The SIC can decrease by 2.0% by reducing the inner diameter when the wall thickness is 0.5 mm. However, reducing the inner diameter alone does not necessarily reduce the SIC. As shown in Fig. 2, reducing the inner diameter will lead to the increase of SIC and the deterioration of thermo-economic performance, when the wall thickness exceeds 1.5 mm. The main reason is that the Reynolds number of evaporation heat transfer decreases with decreasing inner diameter, which reduces the overall heat transfer coefficient and

Fig. 2 Influences of the inner diameter and wall thickness on the minimum SIC of ORC system

Cooperative Optimization of System Parameters …

91

increases the required heat transfer area. The larger the wall thickness, the greater the increase of SIC caused by reducing the inner diameter, and the more obvious the thermo-economic deterioration is. At the wall thickness of 3 mm, the SIC will increase by 6.1% with decreasing the inner diameter. Hence, the reasonable selection of inner diameter should consider the value of wall thickness. According to the results in this work, reducing inner diameter is conducive to improving the thermo-economic performance when the ratio of wall thickness to inner diameter is no more than 0.05. The optimal system parameters will decrease with decreasing the wall thickness. The maximum decreases of the optimal condensing pressure, evaporating pressure, evaporator outlet temperature, and PPTDs are 36.0 kPa (relative decrease is 5.6%), 111.7 kPa (relative decrease is 4.3%), 2.4 °C, and 2 °C, respectively. The smaller the inner diameter, the more remarkable the impacts of wall thickness on the optimal system parameters are. Low operating pressures are beneficial to safety, which provides favorable conditions for using a smaller wall thickness. The influences of inner diameter on the optimal system parameters are still closely related to the wall thickness. The optimal system parameters tend to decrease with decreasing the inner diameter when the wall thickness is low; while, they tend to increase when the wall thickness is large. While, the variations of optimal system parameters are low with decreasing inner diameter, and the influence is relatively weak. Furthermore, the reasonable selections of the inner diameter and wall thickness can decrease minimum SIC and increase net power simultaneously, and thereby effectively improving the performance and application potential of ORC technology. At the situations in Fig. 2, the minimum SIC decreases by 6.1%, and the net power increases by 2.6% by appropriately reducing the inner diameter and wall thickness. The benefits are very attractive. The results also indicate that the conclusions and quantitative results based on the specific inner diameter and wall thickness may not apply to other inner diameter and wall thickness situations because there may be a large gap in thermo-economic performance. Otherwise, it will easily lead to misleading conclusions.

3.2 Effects at Different Geothermal Temperatures As shown in Fig. 3, the decrement of minimum SIC obtained by reducing the wall thickness will increase with decreasing geothermal temperature. The decrement of SIC is only 7.3–14.3% at 150 °C, but it increases to 8.0–15.6% at 100 °C. It shows that the lower the geothermal temperature, the greater the SIC benefit obtained by reducing the wall thickness. The reason is that the proportion of the purchased cost of heat exchangers increases with decreasing geothermal temperature in ORC system, and the impacts of heat exchanger performance on system thermoeconomic performance become more remarkable. Therefore, the quantitative effect of heat exchanger structural parameters is also enhanced. The optimum evaporating and condensing pressures decrease with decreasing geothermal temperature, which provides favorable conditions for using a lower wall thickness.

92

J. Li et al.

Fig. 3 Decrements of SIC obtained by reducing the wall thickness (from 3 to 0.5 mm) at different geothermal temperatures

On the other hand, the SIC benefits obtained by reducing the inner diameter are shown in Fig. 4. Therein, the positive decrement represents that reducing the inner diameter is beneficial to improve the thermo-economic performance. On the contrary, the negative decrement represents that reducing the inner diameter will deteriorate the thermo-economic performance. The variations of SIC will increase with decreasing geothermal temperature in general, which indicates that the lower the geothermal temperature, the more remarkable the impacts of inner diameter. By reducing the inner diameter, the minimum SIC can decrease by 2.0–2.3% when the wall thickness is 0.5 mm, but it will increase by 6.0–6.5% when the wall thickness is 3 mm. In summary, the lower the geothermal temperature, the more significant the inner diameter and wall thickness influences the thermo-economic performance, and the greater the SIC benefits can be obtained by optimizing heat exchanger structure. The impacts of wall thickness on the optimal condensing pressure and PPTDs generally

Fig. 4 Decrements of SIC obtained by reducing the inner diameter (from 20 to 5 mm) at different geothermal temperatures

Cooperative Optimization of System Parameters …

93

enhance with increasing geothermal temperature. The quantitative influences of inner diameter on the optimal system parameters are different but the differences are very low at different geothermal temperatures.

3.3 Effects at Different Geothermal Flow Rates The minimum SIC is more sensitive to the change of wall thickness with increasing the geothermal flow rate, as presented in Fig. 5. Reducing the wall thickness, the minimum SIC decreases by 6.4–12.5% at 3 kg·s−1 , but the decrement is as high as 7.6–14.6% at 10 kg·s−1 . The proportion of the purchased cost of heat exchangers decreases as the geothermal flow rate increases. While, the cost decrement obtained by reducing the wall thickness becomes larger, and this impact on the SIC is more remarkable. Therefore, the decrement of SIC obtained by reducing the wall thickness increases with increasing the geothermal flow rate. On the other hand, the larger the geothermal flow rate, the more sensitive the SIC to the change of inner diameter, as illustrated in Fig. 6. With the flow rate increasing from 3 to 10 kg·s−1 , the decrement of SIC obtained by reducing the inner diameter increases from 1.7 to 2.0% at the wall thickness of 0.5 mm. While, the increment of SIC will increase from 2.6 to 3.2% at the wall thickness of 2 mm. In summary, the SIC benefits obtained by optimizing heat exchanger structure will be larger with increasing the geothermal flow rate. The quantitative effects of the inner diameter and wall thickness on optimal system parameters only differ slightly at various flow rates.

Fig. 5 Decrements of SIC obtained by reducing the wall thickness (from 3 to 0.5 mm) at different geothermal flow rates

94

J. Li et al.

Fig. 6 Decrements of SIC obtained by reducing the inner diameter (from 20 to 5 mm) at different geothermal flow rates

3.4 Effects for Different Organic Fluids The minimum SIC decreases by 7.5–15.9% for four organic fluids by reducing the wall thickness, and the decrements exceed 2% by reducing inner diameter, which substantially improves the thermo-economic performance. Within the studied scope of this work, the compared results in minimum SIC of four organic fluids will not change in the cases of different inner diameters and wall thicknesses. The decrements of SIC obtained by reducing the wall thickness are relatively close for four organic fluids. Although the lower the inner diameter, the larger the difference in SIC decrement of different organic fluids; while, the largest difference is less than 1.3%. Reducing the inner diameter, the decrements of SIC are different for various organic fluids, but the absolute difference is also very low, less than 0.9%. In summary, although the quantitative impacts of the inner diameter and wall thickness are different for various organic fluids, the differences are slight. In addition, the lower the critical temperature of organic fluid, the larger the decrement of optimal evaporating pressure with decreasing the wall thickness, and the more sensitive the optimal evaporating pressure and evaporator outlet temperature to the inner diameter.

3.5 Benefits of Cooperative Optimization The thermo-economic comparisons between traditional optimization (only systemlevel optimization) and cooperative optimization are shown in Fig. 7. The cooperative optimization of system parameters and heat exchanger structure can substantially enhance the thermo-economic performance of geothermal ORC system, compared

Cooperative Optimization of System Parameters …

95

Fig. 7 Thermo-economic comparisons between traditional optimization and cooperative optimization: a Different geothermal temperatures; b Different geothermal flow rates

to the traditional optimization. The minimum SIC can decrease by 12.5–15.6% at the scenarios of Fig. 7. The superiority of cooperative optimization will enhance with decreasing geothermal temperature and increasing geothermal flow rate. The substantial decrease of SIC will enhance the competitive advantage of ORC technology in geothermal energy development. The cooperative optimization of system parameters and heat exchanger structure will become a new trend in geothermal ORC systems.

4 Conclusions In this work, the cooperative optimization of system parameters and heat exchanger structure is carried out for the geothermal ORC system. The influences of the inner diameter and wall thickness of heat exchange tube on the optimal system parameters and thermo-economic performance are analyzed for various geothermal conditions

96

J. Li et al.

and organic fluids. The thermo-economic benefits of cooperative optimization way are evaluated. The main conclusions are summarized as below. The inner diameter and wall thickness have non-negligible influences on the optimum design and thermo-economic performance of geothermal ORC system. Influence rules of the inner diameter and wall thickness are similar in various scenarios. Reducing wall thickness is conducive to improving the thermo-economic performance, but reducing inner diameter is beneficial only when the ratio of wall thickness to inner diameter is no more than 0.05. A larger thermo-economic benefit can be obtained by optimizing the inner diameter and wall thickness at low geothermal temperature and large geothermal flow rate. While, the thermo-economic benefits only differ slightly for various organic fluids. Cooperative optimization of system parameters and heat exchanger structure can substantially enhance the thermo-economic performance of geothermal ORC system, compared to the traditional system-level optimization. The decrease of minimum SIC can be 12.5–15.6%. Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant Nos. 52106017 and 51736005), Beijing Natural Science Foundation (Grant No. 3222031), State Key Laboratory of Clean Energy Utilization (Open Fund Project No. ZJUCEU2020013), National Postdoctoral Program for Innovative Talents (Grant No. BX20200178), and China Postdoctoral Science Foundation (Grant No. 2020M680548). Jian Li thanks the supports of Shuimu Tsinghua Scholar Program (Grant No. 2020SM013).

References 1. Tchanche BF, Petrissans M, Papadakis G (2014) Heat resources and organic Rankine cycle machines. Renew Sustain Energy Rev 39:1185–1199 2. Velez F, Segovia JJ, Martin MC, Antonlin G, Chejne F, Quijano A (2012) A technical, economical and market review of organic Rankine cycles for the conversion of low-grade heat for power generation. Renew Sustain Energy Rev 16(6):4175–4189 3. Li J, Liu Q, Ge Z, Duan YY, Yang Z (2017) Thermodynamic performance analyses and optimization of subcritical and transcritical organic Rankine cycles using R1234ze(E) for 100–200°C heat sources. Energy Convers Manage 149:140–154 4. Li J, Ge Z, Duan YY, Yang Z, Liu Q (2018) Parametric optimization and thermodynamic performance comparison of single-pressure and dual-pressure evaporation organic Rankine cycles. Appl Energy 217:409–421 5. Luo XL, Huang RL, Yang Z, Chen JY, Chen Y (2018) Performance investigation of a novel zeotropic organic Rankine cycle coupling liquid separation condensation and multi-pressure evaporation. Energy Convers Manage 161:112–127 6. Heberle F, Bruggemann D (2015) Thermo-economic evaluation of organic Rankine cycles for geothermal power generation using zeotropic mixtures. Energies 8(3):2097–2124 7. Li J, Yang Z, Hu SZ, Yang FB, Duan YY (2020) Thermo-economic analyses and evaluations of small-scale dual-pressure evaporation organic Rankine cycle system using pure fluids. Energy 206:118217

Cooperative Optimization of System Parameters …

97

8. Li J, Yang Z, Hu SZ, Duan YY (2021) Influences of fluid corrosivity and heat exchanger materials on design and thermo-economic performance of organic Rankine cycle systems. Energy 228:120589 9. Walraven D, Laenen B, D’haeseleer W (2014) Optimum configuration of shell-and-tube heat exchangers for the use in low-temperature organic Rankine cycles. Energy Convers Manag 83:177–187 10. Li J, Yang Z, Hu SZ, Yang FB, Duan YY (2020) Effects of shell-and-tube heat exchanger arranged forms on the thermo-economic performance of organic Rankine cycle systems using hydrocarbons. Energy Convers Manage 203:112248 11. Shi MZ, Wang ZZ (2014) Principle and design of heat exchangers. Southeast University Press, Nanjing 12. Sadeghi M, Nemati A, Ghavimi A, Yari M (2016) Thermodynamic analysis and multi-objective optimization of various ORC (organic Rankine cycle) configurations using zeotropic mixtures. Energy 109:791–802 13. Turton R, Bailie RC, Whiting WB, Shaeiwitz JA, Bhattacharyya D (2012) Analysis, synthesis and design of chemical processes. Prentice Hall, Old Tappan 14. Lecompte S, Lemmens S, Huisseune H, van den Broek M, De Paepe M (2015) Multi-objective thermo-economic optimization strategy for ORCs applied to subcritical and transcritical cycles for waste heat recovery. Energies 8:2714–2741 15. Quoilin S, Declaye S, Tchanche BF, Lemort V (2011) Thermo-economic optimization of waste heat recovery organic Rankine cycles. Appl Therm Eng 31(14–5):2885–2893

Thermodynamic Evaluation of CCS Waste Heat Recovery by Organic Rankine Cycle Ran Li, Zhen Yang , and Yuan-Yuan Duan

Abstract To reduce the efficiency penalty of CO2 Capture and Storage (CCS) for generation system, the organic Rankine cycle is used to recover the waste heat of CO2 compression process in CCS. Considering the evaporation temperature is the key parameter affecting the output work of ORC, the minimum work consumption of the compression system is selected as the optimization objective. Simultaneously, eight kinds of fluids are considered, and R1233zd(E) has the best performance. The overall compression work consumption is 21,804.1 kW and the maximum energy saving rate is 27.7%. In addition, the mass flow rate of organic fluid is considered. The results show that he application of R601 is also a better choice because its energy-saving rate can reach 25.1%, which is a small difference from the maximum value, but the mass flow rate will be greatly reduced. The conclusions indicate that using waste heat recovery technology can effectively mitigate compression power consumption and improve system efficiency. Keywords IGCC · Carbon capture · Multistage compression · ORC · Heat recovery

1 Introduction With the growth of world energy demand and the severe challenge of carbon emission, power production is an important industry of carbon emission reduction. To achieve the “2 °C” target in the Paris Agreement, improving the efficiency of the existing power generation system, promoting renewable energy power generation, and increasing CO2 Capture and Storage (CCS) units are important measures. By 2020, global coal power generation still accounts for 38% [1]. Coal-rich countries such as China, India, and Pakistan et al., coal power generation is still the main electric generation way. However, coal is an important source of greenhouse gas emissions. To solve this problem, many experts and scholars have proposed and R. Li · Z. Yang (B) · Y.-Y. Duan Tsinghua University, Beijing 100084, People’s Republic of China e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_9

99

100

R. Li et al.

demonstrated that using coal power generation with CCS is an effective technology to reduce greenhouse gas emissions [2–4]. Although the addition of CCS solves the problem of carbon emission, it will cause great efficiency and economic punishment to the original power system. Coal-fired power generation technologies include gasifier, subcritical pulverized coal furnace, and supercritical pulverized coal furnace. The integrated gas combined cycle system built by gasifier and the combined cycle has higher energy efficiency. The research of the National Energy Technology Laboratory (NETL) shows that [5] when different coal-fired power generation technologies are coupled with CCS, the net efficiency penalty of the IGCC system is the smallest, about 7.3%, and the efficiency penalties of the other two power generation technologies are 8.6% and 9.8%. Simultaneously, with the development of gas turbines, the efficiency improvement potential of the combined cycle is greater in the future. IGCC with the CCS system can realize the demand for efficient and low-carbon coal utilization. For the IGCC system, the coupling of CCS causes a huge efficiency penalty. At present, capture technology mainly includes the physical solvent or chemical solvent absorption and membrane separation technology. The captured CO2 needs to be compressed before transportation and storage. Multi-stage compression and interstate cooling are usually adopted in engineering applications to reduce work consumption for the compression process. However, the literature shows that the compression work still accounts for about 17% of the total auxiliary power [5], which greatly impacts the overall generation efficiency. This paper proposes recovering the heat between compressor stages through thermal power conversion to realize the step utilization of energy. The CO2 flow between different stages belongs to the medium and low-grade heat source (200– 300 °C). So, the conventional steam Rankine cycle is no longer applicable. Under the low-temperature heat source condition, the organic Rankine cycle (ORC) can achieve higher thermal efficiency due to its low working medium critical parameters and various types [6, 7]. ORC can be divided into supercritical and subcritical systems according to evaporation pressure. For supercritical ORC, the evaporation pressure of working fluid is above the critical point, which can avoid the two-phase heat exchange process. However, the density of the supercritical fluid is close to liquid, and the viscosity is close to gas, the physical property changes are more complex, and the related expander technology is still in the development stage and immature; In addition, supercritical ORC has high requirements for the heat source. In contrast, subcritical ORC is more widely used because of its simple structure, safety, reliability, low investment cost, and operation cost. There are many factors affecting the performance of ORC, and the evaporation temperature is the key factor. Subcritical ORC mainly comprises evaporation, expansion, condensation, and compression processes. Studies show that the exergy damage in the evaporator exceeds 60% of the total cycle [8, 9]. Therefore, for the ORC system, evaporation temperature greatly impacts system efficiency. In addition, the

Thermodynamic Evaluation of CCS Waste Heat …

101

temperature of CO2 is different between different compressor stages. The evaporation temperature needs to be optimized to increase the temperature matching degree between organic working fluid (OWF) and heat source. On the other hand, evaporation temperature affects the inlet temperature of CO2 in the next compressor stage. However, it is not clear how to set the evaporation temperature of OWF between each stage. In previous studies, the same evaporation temperature is often used, so the system design that minimizes the work consumption of the system cannot be obtained, and the maximum waste heat recovery cannot be realized. This paper takes the system work consumption as the research objective, optimizes the evaporation temperature of OWF in different interstate heat exchangers to obtain the optimal fluid and design of the system.

2 Simulation 2.1 System Construction The system structure layout is shown in Fig. 1. with four stages of compression, three intercoolers, and an outlet cooler. Keep the interstate compression ratio is constant. The CO2 inlet temperature is 40 °C, the inlet pressure is 0.3 MPa, the outlet pressure is 15 MPa. The system operating parameters are shown in Table 1. Table 2 shows the critical temperature, critical pressure, global warming potential value (GWP), and ozone destruction potential value (ODP) of eight organic fluids.

Fig. 1 System configuration

102

R. Li et al.

Table 1 ORC system parameters Parameters

Value

Turbine isentropic efficiency (X t ) /%

84.4

Evaporator pinch temperature difference/°C

10

Evaporator pinch temperature difference/°C

5

Cool feed inlet temperature/°C

20

Pump isentropic efficiency (X p )/%

75

CO2 flowrate/(kg/s)

98

Superheat degree/°C

5

Compressor isentropic efficiency (X c )/%

84

Environment temperature/°C

20

Table 2 Thermodynamic parameters of different organic working fluids Fluid

T c /°C

pc /MPa

GWP

ODP

R601

196.6

3.37

< 20

0

R601a

187.2

3.38

< 20

0

R245ca

174.4

3.93

693

0

R1233zd(E)

165.6

3.57

7

0

R1224yd(Z)

155.5

3.33

0.88

0

R600

152.0

3.80

20

0

R600a

134.7

3.63

20

0

R236ea

139.3

3.50

710

0

2.2 Thermodynamic Process The thermodynamic process is CO2 is compressed in four stages, and heat is released after each stage of compression, used as the heat source of ORC. The OWF is heated to the superheated state in the heat exchanger, and the superheated steam does work through the turbine. Turbine outlet fluid is condensed to the saturated liquid state in the condenser and then pressurized in working fluid pumps. The definitions of thermal insulation efficiency of OWF pump, CO2 compressor, and turbine are respectively shown below X p = (h outs − h in )/(h out − h in )

(1)

X c = (h outs − h in )/(h out − h in )

(2)

X t = (h in − h out )/(h in − h outs )

(3)

Thermodynamic Evaluation of CCS Waste Heat …

103

where, hin , hout , houts represent the actual enthalpy of the fluid at the inlet of the component, the actual enthalpy of the fluid at the outlet of the component, and the isentropic enthalpy of the fluid at the outlet of the component. The work consumption of the system is Wnet = Wt − Wp − Wc − Wcool

(4)

where W t , W p , W c , W cool respectively represent the work of the turbine, OWF pump, compressor, and cooling water pump.

2.3 Optimization Algorithm This paper uses the Genetic Algorithm (GA) for optimization calculation. GA is a common method for solving complex nonlinear problems. It has been widely used in engineering optimization design [10, 11]. Therefore, GA is used for optimization calculation in this paper. The evaporation temperature of OWF in four heat exchangers is taken as the independent variable, and the net output work is carried as the dependent variable. The thermophysical property data of the fluid is from refprop10.0. The following assumptions are adopted: (a) (b)

The system operates under steady-state conditions. Ignore the heat loss and flow pressure drop in the heat exchanger.

3 Results In this paper, eight fluids are used. The minimum work of the whole compression system is obtained by optimizing the evaporation temperature. Results show that the optimal evaporation temperature of each fluid is obtained. The optimization results show that when the net output work of the overall system is taken as the optimization objective, the mass flow of OWF in the second heat exchanger corresponding to the optimal system design is 0, indicating that there is no need for the second intercooler. The optimal evaporation temperatures for the three heat exchangers are T1 = 359.7 K, T3 = 430 K, and T4 = 328.6 K. Therefore, the three-stage compressor is the optimal system design. The system work consumption is -21,804 kW. The proportion of turbine work, OWF pump work, cooling water pump work, and CO2 compression work in the system is shown in Fig. 2 The work consumption of the cooling water pump and OWF pump is very small, and the sum of the two is less than 2%. Turbine work of the ORC system can offset more than 20% of the work consumption of the overall compression system, indicating that the use of ORC to recover waste heat can effectively reduce the CO2 compression work. This paper also compares this system with the system without ORC waste heat recovery. The comparison results show that when the intercooler reaches the same

104

R. Li et al.

Fig. 2 Distribution of absolute value of net output work of pressure components in the system

interstage temperature and there is no ORC heat recovery, the CO2 compression work is 30171 kW, and the increased cooling water flow rate is 1110.2 kg/s. The calculation results of several other OWF are displayed. Figure 3 shows the work distribution of different pressure components with other fluids. The figure also lists the minimum work consumption of the system that can be achieved by using various fluids. R1233zd(E) is the optimal fluid when only considering the work consumption. The work consumption of the CO2 compression system is 21804.1 kW, the system energy-saving rate is 27.7%, and the performance of R600 is the worst. The system power consumption is 24464.4 kW, and the energysaving rate is 16.3%.

Fig. 3 The system’s power distribution with different organic working fluids (OWF)

Thermodynamic Evaluation of CCS Waste Heat …

105

Fig. 4 The flow of organic working fluid (OWF) and corresponding net output work of the system

Figure 4 shows the relationship between the minimum system work consumption and the mass flow using different fluids. Although R1233zd(E) can achieve the minimum system work consumption, the OWF mass flow consumed is also much larger than the other fluids, indicating that if only consider the process of the ORC, the thermal efficiency of this cycle is not the largest. In addition, considering the consumption of OWF, the application of R601 is also a better choice because its energy-saving rate can reach 25.1%, which is a small difference from the maximum value. However, the mass flow will be greatly reduced (Table 3). Table 3 The maximum energy-saving rate of the different fluids Fluid

Energy saving rate (%)

R601

25.10

R601a

23.60

R245ca

21.50

R1233zd(E)

27.70

R1224yd(Z)

21.10

R600

16.30

R600a

16.70

R236ea

16.60

106

R. Li et al.

4 Conclusion The optimization results indicate that the waste heat recovery method has a significant impact on system efficiency and is worth popularizing in the future. The main conclusions are as follows: (a) (b)

(c)

The optimization results show that the waste heat can be recovered more efficiently by optimizing evaporation temperature. The minimum work consumption of the system when different working fluids are applied is obtained, in which the performance of R1233zd(E) is the best. The overall compression work consumption is 21804.1 kW and the maximum energy saving rate is 27.7%. Comparing the relationship between the minimum work consumption and the mass flow rate of the organic working fluid, R1233zd(E) can achieve the minimum work consumption, but the required mass flow is also the largest. If the mass flow and work are comprehensively considered, R601 has great application potential.

Acknowledgements This work is supported by National Science and Technology Major Project(J2019-I-0009-0009).

References 1. Bp. Energy Outlook 2020 (2021) 2. Ren S, Feng X, Wang Y (2021) Energy evaluation of the integrated gasification combined cycle power generation systems with a carbon capture system. Renew Sustain Energy Rev 147:111208 3. Szima S, Arnaiz Del Pozo C, Cloete S et al (2021) Finding synergy between renewables and coal: flexible power and hydrogen production from advanced IGCC plants with integrated co2 capture. Energy Convers Manage 231:113866 4. Georgousopoulos S, Braimakis K, Grimekis D et al (2021) Thermodynamic and technoeconomic assessment of pure and zeotropic fluid orcs for waste heat recovery in a biomass IGCC plant. Appl Therm Eng 183:116202 5. James Iii RE, Kearins D, Turner M (2019) Cost and performance baseline for fossil energy plants volume 1: bituminous coal and natural gas to electricity. NETL 6. Pethurajan V, Sivan S, Joy GC (2018) Issues, comparisons, turbine selections and applications – an overview in organic rankine cycle. Energy Convers Manage 166:474–488 7. Xu B, Rathod D, Yebi A et al (2019) A Comprehensive review of organic rankine cycle waste heat recovery systems in heavy-duty diesel engine applications. Renew Sustain Energy Rev 107:145–170 8. Altun AF, Kilic M (2020) Thermodynamic performance evaluation of a geothermal orc power plant. Renew Energy 148:261–274 9. Sun Q, Wang Y, Cheng Z et al (2020) Thermodynamic and economic optimization of a double-pressure organic rankine cycle driven by low-temperature heat source. Renew Energy 147:2822–2832

Thermodynamic Evaluation of CCS Waste Heat …

107

10. Roetzel W, Luo X, Chen D (2020) Optimal design of heat exchanger networks, pp 231–317 11. Wang J, Qi X, Ren F et al (2021) Optimal design of hybrid combined cooling, heating and power systems considering the uncertainties of load demands and renewable energy sources. J Clean Prod 281:125357

Effects of Initial Pressure on the Explosion Temperature Peak of Ethanol-Air Mixture and the Time Difference Between Temperature and Pressure Peak Xiaoyao Ning, Xuehui Wang, and Jian Wang Abstract In the previous studies, the flammability limits and explosion pressure characteristics were the main topics of the explosion of combustible gas or vapor and the temperature change characteristics in the explosion process were rarely studied. Therefore, this paper takes ethanol as the research object and studies the influence of initial pressure on the temperature peak and the time difference between temperature and pressure peak in the process of ethanol-air mixture explosion at different ambient temperatures. This work can provide some reference for the similar research in the future. Keywords Ethanol · Explosion characteristics · Explosion temperature peak · Time to the peak · Time difference

1 Introduction Ethanol is a common compound widely used in food industry, medical and health, chemical industry and other fields. As a flammable liquid, ethanol vapor in a certain concentration range will explode when it is mixed with air and encounters an ignition source. Because the ambient temperature and pressure in production and processing are often different from room temperature and standard atmospheric pressure, researchers have studied the explosion characteristics of ethanol vapor under different ambient temperature and pressure to determine the influence of different initial conditions on the explosion characteristics of ethanol. Brooks et al. [1] determined the explosion range of four combustible liquids including ethanol at room temperature and standard atmospheric pressure through experiments, and predicted the explosion range by using CAFT method. Coronado et al. [2, 3] reviewed the literature on the flammability limits of combustible mixtures focused on ethanol, discussed the experimental steps to determine the flammability limit, and gave the X. Ning · X. Wang · J. Wang (B) State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei, Anhui 230026, China e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_10

109

110

X. Ning et al.

flammability limits of ethanol under atmospheric pressure. After that, the flammability limits of hydrated and anhydrous ethanol at different temperatures were studied experimentally at reduced pressures. Wan et al. [4] predicted the lower flammability limits of methanol, ethanol, methyl format and dimethyl ether in air under high temperature and pressure based on the calculated adiabatic flame temperature, and discussed the influence of initial pressure and temperature on the lower flammability limit. Based on the experimental results, Velasquez et al. [5] proposed a flammability limit prediction model for ethanol-air mixture with different water content by using Kriging regression model and response surface method. Li et al. [6] conducted experiments on the explosion characteristics of five alcohol-air mixtures at different initial temperatures and pressures, and studied the effects of temperature and pressure on different explosion behaviors of various alcohols. Mitu et al. [7] experimentally studied the explosion pressure characteristics of ethanol-air mixtures with different concentrations at different initial pressures, initial temperatures and explosion vessels. Li et al. [8] conducted a series of experiments on ethanol-gasoline with different proportions and analyzed the explosion characteristics. Xu et al. [9] studied the explosion characteristics of hydrous bio-ethanol in a constant volume combustion chamber under different initial pressures, initial temperatures, oxygen concentrations and equivalence ratios. All the above studies focus on the flammability limit of ethanol and the pressure change in the explosion process, but few of them focus on the temperature change in the ethanol explosion process. However, the high temperature caused by the explosion can also cause great damage to personnel and equipment. Therefore, in this work, ethanol explosion experiments at different initial pressures were carried out at the lower flammability limits of ethanol of different temperatures. The explosion temperature peaks of ethanol-air mixture under different initial conditions were discussed, and the time difference between temperature and pressure peak were compared.

2 Experiment 2.1 Experimental System Figure 1 is a schematic diagram of the composition of the experimental system. The experimental system consists of five parts, including the explosion vessel, gas distribution module, ignition module, heating module and detection module. The explosion vessel is a 5L stainless steel sphere in which the explosion experiment is carried out. The gas distribution module is responsible for the preparation of the gas mixture required by the experiment. The vacuum pump is responsible for exhausting the exhaust gas of the last experiment and creating a vacuum environment, the stirrer is responsible for mixing samples and air evenly. The ignition module provides ignition energy for the mixed gas through spark ignition. The ignition voltage is 15 kV, the current is 20 mA, and the distance between the two electrodes is 3 mm. The heating

Effects of Initial Pressure on the Explosion Temperature …

111

Fig. 1 Schematic diagram of experimental system

module increases the temperature inside the explosion vessel by heating the air outside. The detection module consists of two pressure sensors and a temperature sensor, which are used to prepare the gas mixture and record the data of pressure and temperature changing with time during the experiment, and transmit the data to the computer through a data acquisition unit.

2.2 Procedure Preparation: Check the air tightness of the explosion vessel to ensure that the air tightness is good; Empty the vessel, pump the pressure in the explosion vessel below 2 kPa and restore it to atmospheric pressure, for two consecutive times. Experiment: Open the heating module and heat up to the experimental temperature; Pump the pressure in the explosion vessel below 1.3 kPa; Use partial pressure method to prepare the gas mixture for the experiment; Turn on the stirrer and stir for 5 min, and the stirrer speed should not be less than 400r/min; Ignition; Collect and output the experimental pressure and temperature data; Empty the vessel, pump the pressure in the explosion vessel below 2 kPa and restore it to atmospheric pressure, for two consecutive times.

2.3 Sample and Conditions Ethanol with content greater than or equal to 99.7% was used as the experimental sample without further purification.

112

X. Ning et al.

In this paper, the initial temperatures of 60, 90, 120 and 150 °C were taken as initial temperatures, and the explosion experiments of ethanol vapor at 60 kPa, 80 kPa and 100 kPa were carried out at each temperature. The concentration of ethanol vapor was controlled at the lower flammability limit of this temperature. The lower flammability limit of each condition is as follows. Table 1 Lower flammability limit of ethanol vapor Initial temperature (°C)

Initial pressure (kPa)

LFL (Vol%)

60

60

3.6

90

120

150

80

3.6

100

3.6

60

3.8

80

3.7

100

3.6

60

3.5

80

3.6

100

3.5

60

3.1

80

3.2

100

3.0

3 Results and Discussion In order to ensure the reliability of experimental data, at least three experiments should be carried out under each condition. Taking the pressure and temperature curve of an experiment when the initial temperature was 150°C and the initial pressure was 80 kPa as an example, the parameters discussed in this paper were visually explained, as shown in Fig. 2. Figure 2 shows the curves of explosion pressure and temperature changing with time within 5 s after ignition. It can be seen from the figure that, after ignition, pressure and temperature have the same trend of change over time. Both of them are rising rapidly to the peak value and then falling rapidly. However, by comparing the two curves of pressure and temperature, it can be found that the pressure rises and falls more sharply with time, and the pressure peak appears before the temperature peak. In this paper, the peak value of temperature curve is expressed by tmax , the time to the pressure peak is expressed by θ1 , the time to the temperature peak is expressed by θ2 , and the difference between θ2 and θ1 is expressed by △θ. In data processing, the data of multiple experiments under each condition are screened and then averaged to obtain the final values of the four parameters.

Effects of Initial Pressure on the Explosion Temperature …

113

Fig. 2 Pressure and temperature curves of an experiment at initial temperature of 150 °C and initial pressure of 80 kPa

3.1 Temperature Peak Figure 3 shows the variation trend of temperature peak with initial pressure at different initial temperatures. It can be seen from the figure that within the experimental range, the lowest temperature peak is about 460 °C and the highest temperature peak is about 610 °C. Except for the initial temperature of 60 °C, the temperature peaks at the other three initial temperatures basically increase with the increase of the initial pressure. This may be because the average intermolecular distance is smaller at higher initial pressures, resulting in faster chemical reaction rate, faster release of reaction heat, and therefore a higher temperature peak can be achieved under exactly the same external cooling condition. When the initial temperature is 60 °C, the temperature peak at the initial pressure of 80 kPa is too high, leading to a huge difference in trend with other initial temperatures, which may be caused by large experimental errors under

Fig. 3 Variation of temperature peak with initial pressure at different initial temperatures

114

X. Ning et al.

this condition. In addition, at the initial temperature of 90 °C, the temperature peak at 100 kPa was smaller than that at 80 kPa, which may also come from experimental error.

3.2 Time to Pressure and Temperature Peak Figure 4 shows the variation trend of θ1 and θ2 with initial pressure at four initial temperatures. It can be seen from the figure that θ1 and θ2 have the same trend of change: When the initial temperature is 60 and 90 °C, basically, θ1 and θ2 increase linearly with the increase of the initial pressure. When the initial temperature rises to 120 and 150 °C, the increases of θ1 and θ2 decreases. By comparing θ1 and θ2 , it can be found that θ2 is greater than θ1 in all conditions, which indicates that the mass transfer process is ahead of the heat transfer process in experimental process, and the difference between them represents the time difference between the pressure peak and the temperature peak. The difference value △θ was calculated under each condition, and its variation trend with the initial pressure was drawn as Fig. 5.

Fig. 4 Variations of θ1 and θ2 with initial pressure at different initial temperatures

Effects of Initial Pressure on the Explosion Temperature …

115

Fig. 5 Variation of △θ with initial pressure at different initial temperatures

It can be seen from Fig. 5 that △θ varies differently with the initial pressure at different initial temperatures. When the initial temperature is 60 and 120 °C, △θ decreases with the increase of the initial pressure. When the initial temperature is 90 and 150 °C, △θ is almost unaffected by the change of initial pressure. There may be two reasons for this phenomenon: Firstly, too few initial pressure points were selected to further determine the variation trend of △θ. Secondly, taking the lower flammability limit of each initial temperature as the experimental concentration may cause great difference in flame propagation process of repeated experiments and poor data repeatability, resulting in unclear variation trend of △θ.

4 Conclusion In this paper, ethanol explosion experiments with different initial pressures were carried out at the lower flammability limits of ethanol of different temperatures. The variation trend of explosion temperature peak (tmax ), the time to the pressure peak (θ1 ), the time to the temperature peak (θ2 ) and the difference between θ2 and θ1 (△θ) with the initial pressure are discussed. We found that: 1. tmax basically increases with the increase of initial pressure. However, the temperature peak at the initial temperature of 60 °C and the initial pressure of 80 kPa is too high, which may be caused by large experimental errors. 2. θ1 and θ2 have the same trend of variation: when the initial temperature is 60 and 90 °C, basically, θ1 and θ2 increase linearly with the increase of the initial pressure. When the initial temperature rises to 120 and 150 °C, the increase of θ1 and θ2 decreases. 3. θ2 is greater than θ1 in all conditions, indicating that the mass transfer process is ahead of the heat transfer process in the experimental process. 4. The variation trend of △θ with the initial pressure is not clear: when the initial temperature is 60 and 120 °C, △θ decreases with the increase of the initial pressure; When the initial temperature is 90 and 150 °C, △θ is almost unaffected by the change of initial pressure. This may be caused by too few initial pressure points and improper selection of ethanol vapor concentration. Therefore,

116

X. Ning et al.

further studies need to conduct experiments on more initial pressure points and more ethanol vapor concentrations. This work was supported by the National Science and Technology Major Project (J2019-VIII-0010–0171), Key R & D Program of Yunnan Province (202003AC100001) and the Fundamental Research Funds for the Central Universities under Grant No.WK2320000052.

References 1. Brooks MR (2007) Flammability envelopes for methanol, ethanol, acetonitrile and toluene. J Loss Prev Process Ind 20(2):144–150 2. Coronado CJ (2012) Flammability limits: a review with emphasis on ethanol for aeronautical applications and description of the experimental procedure. J Hazard Mater 241–242:32–54 3. Coronado CJ (2014) Flammability limits of hydrated and anhydrous ethanol at reduced pressures in aeronautical applications. J Hazard Mater 280:174–184 4. Wan X (2016) Numerical study of influence of initial pressures and temperatures on the lower flammability limits of oxygenated fuels in air. J Loss Prev Process Ind 41:40–47 5. Velasquez EIG (2017) Prediction of flammability limits for ethanol-air blends by the kriging regression model and response surfaces. Fuel 210:410–424 6. Li QQ (2015) Comparative assessment of the explosion characteristics of alcohol-air mixtures. J Loss Prev Process Ind 37:91–100 7. Mitu M (2017) Influence of pressure, temperature and vessel volume on explosion characteristics of ethanol/air mixtures in closed spherical vessels. Fuel 203:460–468 8. Li GC (2020) Experimental study on explosion characteristics of ethanol-gasoline blended fuels. J Loss Prev Process Ind 64(104083) 9. Xu CS (2020) Explosion characteristics of hydrous bio-ethanol in oxygen-enriched air. Fuel 271(117604)

Environmental and Chemical Engineering

Preparation and Photocatalytic Activity of Bi2 WO6 /RGO Composite Photocatalyst Ting Lin, Haiyan Fu, Yicheng Wu, Tian Chai, Guoxin Su, and Shuguang Wang

Abstract Bi2 WO6 and Bi2 WO6 /RGO composite synthetized by hydrothermal route in this study. The adsorption and photocatalytic properties of Bi2 WO6 and Bi2 WO6 /RGO composites for rhodamine B (RhB) and tetracycline hydrochloride (TC) were investigated under the irradiation of a 500 W xenon lamp. The results showed that a higher adsorption and photocatalytic properties using Bi2 WO6 /RGO composite compared with single Bi2 WO6 was measured. The photocatalytic removal calculation to RhB over Bi2 WO6 /RGO reached to 79% in photoreaction 120 min, and the total removal that exceeded 90%. The photocatalytic removal rate of TC by Bi2 WO6 /RGO increased to 90%, and the total removal effect exceeded 95%. In addition, the effect of different dosage of Bi2 WO6 /RGO composite on photocatalytic performance was studied. The results showed that the highest photocatalytic activity was achieved for 50 mg dosage of Bi2 WO6 /RGO composite. Keywords Photocatalysis · Bismuth tungstate · Graphene · Water treatment

1 Introduction In recent years, all kinds of dye wastewater and antibiotic wastewater discharged caused by population growth and industrial development, which has caused a serious menace to the natural environment and human health [1–3]. With the rapid development of the pharmaceutical industry, although the use of drugs brings people a lot of convenience, the discharge of antibiotic wastewater is becoming increasingly prominent. After the antibiotic wastewater is discharged into the water body, it remains in the water body for a long time, which is difficult to decompose and most of them are highly toxic. In addition, it can accumulate and enrich through the food chain, and T. Lin · H. Fu · Y. Wu · T. Chai · S. Wang (B) Key Laboratory of Environmental Biotechnology (XMUT), Fujian Province University, Xiamen University of Technology, Xiamen 361024, China e-mail: [email protected] G. Su Xiamen Ocean Vocational College, Xiamen 361102, China © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_11

119

120

T. Lin et al.

finally enter the human body, causing harm to human health. At the same time, trace drug compounds and their metabolites also pose potential risks to aquatic organisms and humans [4]. Tetracycline hydrochloride is one of the most common tetracycline antibiotics, because it is toxic and difficult to degrade, it will produce drug resistance genes in the environment, which is harmful to human health. Therefore, people pay attention to how to effectively remove this kind of antibiotic wastewater. At present, the common wastewater treatment methods include biological method [5], adsorption method [6], electrochemical treatment method [7] and so on. However, these methods either have long treatment time or easy to cause secondary pollution and other problems. Photocatalysis has become a research hotspot in wastewater treatment because of its high catalytic activity, photostability and low cost [8–10]. Bi2 WO6 is one of the Aurivillius bismuth-based visible light catalysts. The layered structure of Bi2 WO6 is regarded as an ideal visible light catalyst to replace TiO2 due to unique (WO4 )2− layers and (Bi2 O2 )2+ layers, which has the advantages of visible light response activity, high chemical stability and non-toxicity [11–13]. The layered structure of Bi2 WO6 increases the specific surface area and porosity of the catalyst, which makes more active sites between the layers. However, the photogenerated electron–hole of Bi2 WO6 is prone to reunite, and the separation efficiency of photogenerated carriers is generally low [14]. The two-dimensional material graphene (RGO) has a large specific surface area, which can improve the dispersibility and adsorption capacity of semiconductor photocatalytic materials [15]. At the same time, the outstanding electron mobility of RGO could promote the transfer rate of photogenerated carriers, thus cut down the reunite rate of electron–hole pairs and advance the photocatalytic efficiency of the materials [16, 17]. Therefore, graphene was chosen to compound with Bi2 WO6 . In this study, we prepared Bi2 WO6 photocatalyst and Bi2 WO6 /RGO composite photocatalyst by simple hydrothermal method. Graphene were doped into Bi2 WO6 for the enhancement of photocatalytic activities on dye wastewater and antibiotic wastewater. The photocatalytic degradation performance of the two photocatalysts for RhB and TC was investigated. In addition, the effect of different dosage of Bi2 WO6 /RGO composite on photocatalytic performance was studied.

2 Preparation of Experimental Materials 2.1 Experimental Materials Bi(NO3 )3 ·5H2 O, Na2 WO4 ·2H2 O, NaNO3 , H2 SO4 , KMnO4 , H2 O2 , BaCl2 , tetracycline hydrochloride (TC, analytical purity) and rhodamine B (RhB, analytical purity) were purchased from Sinopharm Chemical Reagent Co., Ltd.

Preparation and Photocatalytic Activity of Bi2 WO6 /RGO …

121

2.2 Preparation of Photocatalyst Synthesis of Bi2 WO6. Na2 WO4 · 2H2 O and Bi(NO3 )3 · 5H2 O were successively dissolved in 150 mL deionized water, which strongly stirred for 2 h to form suspension. The dosage of Na2 WO4 · 2H2 O and Bi(NO3 )3 · 5H2 O were respectively 1.5 mmol and 3 mmol. The mix liquid was then transferred to 100 mL Teflflon-lined stainless steel autoclaves, sealed, heated at 150 °C and reacted for 24 h. After the reaction was finished, the product was centrifuged, washed with deionized water and absolute ethanol. Finally, the product was dried at 60 °C overnight. Synthesis of Graphene. Graphene (RGO) was prepared by improved Hummers method [18]. Low temperature reaction stage: put the round bottom flask in ice water bath, add 1.00 g graphite and 0.50 g sodium nitrate (NaNO3 ), slowly add 23 mL concentrated sulfuric acid (H2 SO4 ), magnetic stirring reaction 30 min. Then 3.00 g potassium permanganate (KMnO4 ) was slowly added to the reaction system for several times and reacted for 2 h under magnetic stirring. In the middle temperature reaction stage, the ice water bath was converted into water bath, and then the temperature was raised to 35 °C to react 30 min. High temperature reaction stage: slowly adding 46 mL distilled water, heating up to 98 °C to react 15 min. Finally, 10 mL 30% hydrogen peroxide (H2 O2 ) was slowly added to neutralize the incomplete reaction of KMnO4 , and then added 70 mL distilled water. The graphene (RGO) was obtained by alternately centrifuging with distilled water, dilute hydrochloric acid and anhydrous ethanol until no SO4 2− was detected in the supernatant with barium sulfate (BaCl2 ) solution. After ultrasonic dispersion for 2 h, RGO was obtained by freeze-drying at −55 °C for 48 h. Synthesis of Bi2 WO6 /RGO. Moderate amount reduced graphene oxide and Na2 WO4 · 2H2 O were dissolved 50 mL deionized water in turn, and uniform brown aqueous solution A was formed by ultrasonic oscillation. Pre-determined amount of Bi(NO3 )3 · 5H2 O was put into deionized water, and until the solid substance dissolved by means of ultrasound to form solution B. Then, dribble solution B slowly into solution A, and they were completely mixed by ultrasound. The mix solution was devolved to a 100 mL Teflflon-lined stainless steel autoclave, and heated at 160 °C for 16 h. At the end of the reaction, waiting for the reactor to cool naturally to room temperature, filtered to dry.

3 Photocatalytic Degradation Experiment The photocatalytic activity of the prepared samples was evaluated by visible light degradation of rhodamine B and tetracycline hydrochloride. First of all, 0.005 g rhodamine B was dissolved in the volumetric flask of 1000 mL, and the rhodamine B solution with 5 mg/L was placed in a place away from light for use. Similarly,

122

T. Lin et al.

0.02 g tetracycline hydrochloride was dissolved in the volumetric flask of 1000 mL, and the 20 mg/L tetracycline solution was placed in a place away from light for use.

3.1 RhB Degradation Experiment The photocatalytic effect of the synthetized samples was tested in photochemical reactors under the irradiation of a 500 W xenon lamp, RhB was chosen as degradation product. The input of sample Bi2 WO6 and sample Bi2 WO6 /RGO was the same 40 mg, degraded 50 mL RhB solution (5 mg/L). Stirring 60 min under dark conditions makes the catalyst and dye molecules to achieve adsorption–desorption equilibrium. The light was then turned on to provide lighting conditions for photodegradation reaction, and the 2.5 mL reaction solution was taken at regular 60 min intervals. After the sample was filtered by 0.22 µm filter head, the sample liquid was detected by an ultraviolet spectrophotometer. According to the absorbance value of the test solution at 554 nm, the catalyst for degradation efficiency of RhB was determined.

3.2 Degradation TC Experiments The photocatalytic performance of the prepared samples was evaluated by visible light degradation effect of tetracycline hydrochloride. Catalyst Bi2 WO6 and Bi2 WO6 /RGO were used to degrade 50 mL TC (20 mg/L), the dosages of catalysts were 50 mg, the sampling interval was 30 min, and the wavelength was detected at 357 nm.

3.3 Optimization Experiment of Bi2 WO6 /RGO Dosage The optimal dosage of Bi2 WO6 /RGO sample with the best catalytic activity was analyzed in the experiment. And the dosage of Bi2 WO6 /RGO was set as the dosage in the range of 10 mg gradient.

4 Results and Discussion 4.1 Experimental Results and Analysis of RhB Degradation As shown in Fig. 1, we evaluated and compared Bi2 WO6 and Bi2 WO6 /RGO for the degradation effect of dye RhB. Significantly, Bi2 WO6 /RGO exhibited outstanding

Preparation and Photocatalytic Activity of Bi2 WO6 /RGO …

123

100%

Bi2WO6 Bi2WO6/RGO

C/C0

80%

60%

40%

20%

0% 0

black60

light60

light120

light180

Time(min)

Fig. 1 Degradation efficiency of RhB by Bi2 WO6 and Bi2 WO6 /RGO

catalyst performance for RhB in comparison to that of the pure Bi2 WO6 . The catalysts with removal through adsorption of RhB to varying degrees for dark reaction 60 min, and the removal rate by Bi2 WO6 /RGO increased 62% and was 2 times compared with single Bi2 WO6 . Catalyst Bi2 WO6 /RGO for the total removal efficiency of RhB increased to 79% and was 1.8 times higher than that of Bi2 WO6 in 60 min photoreaction. Materials Bi2 WO6 and Bi2 WO6 /RGO for the total removal rates of RhB separately exceeded 85% and 95% in photoreaction 180 min, it implied the photocalytic performance of material Bi2 WO6 /RGO was better than material Bi2 WO6 .

4.2 Experimental Results and Analysis of TC Degradation As shown in Fig. 2, we evaluated and compared single photocatalyst Bi2 WO6 and composite photocatalyst Bi2 WO6 /RGO for the photocatalytic degradation of TC. Composite Bi2 WO6 /RGO for the photocatalytic removal rate of TC increased to 61% and was 2 times compared with pure Bi2 WO6 in photoreaction 30 min. When the photoreaction was 180 min, Bi2 WO6 for the removal efficiency of TC exceeded 70%, while the concentration of TC was basically completely degraded by Bi2 WO6 /RGO. It stated that the photocatalytic activity of composite Bi2 WO6 /RGO was better than that of single material Bi2 WO6 . The photocatalytic removal rate of catalyst Bi2 WO6 /RGO became slow when time of light reaction exceeded 90 min, so it is necessary to optimize the dosage of catalyst Bi2 WO6 /RGO.

124

T. Lin et al.

1.0 Bi2WO6 Bi2WO6/RGO

C/C0

0.8 0.6 0.4 0.2 0.0

0

30

60

90

120

150

180

Time(min) Fig. 2 Catalyst Bi2 WO6 and Bi2 WO6 /RGO photodegradation efficiency of TC

4.3 Experimental Results and Analysis of Bi2 WO6 /RGO Dosage Optimization The experimental results of optimizing the dosage of Bi2 WO6 /RGO were shown in Fig. 3. The dosage of 30 mg compared with 40 mg, the latter photocatalytic process became faster as the dosage of Bi2 WO6 /RGO composite was increased when degrading TC. The result indicated that the dosage of Bi2 WO6 /RGO must be higher than that of 40 mg in order to have obvious photocatalytic effect. In addition, the photocatalytic efficiency of Bi2 WO6 /RGO was increased to the highest by the dosage with 50 mg and then decreased by being dosed with 60 mg. It should be noticed that a high amount of Bi2 WO6 /RGO dosage may serve as recombination centers and hinder the separation of photogenerated carriers. Therefore, the photocatalytic performance of Bi2 WO6 /RGO first gradually increased with the dosage and then decreased. When the dosage of 40 mg, 50 mg and 60 mg, degraded TC completely in photoreaction 180 min. It was noticed that the dosage of 50 mg for the removal effect was slightly higher than the other two. Consequently, it implied that 50 mg was the best dosage of Bi2 WO6 /RGO.

Preparation and Photocatalytic Activity of Bi2 WO6 /RGO …

125

1.0

20 mg 30 mg 40 mg 50 mg 60 mg

0.8

C/C0

0.6 0.4 0.2 0.0 0

30

60

90

120

150

180

Time(min) Fig. 3 The effect of Bi2 WO6 /RGO dosage on the photodegradation efficiency of TC

5 Conclusion (a)

(b)

Bi2 WO6 and Bi2 WO6 /RGO were prepared by hydrothermal method, which showed several performances for RhB and TC degradation. Photocatalytic removal and the total removal rate of RhB for Bi2 WO6 /RGO increased to 79% and 92% for photoreaction 120 min, respectively. The catalyst Bi2 WO6 /RGO for photocatalytic removal effect of TC increased to 90%, and the total removal effect exceeded 97%. Therefore, a higher adsorption and photocatalytic properties using Bi2 WO6 /RGO composite compared with single Bi2 WO6 was measured. We optimized the dosage of the composite Bi2 WO6 /RGO for degrading TC (20 mg/L). When the dosage was 50 mg, the degradation of TC had the best photocatalytic effect. In addition, the photodegradation effect of Bi2 WO6 /RGO to TC reached to 93% for 90 min photocatalytic reaction.

Acknowledgements This work was funded by Fujian provincial industry-university-research collaborative innovation (2021Y4005), Fujian Science and Technology Guiding Project (2020Y0056), and Fujian Engineering and Research Center of Rural Sewage Treatment and Water Safety.

126

T. Lin et al.

References 1. Volnistem EA, Bini RD, Silva DM, Rosso JM, Dias GS, Cótica LF, Santos IA (2020) Intensifying the photocatalytic degradation of methylene blue by the formation of BiFeO3 /Fe3 O4 nanointerfaces. Ceram Int 46(11):18768–18777 2. Guo F, Huang XL, Chen ZH, Sun HR, Chen LZ (2020) Prominent co-catalytic effect of CoP nanoparticles anchored on high-crystalline g-C3 N4 nanosheets for enhanced visible-light photocatalytic degradation of tetracycline in wastewater. Chem Eng J 395:125118 3. Ma Z, Hu L, Li X, Deng L, Fan G, He Y (2019) A novel nano-sized MoS2 decorated Bi2 O3 heterojunction with enhanced photocatalytic performance for methylene blue and tetracycline degradation. Ceram Int 45(13):15824–15833 4. Carlsson C, Johansson AK, Alvan G, Bergman K, Kühler T (2006) Are pharmaceuticals potent environmental pollutants? Part II: environmental risk assessments of selected active pharmaceutical ingredients. Sci Total Environ 364(1–3):67–87 5. Wu H, Lai C, Zeng G, Liang J, Chen J, Xu J, Dai J, Li X, Liu J, Chen M (2017) The interactions of composting and biochar and their implications for soil amendment and pollution remediation: a review. Crit Rev Biotechnol 37(6):754–764 6. Liang J, Yang Z, Tang L, Zeng G, Yu M, Li X, Wu H, Qian Y, Li X, Luo Y (2017) Changes in heavy metal mobility and availability from contaminated wetland soil remediated with combined biochar-compost. Chemosphere 181:281–288 7. Zhang Y, Zeng G, Tang L, Chen J, Zhu Y, He X, He Y (2015) Electrochemical sensor based on electrodeposited graphene-Au modified electrode and nanoAu carrier amplified signal strategy for attomolar mercury detection. Anal Chem 87(2):989–996 8. Khan M, Fung C, Kumar A, Lo I (2019) Magnetically separable BiOBr/Fe3 O4 @SiO2 for visible-light-driven photocatalytic degradation of ibuprofen: mechanistic investigation and prototype development. J Hazard Mater 365:733–743 9. Zhao J, Liu B, Meng L, He S, Yuan R, Hou Y, Ding Z, Lin H, Zhang Z, Wang X, Long J (2019) Plasmonic control of solar-driven CO2 conversion at the metal/ZnO interfaces. Appl Catal B Environ 256:117823 10. Yu C, Wu Z, Liu R, Dionysiou D, Yang K, Wang C, Liu H (2017) Novel fluorinated Bi2 MoO6 nanocrystals for efficient photocatalytic removal of water organic pollutants under different light source illumination. Appl Catal B Environ 209:1–11 11. Hou J, Cao S, Wu Y, Liang F, Sun Y, Lin Z, Sun L (2017) Simultaneously efficient light absorption and charge transport of phosphate and oxygen-vacancy confined in bismuth tungstate atomic layers triggering robust solar CO2 reduction. Nano Energy 32:359–366 12. Liang L, Lei F, Gao S, Sun Y, Jiao X, Wu J, Qamar S, Xie Y (2015) Single unit cell bismuth tungstate layers realizing robust solar CO2 reduction to methanol. Angew Chem Int Ed 54(47):13971–13974 13. Zhang N, Ciriminna C, Pagliaro M, Xu Y (2014) Nanochemistry-derived Bi2 WO6 nanostructures: towards production of sustainable chemicals and fuels induced by visible light. Chem Soc Rev 43:5276–5287 14. Zhou H, Wen Z, Liu J, Ke J, Duan X, Wang S (2019) Z-scheme plasmonic Ag decorated WO3 /Bi2 WO6 hybrids for enhanced photocatalytic abatement of chlorinated-VOCs under solar light irradiation. Appl Catal B Environ 242:76–84 15. Huang X, Yin Z, Wu S, He Q, Zhang Q, Yan Q, Freddy B, Zhang H (2011) Graphene-based materials: synthesis, characterization, properties, and applications. Small 7(14):1876–1902 16. Kong X, Chen C, Chen Q (2014) Doped graphene for metal-free catalysis. Chem Soc Rev 43(8):2841–2857 17. Han C, Zhang N, Xu Y (2016) Structural diversity of graphene materials and their multifarious roles in heterogeneous photocatalysis. Nano Today 11(3):351–372 18. Hummers WS, Offeman RE (1958) Preparation of graphitic oxide. J Am Chem Soc 80(6)

Study on Carbon Emission Accounting of Technological Processes in Synergistic Treatment and Disposal of Sludge and Food Waste Changhao Xiao and Jieying Chen

Abstract To solve the problem that sludge and food waste production increases year by year and consider the characteristics of both, the synergistic treatment of sludge and food waste is a hot research topic at present. Still, the evaluation of a synergistic treatment system from low carbon emission reduction is rarely discussed. Based on the methodology principle of IPCC, this paper constructs the calculation method of carbon emissions and carbon sinks for each process link in the collaborative treatment process of sludge and kitchen waste. Taking a sizeable domestic sewage treatment plant as an example, the carbon emissions under typical working conditions are calculated. The carbon emission characteristics are analyzed to provide a reference for the optimization of future processes. The results show that the dewatering and drying of biogas residue is the primary source of carbon emissions in collaborative treatment and disposal, followed by sewage treatment. The biogas produced in the synergistic anaerobic digestion process can provide energy, so the process has a significant carbon emission reduction effect. Considering the limitations of the current dehydration and drying methods of biogas residue, this study provides theoretical support and policy suggestions for future process innovation. Keywords Carbon emission accounting · Synergistic treatment · Sludge · Food waste

1 Introduction At present, China’s food waste production is large and growing fast. According to statistics, in 2020, the total amount of restaurant kitchen waste in China has reached 127.75 million tons, and it shows a trend of increasing year by year, with an average annual growth rate of 6.16% from 2015 to 2020. Food waste includes restaurant food waste and kitchen waste. The former refers to food leftovers from C. Xiao and J. Chen—These authors contributed equally. C. Xiao (B) · J. Chen School of Environment, Beijing Normal University, Beijing 100875, China e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_12

127

128

C. Xiao and J. Chen

restaurants and canteens and other processing wastes such as fruits and vegetables, meat, grease, and pastries. The latter refers to perishable organic waste such as discarded fruits and vegetables, food scraps, leftovers, and pericarp in household daily life [1]. Nowadays, food waste treatment and disposal methods at home and abroad mainly include pyrolysis, high temperature aerobic composting, high-temperature disinfection, anaerobic fermentation. The anaerobic fermentation technology is more mature and can use generated biogas as energy, so its market application rate is high [2]. Besides, with the rapid development of China’s social economy, the scale of urban sewage treatment in China is also increasing. Therefore, the treatment of sludge is also facing challenges. Sludge treatment refers to the implementation of thickening, conditioning, dewatering, stabilization, drying, and burning [3]. China’s sludge output exceeded 60 million tons (calculated with a moisture content of 80%) in 2019 and is expected to exceed 90 million tons in 2025. Since sludge has the dual characteristics of resources and pollutants, how to realize the stabilization, harmlessness, and resource utilization of sludge has become a key research issue in China’s environmental protection. Until now, anaerobic digestion technology has been widely used in sludge treatment and disposal because of its stable performance and high efficiency [4]. However, due to the high salinity and organic content of food waste, it is prone to ammonia nitrogen inhibition and acidification during the anaerobic digestion process [5], which will reduce the treatment effect and the efficiency of biogas generation. But the sludge has the characteristics of low salinity and organic content. In the method of treatment and disposal, additional carbon sources need to be added. Otherwise, gas production will decline. At the same time, since the carbon–nitrogen ratio (C/N) of the substrate is required to reach 10–20 in the general anaerobic digestion tank [6], while the carbon–nitrogen ratio of sludge is 6–8 and that of food waste is 15– 20. If the synergistic anaerobic digestion of the two can be realized, it can not only solve the process difficulties, improve the anaerobic digestion efficiency and biogas production, but also integrate the process and reduce the branching process of treatment and disposal. Therefore, this has become a hot topic in the current research at home and abroad [7]. Most of the research on the synergistic treatment of sludge and food waste is oriented to improve energy efficiency and yield, save cost, and focus on the characteristics and influencing factors of anaerobic digestion system and innovation of treatment and disposal process. Pan et al. [8] studied the synergistic effect and biodegradation kinetics of anaerobic digestion of sewage sludge and food waste. They confirmed the superiority of synergistic anaerobic digestion and the existence of the synergistic effect, which could significantly accelerate the biodegradation process and restore biological energy more effectively. Zhang et al. [9] studied the combined anaerobic digestion and co-pretreatment function of sludge and food waste. They developed a new physical co-pretreatment method to enhance anaerobic co-digestion while enhancing the hydrolysis and acidification of food waste and activated sludge. Baldi et al. [10] compared the effects of single-stage and two-stage anaerobic co-digestion of kitchen waste and activated sludge on hydrogen and methane production. The

Study on Carbon Emission Accounting of Technological Processes …

129

results showed that the two-stage system could comprehensively improve the anaerobic performance. The biogas production and volatile solid degradation increased by 26% and 9%, respectively, compared with the single-stage treatment. Vinardell et al. [11] studied the co-digestion process of sludge and food waste in a sewage treatment plant based on anaerobic membrane bioreactor technology. They found that the income of biogas and electricity was roughly balanced with the cost of input in co-digestion, indicating that the technology was economically feasible. Liu et al. [12] studied the synergistic digestive system of food waste and sludge from a demand-oriented biogas supply chain perspective. They proposed a way to combine the collection and transportation process with renewable energy to improve biogas’ utilization efficiency. However, although the anaerobic digestion process can produce biogas as energy for system utilization, it will also produce a large number of greenhouse gases, including CO2 , CH4, and N2 O, in the process of synergistic treatment and disposal of sludge and food waste, and consume electric energy, chemicals, and heat, which will cause carbon emissions. Moreover, with the rapid development of the sewage treatment industry and the increase of food waste production year by year, the total amount of greenhouse gas emissions in treatment and disposal will also show an increasing trend year by year. It is a challenge to China’s goal of peaking carbon dioxide emissions by 2030 and achieving carbon neutrality by 2060. Nevertheless, the current research ignores the evaluation of the synergistic anaerobic digestion system from this perspective. Therefore, it is necessary to construct a relatively perfect carbon emission accounting system and analysis method for the synergistic treatment and disposal of sludge and food waste. That is an excellent application prospect for developing this method and implementing energy-saving and emission reduction in the future. Based on the methodology principles of IPCC (Intergovernmental Panel on Climate Change), this study takes the synergistic treatment and disposal processes of sludge and food waste, including pretreatment, collaborative anaerobic digestion, sewage treatment, biogas residue dehydration and drying, and land use, as the objects to build a relatively complete carbon emission accounting method. Taking a largescale treatment plant as a specific case, the carbon emissions of each process link and operation stage under typical working conditions of the plant were calculated. The carbon emission characteristics of each link were analyzed and compared to provide a reference for low carbonization and parameter optimization of sludge and food waste synergistic anaerobic digestion.

130

C. Xiao and J. Chen

2 Accounting Boundary and Method 2.1 Accounting Boundary The overall accounting boundary includes the whole process from pretreatment to final land use, and determining the boundary is the basis for calculating the carbon emission of the process flow. According to the source and destination of carbon emissions, this study divides the accounting types into three categories: direct carbon emissions, indirect carbon emissions, and carbon sinks [13]. Direct carbon emission refers to the greenhouse gases directly generated in each process link during the synergistic treatment and disposal of sludge and food waste, mainly for the synergistic anaerobic digestion and sewage treatment processes. Indirect carbon emission refers to the carbon emission caused by power consumption, reagent addition, and heat input in each process. Carbon sink refers to the mechanism that the output of products or energy replaces the information of other raw materials or fuels to relatively reduce greenhouse gas emissions to the atmosphere. It is worth noting that anaerobic digestion technology uses facultative bacteria and anaerobic bacteria to degrade organic matter. The CO2 produced by microbial decomposition is a biological cause, not included in the scope of carbon emission accounting the IPCC guidelines [14]. Therefore, the carbon emission accounting boundary is shown in Fig. 1. Considering the diversity of greenhouse gas emissions and greenhouse effects, this study selects Global Warming Potential (GWP) as an indicator to measure the greenhouse effect. It reflects the radiation characteristics of greenhouse gases, representing the mass of carbon dioxide corresponding to various greenhouse gases with the exact greenhouse effect within the one century. The GWP value of carbon dioxide is 1. In the process of synergistic treatment and disposal of sludge and food waste, CH4 and N2 O are produced in addition to CO2 , and their GWP values are 25 and 298, respectively. Based on this, CH4 and N2 O are converted into CO2 emission equivalent to facilitate the accounting and comparison of carbon emissions in different processes.

Fig. 1 Carbon emission accounting boundary for the synergistic treatment and disposal of sludge and food waste

Study on Carbon Emission Accounting of Technological Processes …

131

2.2 Accounting Method Direct carbon emissions. The carbon emission caused by CH4 leakage in the process of synergistic treatment and disposal of sludge and food waste is: Ma,CH4 = Vg × η × α × GWPCH4

(1)

Ma,CH4 represents the CO2 emission equivalent of CH4 leakage, kg/d. Vg represents the biogas yield during synergistic anaerobic digestion, kg/d. η is 5%, meaning the uncontrollable leakage of CH4 in the biogas collection pipeline. α represents CH4 volume fraction in biogas, 65%. GWPCH4 represents the global warming potential of CH4 , 25. The carbon emission caused by CH4 generated in the sewage treatment process is: Mb,CH4 = V1 × (BOD0 − BOD1 ) × EFCH4 × GWPCH4

(2)

Mb,CH4 represents the CO2 emission equivalent of CH4 from sewage treatment, kg/d. Vl is the biogas slurry volume entering the sewage treatment process, t/d. BOD0 and BOD1 represent the BOD5 concentration of inflow and outflow in sewage treatment, respectively, mg/L. EFCH4 is the emission factor of CH4 . According to the default value recommended by IPCC Guidelines [14] and the actual situation of China’s sewage treatment plants, the value of CH4 emission factor in this study is 0.086 kgCH4 /kgBOD [15]. GWPCH4 is the global warming potential value of CH4 , which is 25. The carbon emission caused by N2 O generated in the sewage treatment process is: Mb,N2 O = V1 × (TN0 − TN1 ) × EFN2 O × GWPN2 O

(3)

Mb,N2 O represents the CO2 emission equivalent of N2 O from sewage treatment, kg/d. Vl is the biogas slurry volume entering the sewage treatment process, t/d. TN0 and TN1 represent the total nitrogen concentration of inflow and outflow in sewage treatment, respectively, mg/L. EFN2 O is the emission factor of N2 O. According to the default value recommended by IPCC Guidelines [14] and the actual situation of China’s sewage treatment plants, the value of the N2 O emission factor in this study is 0.035 kgN2 O/kgTN [15]. GWPN2 O is 298, representing the global warming potential value of N2 O. The carbon emission caused by N2 O generated by land use is: Mc,N2 O = E × βN × EFN2 O × GWPN2 O

(4)

Mc,N2 O is the CO2 emission equivalent of N2 O generated in the land use process, kg/d. E is the amount of biogas residue for land use, t/d. βN is nitrogen content

132

C. Xiao and J. Chen

in biogas residue. EFN2 O is N2 O emission factor in the land use process, 0.0109 kgN2 O/kg. GWPN2 O is the global warming potential value of N2 O, which is 298. Indirect carbon emissions. Carbon emission caused by power consumption is: Me,i = EFCO2 ,e × Ci

(5)

i represents different technological processes. Me,i is carbon emission, kgCO2 /d. EFCO2 ,e represents the CO2 emission factor of power consumption. In 2019, the emission factor of power consumption in East China was 0.7921 kgCO2 /kWh [16]. Ci represents the power consumption of each technological process, kWh/d. Carbon emission caused by reagent addition is: Mm,i = EFCO2 ,m,j × Gj

(6)

i represents different technological processes. j represents different types of reagents. Mm,i is carbon emission of reagent addition, kgCO2 /d. EFCO2 ,m,j represents the CO2 emission factor of a specific reagent. The primary reagent added in the treatment and disposal process is methanol and coagulants(PAC, PAM), with the CO2 emission factor of 1.54, 2.5 kgCO2 /d [17], respectively. Gj represents the amount of reagent added, kg/d. Carbon emission caused by heat input is calculated by converting carbon emissions into equivalent energy capacity: Mh,i = Hi × γ × δ−1

(7)

i represents different technological processes. Hi is the heat input required for each operation, MJ/d. Taking natural gas as the equivalent energy, γ means the CO2 emission factor of natural gas and the value is 21.622 t/104 m3 . δ represents the calorific value of natural gas, and the value is 389310 MJ/104 m3 . Carbon sink. Biogas produced by synergistic anaerobic digestion can be recycled. 1 m3 of biogas can generate power of 1.7 kWh and heat of 2 kWh, and its carbon sink is:   MS1 = Vg × (1 − η) × λ × γ × δ−1 + ε × EFCO2 ,e

(8)

MS1 represents the carbon sink generated by biogas recycling, kgCO2 /d. Vg is biogas yield during anaerobic digestion, m3 /d. λ is 2 kWh/m3 , representing the biogas heat production coefficient. γ represents the CO2 emission factor of natural gas, with a value of 21.622 t/104 m3 . δ represents the calorific value of natural gas, and the value is 389310 MJ/104 m3 . η is the uncontrollable leakage of CH4 in the biogas collection pipeline, taking 5%. ε represents the biogas power generation coefficient, which is

Study on Carbon Emission Accounting of Technological Processes …

133

1.7 kWh/m3 . EFCO2 ,e is the CO2 emission factor of power consumption. In 2019, the emission factor of power consumption in East China was 0.7921 kgCO2 /kWh. Besides, in the process of land use, about 61% of nitrogen and 70% of phosphorus in biogas residue can be absorbed and utilized by plants to replace chemical fertilizers. The unit energy consumption for the ammonium nitrate and calcium superphosphate production was 1 GJ/t and 1.3 GJ/t [18], respectively. The carbon sink of this process can be obtained by accounting: MS2 = E × EFCO2 ,e × (484βN + 954βP )

(9)

MS2 represents the carbon sink generated by replacing chemical fertilizer, kgCO2 /d. EFCO2,e is the CO2 emission factor of electricity consumption with the value of 0.7921 kgCO2 /kWh. βN is the nitrogen content of biogas residue, and βP is the phosphorus content of biogas residue. (484βN + 954βP ) is the energy consumption of fertilizer production reduced by biogas residue per unit mass in the process of land use, kWh/t.

3 Case Analysis 3.1 Case Profile Figure 2 shows the technological process of synergistic treatment and disposal process of sludge and food waste in a large domestic treatment plant. The biogas production during anaerobic digestion is 19.83 t/d (16121.13 m3 /d). The biogas

Fig. 2 The technological process and carbon emission accounting boundary of a sizeable domestic treatment plant for synergistic treatment and disposal of sludge and food waste

134

C. Xiao and J. Chen

Table 1 Specific data of power consumption, reagent addition, and heat input in synergistic treatment and disposal process Technological process Pretreatment of kitchen waste Sludge slurrying

Power consumption (kWh/d)

Reagent addition (kg/d)

Heat input (MJ/d)

3370.8

–

37,400

1532.0

–

48,200

2850(CH3 OH)

143,000

–

67,100

630(PAC) + 1.8(PAM)

–

Dewatering and drying of biogas residue

23303.3

Synergistic anaerobic digestion

3945.0

Sewage treatment

13865.5

slurry volume in the sewage treatment process was 404.16 t/d. The BOD5 concentrations of inflow and outflow are 129.88 mg/L and 3.7 mg/L, respectively. The TN concentrations of inflow and outflow are 51.89 mg/L and 11.69 mg/L, respectively. The amount of biogas residue during land use is 49.25 t/d, and the nitrogen and phosphorus contents of biogas residue are 4% and 2%, respectively. According to Eqs. (1)–(9) and the data of power consumption, heat consumption, and drug consumption of each technological process in Table 1, the carbon emissions of sludge and food waste’s synergistic treatment and disposal process in the plant are calculated.

3.2 Result and Discussion Direct carbon emissions. The direct carbon emission from the synergistic treatment and disposal of sludge and food waste in the treatment plant comes from three processes: synergistic anaerobic digestion, sewage treatment, and land use. The calculation results are shown in Table 2, and the total CO2 emission equivalent is 25485.6 kg/d. Among them, synergistic anaerobic digestion is the primary source of direct carbon emissions, accounting for 63.2%. The second one is the land use Table 2 Direct carbon emissions in synergistic treatment and disposal Technological process

CH4 emission (kg/d)

Synergistic anaerobic digestion

644.5

Sewage treatment Land use

N2 O emission (kg/d)

–

CO2 emission equivalent (kg/d) 16111.9

16.7

8.6

2974.7

–

21.5

6399.0

Study on Carbon Emission Accounting of Technological Processes …

135

Table 3 Indirect carbon emissions in synergistic treatment and disposal Technological process

Power consumption (kgCO2 /d)

Pretreatment of food waste

2670.0

Reagent addition (kgCO2 /d) –

Heat input (kgCO2 /d)

Total (kgCO2 /d)

2077.2

4747.2

Sludge slurrying

1213.5

–

2677.0

3890.5

Dewatering and drying of biogas residue

18458.5

517.1

7942.1

26917.8

3124.8

4389.0

3726.7

11240.5

10982.9

1579.5

Synergistic anaerobic digestion Sewage treatment

–

12562.4

process, accounting for 25.1%. The smallest share is the sewage treatment system, only 11.7%. By improving the process, such as changing the method of biogas collection and recycling, the leakage of biogas will be reduced, thereby decreasing the direct carbon emissions caused by synergistic anaerobic digestion. Indirect carbon emissions. Indirect carbon emissions come from the power consumption, reagent addition, and heat input of each technological process, and the accounting results are listed in Table 3. The total indirect carbon emission of the whole process is 59358.3 kgCO2 /d. The indirect carbon emissions of power consumption, reagent addition, and heat input are 36,449.7 kgCO2 /d, 6485.6 kgCO2 /d, and 16,423.0 kgCO2 /d from the perspective of energy consumption respectively. The indirect carbon emission of biogas residue dewatering and drying process is the largest, 26917.8 kgCO2 /d, followed by synergistic anaerobic digestion and sewage treatment system. The smaller is food waste pretreatment and slurrying sludge process. For biogas residue’s dewatering and drying process, the carbon emission of power consumption and heat consumption is higher, mainly affected by biogas residue’s water content and temperature. Therefore, reducing the water content and controlling the temperature will reduce the carbon emission of this process to a certain extent. Carbon sinks. Biogas recycling and biogas residue replacing chemical fertilizer are the main paths of carbon emission reduction in the whole process, and the carbon sink calculation results are shown in Table 4. The total carbon sink is 28246.7 kgCO2 /d. And the synergistic anaerobic digestion’s part is 26747.1 kgCO2 /d, accounting for Table 4 Carbon sinks in synergistic treatment and disposal Technological process Synergistic anaerobic digestion Land use

Carbon sink (kgCO2 /d) Biogas heat production

6124.2

Biogas power generation

20622.9 1499.6

26747.1

136

C. Xiao and J. Chen

Fig. 3 Total carbon emissions in synergistic treatment and disposal

94.7%, where biogas power generation is 20622.9 kgCO2 /d, accounting for 77.1% of the total synergistic anaerobic digestion. It can be seen that biogas power generation technology has a significant contribution to the carbon emission reduction of the synergistic treatment process and has a broad application prospect. At the same time, biogas power generation itself provides clean energy. This can solve environmental problems and reduce greenhouse gas emissions and promote the internal circulation of the synergistic treatment and disposal process since the generated biogas generates a large amount of electricity. The heat energy for the energy supply of the treatment plant also brings significant economic and ecological benefits. Total carbon emissions. Total carbon emissions include direct carbon emissions, indirect carbon emissions, and carbon sinks, and the accounting results are listed in Fig. 3. The total carbon emission of the whole process is 56597.2 kgCO2 /d. The carbon emission of biogas residue dewatering and drying is the largest, which is 26917.8 kgCO2 /d, followed by the sewage treatment system, accounting for 15537.1 kgCO2 /d. The carbon emission of the synergistic anaerobic digestion process is 605.3 kgCO2 /d, which can nearly achieve zero emission. So it’s the operation of low carbon environmental protection and energy saving.

4 Conclusion In this study, according to the process characteristics of the synergistic treatment and disposal process of sludge and food waste in large-scale treatment plants in China, the corresponding carbon emission accounting system is constructed, including direct

Study on Carbon Emission Accounting of Technological Processes …

137

carbon emissions, indirect carbon emissions, and carbon sinks. Through quantitative calculation and emission characteristics analysis, the basis of the optimization of process parameters and the selection of technical routes in the future has been built. The case study results show that the dewatering and drying of biogas residue and sewage treatment systems are the primary sources of carbon emissions in the whole process. Because biogas residue’s dewatering and drying process is the pre-step of land use of biogas residue, the temperature of dewatering and drying and final water content are the basis for determining whether the biogas residue can be used as a resource while the whole process requires energy and heat input. That results in large amounts of carbon emissions. Therefore, it is crucial to select a reasonable process route and optimize the operation parameters. At present, there are mainly three methods for profound dehydration of biogas residue: solar drying, plate-andframe pressure filtration after chemical conditioning, and high-voltage pulse electrolysis dehydration. However, there are some defects: solar drying covers a large area, relies on climatic conditions, and has an extended drying time, and the drying efficiency is low. The plate-and-frame pressure filtration, after chemical conditioning, needs to add flocculants such as lime, iron salt, and PM to change sludge characteristics, which will have a particular impact on the subsequent resource utilization. High voltage pulse electrolysis dehydration has high energy consumption and high dehydration cost. Therefore, how to promote the dehydration and drying process to achieve a balance between economic and ecological benefits, including reducing carbon emissions as much as possible, will be the key issues to be solved in future research.

References 1. Wang K et al (2014) Anaerobic digestion of food waste for volatile fatty acids (VFAs) production with different types of inoculum: effect of pH. Bioresour Technol 161:395–401 2. Yang L et al (2015) Enhancing biogas generation performance from food wastes by high-solids thermophilic anaerobic digestion: effect of pH adjustment. Int Biodeterioration Biodegradat 105:153–159 3. Ngo PL et al (2021) echanisms, status, and challenges of thermal hydrolysis and advanced thermal hydrolysis processes in sewage sludge treatment. Chemosphere 281:130890 4. Li P et al (2020) Experimental study on anaerobic co-digestion of the individual component of biomass with sewage sludge: methane production and microbial community. Biomass Convers Biorefinery 1–14 5. Gao S et al (2021) Synergetic enhancement of methane production and system resilience during anaerobic digestion of food waste in ammonia-tolerant anaerobic sludge system. Environ Sci Pollut Res 28(17):21851–21861 6. Kim M et al (2012) Hydrogen production by anaerobic co-digestion of rice straw and sewage sludge. Int J Hydrogen Energy 37(4):3142–3149 7. Mu L et al (2020) Anaerobic co-digestion of sewage sludge, food waste and yard waste: synergistic enhancement on process stability and biogas production. Sci Total Environ 704:135429 8. Pan Y et al (2019) Synergistic effect and biodegradation kinetics of sewage sludge and food waste mesophilic anaerobic co-digestion and the underlying stimulation mechanisms. Fuel 253:40–49

138

C. Xiao and J. Chen

9. Zhang J et al (2017) Enhancement of biogas production in anaerobic co-digestion of food waste and waste activated sludge by biological co-pretreatment. Energy 137:479–486 10. Baldi F, Pecorini I, Iannelli R (2019) Comparison of single-stage and two-stage anaerobic co-digestion of food waste and activated sludge for hydrogen and methane production. Renew Energy 143:1755–1765 11. Vinardell S et al (2021) Co-digestion of sewage sludge and food waste in a wastewater treatment plant based on mainstream anaerobic membrane bioreactor technology: a techno-economic evaluation. Bioresourc Technol 330:124978 12. Liu Y et al (2021) Optimizing the co-digestion supply chain of sewage sludge and food waste by the demand oriented biogas supplying mechanism. Waste Manag Res 39(2):302–313 13. Masuda S et al (2015) The seasonal variation of emission of greenhouse gases from a full-scale sewage treatment plant. Chemosphere 140:167–173 14. Gitarskiy ML (2019) The refinement to the 2006 IPCC guidelines for national greenhouse gas inventories 15. Sun SP et al (2010) Effective biological nitrogen removal treatment processes for domestic wastewaters with low C/N ratios: a review. Environ Eng Sci 27(2):111–126 16. Zhai M et al (2020) Inter-regional carbon flows embodied in electricity transmission: network simulation for energy-carbon nexus. Renew Sustain Energy Rev 118:109511 17. Liang Z et al (2021) Study on the quantitative evaluation of greenhouse gas (GHG) emissions in sewage-sludge treatment system. In: Eco design and sustainability II. Springer, Singapore, pp 271–287 18. Niu D-J et al (2013) Greenhouse gases emissions accounting for typical sewage sludge digestion with energy utilization and residue land application in China. Waste Manag 33(1):123–128

Preparation of Cotton Straw Based Multi-pore Biomass Charcoal, Characterization and Electrochemical Properties Jing Tao Dai, Ying Yang, Wen Xuan Zheng, and Li Na Wang

Abstract Cotton Straw Based Porous biochar was prepared by using cotton straw as raw material and anhydrous calcium chloride as activator. The morphology and structure of the product were characterized by TG-DSC, XRD, Raman, SEM, TEM and N2 adsorption desorption analysis, and the electrochemical properties were also analyzed. The results show that the specific surface area of the sample prepared at 650 °C is 487.68 m2 · g−1 , the average pore diameter is 5.97 nm, the total pore volume is 0.67 cm3 · g−1 , and the micropore volume is 0.15 cm3 · g−1 . The first discharge capacity of the lithium-ion battery with this sample as anode material is 1533.5 mA · h/g at 0.1C rate, 570.1 mA · h/g after 100 cycles, and 596.1 mA · h/g after 500 cycles at 1C rate, which indicates that the sample has good rate performance and cycle performance. Keywords Cotton straw · Activator · Biomass carbon · Lithium ion battery

1 Introduction Lithium ion battery research began in the 1970s. In 1991, Sony Company from Japan successfully developed lithium-ion battery which can be used in actual production and life. Since then, lithium-ion battery has successfully replaced zinc manganese dry battery, lead-acid battery, nickel cadmium battery with its excellent charge and discharge performance, high energy density, cycle life, safety and stability, widely used in mobile phones, computers, electric vehicles, aerospace and other fields, has a profound impact on our lives [1–4]. With the advancement of industrial technology and the implementation of national new energy strategy, the demand for lithium-ion battery and its materials in the future is immeasurable, and its supporting upstream J. T. Dai · Y. Yang (B) · W. X. Zheng · L. N. Wang College of Mechanical and Electrical Engineering, Tarim University, Alar, Xinjiang 843300, China e-mail: [email protected] The Key Laboratory of Modern Agricultural Engineering, Tarim University, Alar, Xinjiang 843300, China © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_13

139

140

J. T. Dai et al.

and downstream industrial chain is also full of huge market, and negative materials have a very important impact on the performance of lithium-ion batteries. Nowadays, the anode materials for lithium-ion batteries are mainly as follows: graphite, Li4 Ti5 O12 , etc. [5, 6] among them, graphite is the most common and widely used anode material, which has the advantages of high safety, good stability and conductivity, rich reserves and low price. However, the theoretical capacity of lithium-ion battery prepared with graphite as anode material is only 372 mA · h/g, which can not meet the daily needs of people. In the process of high current and long-time charge and discharge, lithium dendrite will appear on the surface of graphite, which will lead to the increase of resistance and short circuit of lithium-ion battery. Therefore, a kind of anode material with better performance is needed, which has the characteristics of graphite and can avoid the formation of lithium dendrite [7–12]. The preparation of lithium-ion batteries from biochar as anode materials is a hot research topic in recent years. It has been demonstrated and analyzed by many researchers. The lithium-ion battery with biochar as anode material not only has higher power density and energy density, longer cycle life and better safety performance, but also contributes to environmental protection and promotes the implementation of the national green energy plan [13–15]. Xinjiang is the most widely planted cotton Province in China. At present, the cotton planting area in China is about 3340 million square meters, while that in Xinjiang is about 2540 million square meters, It accounts for 76.05% of the total cotton planting area in China. At the same time, extensive cotton planting means the emergence of a large amount of cotton straw. Now, the annual output of cotton stalk in Xinjiang is about 13.3 million tons. The utilization rate of such a large amount of cotton straw is less than 10% of the total resources. A large number of cotton straw are directly burned or directly discarded, which leads to a large number of environmental pollution and land occupation. Therefore, the rational utilization of cotton straw can not only prevent pollution, but also generate income for farmers, which is of great significance for the sustainable development of agriculture in Xinjiang [16, 17]. Therefore, taking Xinjiang cotton straw as raw material, the morphology characterization and electrochemical performance test of cotton straw were carried out, which provided certain reference value for the rational use of cotton straw. Wang et al. [18] prepared the lithium ion battery with reed leaves as raw material and treated as negative electrode material, and the specific capacity was 277.8 mA · h/g under a large current density. Wang and Hu [19] prepared the lithiumion battery with rice starch as raw material and the sample obtained by pyrolysis at 1050 °C. The specific capacity of charge and discharge was 656.0 mA · h/g and 495.6ma · h/g under 0.1 C current density. They have high irreversible capacity and unstable cycle performance. In this paper, porous biochar was prepared from cotton straw at different temperatures with anhydrous calcium chloride as activator. It can not only reduce environmental pollution, but also has good flame retardancy and can improve the yield of activated carbon. The lithium-ion battery with this sample as anode material not only has the characteristics of graphite, but also can avoid the formation of lithium dendrite, and has better performance.

Preparation of Cotton Straw Based Multi-pore Biomass …

141

2 Experiment 2.1 Raw Materials and Reagents Cotton straw from farmland around Tarim University was selected as biomass raw material. Anhydrous calcium chloride; 2 mol/L dilute hydrochloric acid; Acetylene Black (Super P, battery grade, Shenzhen Kejing Zhida Technology Co., Ltd.); N-methyl-2-pyrrolidone (NMP, chemically pure, Tianjin Fuyu Fine Chemical Co., Ltd.); Polyvinylidene fluoride (PVDF, battery grade, Shenzhen Kejing Zhida Technology Co., Ltd); Electrolyte (1 mol·L−1 LiPF6 /(EC + DMC + MEC), Guangzhou Tianci High-tech Materials Co., Ltd.); Polypropylene Microporous Membrane (Celgand, CELGARD Diaphragm Technology Co., Ltd., USA); Lithium chip (battery grade, Shanghai Energy Lithium Industrial Co., Ltd.); Button cell Components (CR2032, Shenzhen Kejing Zhida Technology Co., Ltd.); The experimental water is deionized water.

2.2 Preparation of Biomass Carbon Take that stem part of the cotton straw and putting it into the ionize water, and removing the root part and the branches of the cotton straw, and taking the main part of the cotton straw. After cleaning, it is oven-dried at 90 °C in an electrothermal blast drying oven. After drying, the cotton straw is cut into segments of about 2 cm by using multifunctional cutting tongs, put into a high-speed multifunctional crusher, crushed, and passed through a 100-mesh screen. A proper amount of cotton straw powder was mixed with anhydrous calcium chloride at a mass ratio of 1: 2.5, and placed in a beaker. Pour in a certain amount of deionized water and use a glass rod to stir evenly to form a paste. The stirred mixture is laid flat in a glassware, dried in a drying oven at 90 °C, and crushed after drying. After passing through a 100mesh sieve, pouring into a crucible in three parts, and precarbonizing at 300 °C for 3 h in a muffle furnace. After cooling to room temperature, the temperature was raised to 550 °C, 650 °C and 750 °C respectively at a heating rate of 10 °C/min, and the temperature was cooled to room temperature after activation for 1 h. The activated product was put into a beaker, 300 ml of deionized water was first poured to dissolve calcium chloride, and then 100 ml, 2 mol/L of dilute hydrochloric acid was poured to soak at room temperature for 24 h. The mixed solution of deionized water and hydrochloric acid was pumped and filtered by circulating water vacuum pump, and was repeatedly washed with deionized water at 75–85 °C until the filtrate was neutral, and then dried in a drying oven at 90 °C. The obtained black powder is the biomass carbon prepared by the activator activation method, and is labeled as HMC-550, HMC-650 and HMC-750 respectively.

142

J. T. Dai et al.

2.3 Characterization of Biomass Carbon Morphology Test was carried out by STA449 F3 synchronous thermal analyzer (Cnscmp, argon environment, test temperature 30–800 °C, heating rate 10 °C/min). D/MAX-3B Xray diffractometer (Rigaku Company, Japan, Cu-Kα ray, λ = 0.15406 nm, scanning range 10–90°, scanning speed 5°/min) was used to characterize the phase of biomass carbon. The structure of biomass carbon was characterized by XploRA Plus micro Raman spectrometer (HORIBA Company, France, laser wavelength 532 nm, test range 400–2000). The appearance and morphology of biomass carbon were observed by SU8010 scanning electron microscope (Hitachi Company). Tecnai G 2F 20 transmission electron microscope (FEI Company, USA) can be used to observe the internal structural characteristics of biomass carbon. The structural parameters of biomass carbon can be obtained by using ASAP2460 specific surface area tester (Micromeritics). (The experiment was carried out with the help of scientific compass).

2.4 Battery Assembly Cotton stalk-based biomass carbon, acetylene black (super P) and polyvinylidene fluoride (PVDF) are mix in a ratio of 8:1:1 and ground evenly. The N-methyl-2pyrrolidone (NMP) was added under magnetic stir. After mixing evenly, the mixture was coated on smooth copper foil and dried in a vacuum drying oven at 80 °C for 16 h. Take out that product and tabletting in a tabletting machine (PX-CP-20, Shenzhen Pengxiangyunda Co., Ltd.) under a pressure of 2 MPa to prepare a wafer with a diameter of 10 mm, which is the electrode wafer. The battery was assembled in an argon-filled glove box (SUPER, Shanghai Mikrouna Electromechanical Technology Co., Ltd.) using lithium sheet as counter electrode, Celgard 2300 polymer membrane as separator and 1 mol·L−1 LiPF6 (volume ratio of vinyl carbonate (EC), methyl ethyl carbonate (EMC) and dimethyl carbonate (DMC) is 1: 1: 1) as electrolyte. After the assembly of the button cell (CR2032) was completed, the cell was kept at room temperature for 12 h, and then the electrochemical performance of the cell was tested. (The experiment was carried out with the help of scientific compass).

2.5 Test of Electrochemical Performance Test of constant current charging and discharging performance and rate performance of the battery was carried out by using Neware battery test system (CT-3008, Shenzhen Neware Electronics Co., Ltd.). Under the voltage range of 0.01–3.0 V, a low current density charging and discharging test was carried out at a rate of 0.2C (1C = 600 mA/g) for 100 cycles. At the magnification of 0.2C, 0.5C, 1C, 2C, 5C and 0.2C, the magnification performance test is carried out 10 times per current density

Preparation of Cotton Straw Based Multi-pore Biomass …

143

cycle. High current density cycle life test is carried out by cycling 500 turns at a rate of 1C. Cyclic voltammetry (CV) test was carried out using an electrochemical workstation (SP-150, Bio-Logic). The operating voltage range was 0.01–3.0 V, and the scanning speed was 0.1 mV · s−1 . The electrochemical workstation (CHI 760E, Shanghai Chinstruments Co., Ltd.) was used for AC impedance test, the working frequency range was 0.01–105 Hz, and the amplitude was 5 mV. (The experiment was carried out with the help of scientific compass).

3 Result and Discussion 3.1 TG-DSC Analysis Figure 1 shows the TG-DSC curve of cotton stalk. Seen from the figure, we can see that the sample lost weight obviously at 200–500 °C, the weight loss ratio was 73.91%; in such process, the main decomposition in samples were hemicellulose, cellulose and lignin. The weight loss rate gradually slowed down above 500 °C, because the most substances in the sample have been pyrolyzed. The aromatics in the remaining small amounts of lignin decompose more slowly by heat. Seen from the figure, we can see that there are two heat absorption peaks around 68 and 329 °C corresponding to the decomposition of calcium chloride and calcium chloride decomposition of acid components formed, which promoted the dehydration reaction of the main functional groups of cotton stalk. The sample weight loss ratio was 23.08% during the pre-carbonization process from room temperature to 280 °C. The chemical composition of the sample in this process began to change and the unstable components began to break down into CO2 , CO and a small amount of acetic acid. The solid decomposition and calcination process activated between 280 and 500 °C. Then, the sample was further decomposed into gases such as H2 O, CO, CO2 and liquids such as acetic acid, methanol and wood tar. The weight loss ratio of the sample was 50.83%. In the end, temperature range of 550–750 °C was

Fig. 1 TG-DSC curve of cotton stalk

144

J. T. Dai et al.

selected as the pyrolysis of cotton stalk to study the physicochemical properties and electrochemical properties of the samples [20].

3.2 X-ray Diffraction and Raman Spectrum Analysis Figure 2a shows the X-ray diffraction (XRD) of the samples HMC-550, HMC-650 and HMC-750.Seen from the figure, we can see that the three samples had wide diffraction peaks around 2θ = 23° and 41°, which corresponded to crystal planes (002) and (100), respectively. The diffraction peak of crystal plane (002) is stronger, while that of crystal plane (100) is weaker, consistent with the diffraction data of graphite PDF standard spectrum. It shows that the samples obtained by the two pyrolysis methods are amorphous carbon with better crystal type [21, 22]. Also, the sample only contains the characteristic peak of carbon material. There is no hybrid peak, indicating that the sample produced is of high purity. The values of d(002) of the three samples are calculated as follows: 0.38670, 0.38571, 0.38415 nm, values of 2θ are 22.98°, 23.04°, 23.08°. It indicates that with the increase of temperature, the value of 2θ increases gradually. The spacing between the crystal planes (002) gradually decreases. The sample structure is changing into a graphitized structure. Figure 2b is the Raman spectrum of the samples HMC-550, HMC-650 and HMC750. It can be seen from the Fig. 2 that all three samples show strong diffraction peaks around 1355 cm−1 and 1590 cm−1 , respectively, corresponding to “D peak” and “G peak” of carbon materials. “D peak” refers to the double resonance effect of disordered carbon structure. “G peak” was caused by the telescopic resonance of sp2 hybridized carbon atoms in the hexagonal lattice of carbon materials [23, 24]. The degree of disorder of carbon materials can be measured by ID /IG [25, 26]. The ID /IG values of the three samples are: 0.83, 0.97, and 0.74, respectively. The sample value produced at 650 °C was the largest, indicating that the sample produced at this temperature (650 °C) has the highest degree of disorder and the largest structural defect. It is conducive to the transmission of Li+ in the sample, improves the storage capacity of Li+ , and reduces the irreversible capacity. Also, larger layer spacing and higher degree of disorder can effectively inhibit the generation of lithium dendrites.

Fig. 2 XRD patterns (a) and Raman spectra (b) of the HMC-550, HMC-650 and HMC-750 samples

Preparation of Cotton Straw Based Multi-pore Biomass …

145

3.3 Specific Surface Area and Pore Diameter Analysis Figure 3a is the N2 adsorption–desorption isotherms of the samples HMC-550, HMC650 and HMC-750. It can be seen from the figure that the three samples are IV type isotherm with H4 hysteresis loop [27–29]. In the case of relatively low pressure (0–0.1) adsorption, and the adsorption capacity increased rapidly, indicating that the three samples all contained a large number of micropores. After the relative pressure increased (0.1–0.8), the adsorption curve rose uniformly and a hysteresic ring was generated [30, 31]. It indicates that there are many large holes in all three samples. Figure 3b is the pore diameter distribution curves of HMC-550, HMC-650 and HMC-750. It can be seen from the figure that the pore diameter of the three samples was mainly concentrated at 2–20 nm, and the pore volume was large, which was consistent with the results of N2 adsorption–desorption isotherm analysis. Table 1 shows the structure characteristics of three samples obtained from N2 adsorption–desorption isotherm and pore diameter distribution curves. It can be concluded from Table 1. that the specific surface area, average pore diameter, total pore volume and micropore volume of the samples prepared at 650 °C are the maximum values, which are about 487.68 m2 · g−1 , 5.97 nm, 0.666 cm3 · g−1 and 0.147 cm3 · g−1 , respectively. Larger specific surface area and more evenly distributed pore diameters can provide more contact points for the electrolyte and reduce the resistance of charge transfer process, and make the embedding and extruding of Li+ more efficient. Thus, the cycling efficiency of lithium-ion batteries can be improved.

Fig. 3 Nitrogen adsorption–desorption isotherms (a) and pore size distributions (b) of theHMC550, HMC-650 and HMC-750 samples

Table 1 Structure characteristic data of cotton straw biochar prepared by different pyrolysis methods Sample

Specific surface area/m2 · g−1

Average pore size/nm

Vtotal/cm3 · g−1

Micropore volume/cm3 · g−1

HMC-550

447.97

5.94

0.663802

0.136

HMC-650

487.68

5.97

0.665685

0.147

HMC-750

387.77

5.45

0.578407

0.118

146

J. T. Dai et al.

The larger total pore volume and micropore volume can store more Li+ and improve the reversible capacity as well, which is consistent with Raman’s analysis.

3.4 Micro-morphology Characterization The scanning electron microscopy (SEM) diagrams of the produced samples HMC550, HMC-650 and HMC-750 are shown as Fig. 4. It can be known from the SEM diagrams that there are porous structures in all the three samples with relatively rough surfaces, which is because the Cacl2 damages the layered structure of cotton stalks. Cacl2 is embedded into the carbon structure under the action of high temperature, which leads to the rearrangement of crystal structure and the formation of porous structures and further expands the specific surface areas of samples [32]. Owing to the lower temperature of sample HMC-550, the internal structure of cotton stalks cannot be fully activated, so the pore size is relatively small. Owing to the excessive high temperature of sample HMC-750, some internal porous structures inside the cotton stalks collapses, which reduces the specific surface area. While the temperature of sample HMC-650 is proper, so there is a larger specific surface area and more porous structures, which can promote the penetration of electrolyte inside the sample. Meanwhile, the stable porous structures of samples that belong to hard carbon materials facilitates improving the cycle stability of lithium-ion batteries. This is consistent with the results of specific surface area and pore size distribution. The transmission electron microscopy (TEM) diagrams of the produced samples HMC-550, HMC-650 and HMC-750 are shown as Fig. 5. It can be known from the TEM diagrams that there are closed rings and unsealed spots on all the three samples. The closed rings are caused by the hollow structure inside the samples under the vacuum pyrolysis status while the unsealed spots are of open and through channel structure. There are dark spots and disorderly layered structure in the sample HMC-550, which is caused by incomplete activation. The sample HMC-650 has more interlaced branch structure, which is caused by the intersection of layered structure with multiple pores [33]. There are relatively large irregular ring areas in the sample HMC-750, which is caused by pore channel collapse. It indicates that the different

Fig. 4 SEM images of the HMC-550 (a), HMC-650 (b) and HMC-750 (c) samples

Preparation of Cotton Straw Based Multi-pore Biomass …

147

Fig. 5 TEM images of the HMC-550 (a), HMC-650 (b) and HMC-750 (c) sample

pyrolysis temperature will lead to different porous structures, which is consistent with SEM result analysis.

3.5 Formation Mechanism Analysis The formation mechanism diagram of cotton-stalk-based porous biomass charcoal is shown as Fig. 6. After mixing the cotton stalk powder and CaCl2 , during the mixture drying process, CaCl2 will further penetrate into the interlayered structure of cotton stalks. The dehydration of CaCl2 and the acid components formed by CaCl2 decomposition will make the dehydration reaction of main functional groups of cotton stalks easier. In the further pyrolysis process, CaCl2 that has penetrated into cotton stalks will dissolve the substances in the stalks, such as cellulose, hemicellulose and

Fig. 6 Formation mechanism of mesoporous biomass carbon derived from cotton stalk

148

J. T. Dai et al.

lignin, under the catalysis of thermal decomposition. With the rising of temperature, CaCl2 will embed into the carbon structure under the action of high temperature and the cotton stalks will make the crystal structure melting and softening deformation under the high-temperature pyrolysis, which makes the crystal structure recombination. CaCl2 plays the role of frame in the recombined carbon structure, based on which the multi-phase substance constituted by CaCl2 as the frame and amorphous carbon is formed. After the pyrolysis, deionized water and 2 mol/L diluted hydrochloric acid are used to wash CaCl2 away, after which the cotton-stalk-based porous biomass charcoal materials with multiple pore channels and porous structures are obtained [34].

3.6 Electrochemical Performance Analysis Diagrams of the first three cycle curves of the charge–discharge tests of the produced samples HMC-550, HMC-650 and HMC-750 at 0.1C magnification within the potential range of 0.01– 3.0 V are shown as Fig. 7. It can be seen from the diagrams that, in the first discharge curve, there is a relatively flat discharge platform at about 0.7 V in all the three samples, which then disappears in the cycles, at which the biomass charcoal and electrolyte react into solid electrolyte interphase (SEI) in the first charge–discharge process [35].

Fig. 7 Diagrams of charge–discharge curves of preparing; biomass charcoal cathode materials at different temperature

Preparation of Cotton Straw Based Multi-pore Biomass …

149

Table 2 The first charge–discharge capacities and the first coulomb efficiency of preparing biomass charcoal cathode materials at different temperatures Sample

The first charge capacity/mA · h · g−1

The first discharge capacity/mA · h · g−1

The first coulomb efficiency (%)

HMC-550

644.8

1197.9

53.8

HMC-650

741.4

1533.5

48.3

HMC-750

558.7

990.4

56.4

The first charge–discharge capacities and the first coulomb efficiency of the samples HMC-550, HMC-650 and HMC-750 are shown as Table 2. The sample HMC-650 has the largest charge–discharge specific capacity, which is 741.4 mA · h/g and 1533.5 mA · h/g, respectively. The first coulomb efficiency is 48.3% only. Because the structural deficiency of biomass charcoal with 650 °C pyrolysis is relatively large with higher amorphous degrees, which consumes more Li+ to form SEI and further leads to lower first coulomb efficiency. Figure 8a shows the cycle performance curve diagram of the prepared samples HMC-550, HMC-650 and HMC-750 cycling 100 times at a rate of 0.1C, it can be seen from the figure that with the increase of the number of cycles, the specific capacity of the three samples first attenuates and then gradually increases and tends to be stable, and also the Coulomb efficiency is stable accordingly. After the 100 cycles, the specific discharge capacities of the three samples are 535.9 mA · h/g,

Fig. 8 Cycling performance profiles at 0.1C (a), rate; performance (b), cycling performance profiles at 1C (c) of the HMC-550, HMC-650 and HMC-750 samples

150

J. T. Dai et al.

570.1 mA · h/g and 483.9 mA · h/g, respectively, the Coulomb efficiency is basically stable at 98%, indicating that all the three samples have the better cycle performance and the specific charge and discharge capacity. Figure 8b is a curve diagram of the rate performance produced. Under the current density of the 0.2C, 0.5C, 1C, 2C and 5C respectively, the rate performance test is conducted for each sample, when the current density is gradually increasing, the specific capacity of each sample decreases accordingly, however, when the current density becomes small again, the specific volume of the sample will recover accordingly, indicating that every sample has a good rate performance, of which, the rate performance for the sample of pyrolysis under 650 °C is the best. Figure 8c shows the cycle curve diagram for every sample charging and discharging 500 times under the 1 C rate of large current, it is seen from the diagram that after cycling 500 times, the sample HMC-550, HMC-650, and HMC-750 discharge specific capacities are 578.8 mA · h/g, 596.1 mA · h/g, 550.3 mA · h/g, respectively, all the samples show the situation that the specific volume of the sample increases accordingly with the increase of cycling times, this kind of situation is mainly caused by the following reasons: (1) In the case of the large current charge and discharge, more pore passages and interlayer structures inside the sample are impacted by the large current, leading to the interconnection of them, and thus the lithium ion and electrons are made to conduct more rapidly. (2) The increase in the number of cycles leads to more sufficient penetration between the electrolyte and the sample, more activation sites inside the sample have been stimulated, and the specific volume of the sample is increased. In general, the performance of HMC-650 sample is excellent under the 0.1C rate, different rate and 1C high rate, which is mainly due to the rich pore structure and larger specific surface area in the sample, the electrolyte can be permeated rapidly in the sample to ensure the rapid conduction of the lithium ions and electrons under the condition of the high charge and discharge rate. Figure 9a–c correspond to the cyclic voltamogram curve diagram of the HMC550, HMC-650, and HMC-750 samples at a scanning speed of 0.1 mV/s and a voltage window of 0.01–3.0 V, respectively. During the process of first turn discharge, each sample showed a sharp peak near 0.01 V, which is produced by the phase transition caused by the Li+ embedded in the sample. In the reduction stage, all the samples have the two reduction peaks between 0.50–0.95 V and 1.10–1.50 V, the reduction peaks between 0.50 and 0.95 V correspond to the formation of solid electrolyte membrane, the reduction peak between 1.10 and 1.50 V corresponds to the irreversible reaction generated of lithium ion in the electrode surface, all the reduction peaks of the different samples are disappeared in the next two turns of cycle, and the cycle curve tends to overlap, indicating that the reversible capacity shows a trend of stability, and this is in conformity with the test results of cycle performance. In the oxidation stage, all the samples show the two oxidation peaks in the 0.10–0.50 V and 1.0– 1.4 V range, of which, the oxidation peak between 0.10 and 0.50 V corresponds to the embedded in and out reaction of lithium ions in the samples, the oxidation peak shown between 1.0 and 1.4 V corresponds to the peak formed by the embedded in and out of lithium ion in many pore channel structure inside the sample, because both the formation of oxidation peak belongs to the reversible reaction, the two oxidation peaks have

Preparation of Cotton Straw Based Multi-pore Biomass …

151

Fig. 9 Cyclic voltammogram profile (a–c) of the HMC-550, HMC-650, HMC-750 and the third cycle cyclic voltammetry curve of each sample (d)

152

J. T. Dai et al.

not disappeared during the next process of cycle. Figure 9d shows the voltamogram curve diagram for the third turn of cycle in each sample. By comparing the integral area of the third cycle voltamogram curve in each sample, it can be seen that the sample HMC-650 has the largest integral area, indicating that its specific capacity is higher, which is consistent with the results of charge–discharge test. In order to further test the electrochemical performance of cotton straw based porous biochar, AC impedance tests were carried out for each sample before and after 100 cycles. The working frequency range was 0.01–105 Hz, and the amplitude was 5 mV. As shown in Fig. 10a all samples are composed of semicircles in high frequency region and diagonal lines in low frequency region. The semicircles in high frequency region of each sample decrease significantly after cycling, which indicates that the impedance of electrode material decreases. By fitting the Nyquist curve before and after the cycle, the appropriate equivalent circuit model diagram is obtained, as shown in Fig. 10b The specific values obtained by simulation are shown in Table 3. From the table, it can be seen that the sample HMC-650 before and after the cycle has a small charge transfer resistance. At the same time, after 100 cycles, the interface resistance is only 5.9 ῼ, and the charge transfer resistance is reduced to 20.6 ῼ, which can be attributed to After several cycles, on the one hand, lithium ion reacts with the electrode surface to form a stable SEI film. On the other hand, the intercalation and desorption of lithium ions expand the channels and gaps in the

Fig. 10 AC impedance spectroscopy of samples HMC-550, HMC-650 and HMC-750 (a) and simulated circuit diagram (b)

Preparation of Cotton Straw Based Multi-pore Biomass …

153

Table 3 Matching impedance values of samples HMC-550, HMC-650 and HMC-750 before and after cycling Sample

Before Rs /ῼ

After Rct /ῼ

Rs /ῼ

RSEI /ῼ

Rct /ῼ

HMC-550

1.9

159.0

4.1

18.9

25.6

HMC-650

2.1

130.5

3.8

5.9

20.6

HMC-750

2.4

187.5

3.7

15.2

43.1

sample, increase the space for charge and lithium ion movement, and reduce the charge transfer resistance [36].

4 Conclusion Cotton straw based porous biochar was prepared from agricultural waste cotton straw by activation with anhydrous calcium chloride by steps of pyrolysis and carbonization, washing and drying. The microstructure and morphology of the samples prepared at different carbonization temperatures were characterized by XRD, Raman, bet, SEM and TEM, and the formation mechanism of the porous structure was explored. The results show that the samples activated by anhydrous calcium chloride have many mesoporous structures and large specific surface area. Different temperatures affect the amorphous and defect degree of biochar. The results show that the specific discharge capacity is 570.1 mA · h/g and the coulomb efficiency is 98% when the samples are cycled 100 times at 0.1C rate at 650 °C, and the specific discharge capacity is 596.1 mA · h/g when the sample is cycled 500 times at 1C rate.

References 1. Cheng XB, Zhang R, Zhao CZ, Zhang Q (2017) Toward safe lithium metal anode in rechargeable batteries: a review. Chem Rev 117:10403–10473 2. Li M, Lu J, Chen Z, Amine K (2018) 30 years of lithium-ion batteries. Adv Mater 30:1800561– 1000566 3. Winter M, Barnett B; Xu K (2018) Before Li ion batteries.Chem Rev 118:11433–11456 4. Whittingham MS (1979) Electrical energy storage and intercalation chemistry. J Solid State Chem 192(3):303–310 5. Zhang WM, Wu XL, Hu JS et al (2010) Carbon coated Fe3 O4 nanospindles as asuperior anode material for lithium-ion batteries. Adv Funct Mater 18(24):3941–3946 6. Tarascon JM, Armand M (2001) Issues and challenges facing rechargeable lithium batteries. Nature 414(6861):359–367 7. Thackeray MM, David WIF, Bruce PG et al (1983) Lithium insertion into manganese spinels. Mater Res Bull 18(4):461–472

154

J. T. Dai et al.

8. Scrosati B, Garche J (2010) Lithium batteries: status, prospects and future. J Power Sourc 195(9):2419–2430 9. Huo K, An W, Fu J et al (2016) Mesoporous nitrogen-doped carbon hollow spheres as highperformance anodes for lithium-ion batteries. J Power Sourc 324:233–238 10. Jiang Q, Zhang Z, Yin S et al (2016) Biomass carbon micro/nano-structures derived from ramie fibers and corncobs as anode materials for lithium-ion and sodium-ion batteries. Appl Surf Sci 379:73–82 11. Kundu D, Talaie E, Duffort V et al (2015) The emerging chemistry of sodium ion batteries for electrochemical energy storage. Angew Chem Int Ed Engl 46(21):3431–3448 12. Mu L, Xu S, Li Y et al (2016) Prototype sodium-ion batteries using an air-stable and Co/Ni-free O3 -layered metal oxide cathode. Adv Mater 27(43):6928–6933 13. Liang YP, Li CC, Chen WJ et al (2013) Hydrothermal synthesis of lithium iron phosphate using pyrrole as an efficient reducing agent. Electrochim Acta 87(1):763–769 14. Deng J, Li M, Wang Y (2016) Biomass-derived carbon: synthesis and applications in energy storage and conversion. Green Chem 18(18):4824–4854 15. Kalyani P, Anitha A (2013) Biomass carbon & its prospects in electrochemical energy systems. Int J Hydrogen Energy 38(10):4034–4045 16. National Bureau of Statistics (2020) Announcement on cotton yield in 2019. Econ Res China’s Cotton Hemp Indus 01(01):27–28 17. Xu D, Wang M (2016) Status quo of feed utilization of cotton stalk resources in Xinjiang and ways of industrial production. Xinjiang Animal Husbandry 10:17–18 18. Wang J, Zhang J, Yu Y (2019) Synthesis of Si/C composites derived from directly-carbonized reed plants as high-performance anode for lithium ion batteries. J Forest Eng 4(5):84–91 19. Wang L, Hu PF (2018) Preparation and electrochemical performance of rice starch hard carbon cathode material. New Chem Mater 46(1):229–232+236 20. Zuo PJ, Yin GP, Ma YL (2007) Electrochemical stability of silicon/carbon composite anode for lithium ion batteries. Electrochimica Acta 52(15):4878–4883 21. Zhang LL, Zhao X, Stoller MD et al (2012) Highly conductive and porous activated reduced graphen oxide films for high-power supercapacitors. Nano Lett 12(4):1806–1812 22. Ng SH, Wang JZ, Wexle RD et al (2007) Amorphous carbon-coated silicon nanocomposites: a low-temperature synthesis via spray pyrolysis and their application as high-capacity anodes for lithium-ion batteries. J Phys Chem C 111(29):11131–11138 23. Bonnet R (2000) Evaluation of surface strain due to the reconstruction of atomically closepacked crystalline surfaces. Phys Rev B 61(20):14095–14107 24. Liu N, Wu H, Mcdowell MT et al (2012) A yolk-shell design for stabilized andscalable Li-ion battery alloy anodes. Nano Lett 12(6):3315–3321 25. Han SW, Jung DW, Jeong JH et al (2014) Effect of pyrolysis temperature on carbon obtained from green tea biomass for superior lithium ion battery anodes. Chem Eng J 254(7):597–604 26. Beyaz Kayiran S, Darkrim Lamari F, Levesque D (2013) Adsorption properties and structural characterization of activated carbons and nanocarbons. J Phys Chem B 108(39):15211–15215 27. Lee K, Lytle J, Ergang N et al (2005) Synthesis and rate performance of monolithic macroporous carbon electrodes for lithium-ion secondary batteries. Adv Funct Mater 15(4):547–556 28. Liu J, Wen Y, Wang Y et al (2014) Carbon-encapsulated pyrite as stable and earth-abundant high energy cathode material for rechargeable lithium batteries. Adv Mater 26(34):6025–6030 29. Lu J, Zhou C, Liu Z et al (2016) LiMn2 O4 cathode materials with large porous structure and radia l interior channels for lithium ion batteries. Electrochim Acta 212:553–560 30. Regufe MJ, Ferreira AFP, Loureiro JM et al (2018) New hybrid composite honeycomb monolith with 13X zeolite and activated carbon for CO2 capture. Adsorption 24(3):249–265 31. Yue L, Xia Q, Wang L et al (2018) CO2 adsorption at nitrogen-doped carbons prepared by K2 CO3 activation of urea-modified coconut shell. J Colloid Interf Sci 511:259–267 32. Liu J, Kopold P, Van Aken PA et al (2015) Energy storage materials from nature through nanotechnology: a sustainable route from reed plants to a silicon anode for lithium-ion batteries. Angew Chem Int Ed 54(33):9632–9636

Preparation of Cotton Straw Based Multi-pore Biomass …

155

33. Xue GZ, Hou GH, Qiao RJ et al (2016) Preparation and activation mechanism of rice husk based mesoporous carbon. Chin J Environ Eng 10(1):375–378 34. Yi L, Chun L, Kaifeng Y (2018) Preparation, characterization and electrochemical properties of mesoporous biomass carbon derived from corn stalk. Chem Res Chin Univ 39(4):607–613 35. Yang Y, Yan C, Huang J (2020) Research progress of solid electrolyte interphase in lithium batteries. Acta Phys Chim Sin 2020(1):1–13 36. Li J-T, Wu J-H, Zhang T, Huang L (2017) Preparation of biochar from different biomasses and their application in the Li-S battery. Acta Phys Chim Sin 33(5):968–975

Dynamic Gas Control Strategy for Mode Switching in a Proton Exchange Membrane Unitized Regenerative Fuel Cell Mengdi Guo, Zhonghao Zhang, Zhonghao Yu, Siyue Yao, Diankai Qiu, and Linfa Peng Abstract Proton exchange membrane unitized regenerative fuel cell (PEM-URFC) is a particular kind of fuel cell which combines fuel cell (FC) and water electrolyzer (WE) together. The mode switching from FC operating mode to WE operating mode is an essential process in PEM-URFC. In this study, we explored the change of high frequency resistance (HFR) at different purging speeds when the cell was converted from WE to FC mode through experiments and divided the purging into three stages. We designed an efficient and stable dynamic purging strategy, which increased the purging efficiency by 62.8% and reduced the gas consumption by 40.4%. Keywords URFC · Purge · HFR · Mode switch · Dynamic control strategy

1 Introduction Hydrogen, with its high calorific value and cleanness, is considered to be the next generation of energy source replacing traditional fossil fuels [1–3]. Proton exchange membrane unitized regenerative fuel cell (PEM-URFC), combining a fuel cell (FC) and a water electrolyzer (WE) together, is a promising energy device which is capable of both storing and generate energy using hydrogen as the medium. Due to its longterm energy storing and generating capability, high power density, light weight and relatively low operating temperature, PEM-URFC is deemed to be potential for various applications such as unmanned aerial vehicle (UAV) and power grid peak shaving [4, 5]. Mode switching between FC mode and WE mode is an important process in PEMURFCs. Obviously, there is no technological difficulty in switching from FC mode to WE mode by turning off the gas supply at the end of the FC mode and supplying the M. Guo · Z. Zhang · Z. Yu · S. Yao · D. Qiu (B) · L. Peng State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University, Shanghai 200240, People’s Republic of China e-mail: [email protected] Key Laboratory of Digital Manufacture for Thin-Walled Structures, Shanghai Jiao Tong University, Shanghai 200240, People’s Republic of China © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 M. L. Kolhe (ed.), Renewable Energy Systems in Smart Grid, Lecture Notes in Electrical Engineering 938, https://doi.org/10.1007/978-981-19-4360-7_14

157

158

M. Guo et al.

cell with deionized water to power the solution. However, during the transition from electrolysis mode to electric generation mode, the cell is filled with water, which may cause the flooding problem in FC mode, preventing gas from entering the fuel cell. So the remaining water needs to be purged. It should be noted that the scavenging of residual water requires not only to damage the soft components of the cell, such as the membrane and the catalyst layer, but also to improve the scavenging efficiency as much as possible, reduce the time consumed by the scavenging, so as to improve the efficiency of the URFC. URFC needs to discharge most of the water in the battery during the transformation from WE mode to FC mode. If the purge time is too long, the membrane electrode will be too dry and the diffusion layer and catalytic layer will be damaged. However, if the purge time is too short, the water in the battery will be too much, the gas mass transfer efficiency will be reduced, and the power generation performance will be reduced. If the purge pressure is too large, the pressure difference between the two sides of the membrane electrode will be too large and the membrane electrode will be damaged. Meanwhile, the switching time, purge gas and energy consumption in mode switching will have a great impact on the performance of URFC. In the previous literature, Niu et al. [6] studied the three stages of water removal after purging of fuel cells based on HFR characteristics, analyzed and discussed the different distribution of water in the membrane caused by different purging methods. Ito et al. [7] developed a gas purge model for URFC mode switching from an electrolysis cell to FC mode to study the pure time under different gas flow rates. In their research, relationships between HFR and water content in fuel cells are explored. They found that, within a certain range, the growth rate of HFR increases as the purge volume increased. However, the long-term prediction effect is poor, and no corresponding purge strategy has been put forward. Xiao et al. [8] analyzed the dynamic response of mass fraction operating voltage and cell temperature of a unitized regenerative fuel cell under various ways of mode switching. Liu et al. [9] conducted an experimental investigation on two-phase flow in a unitized regenerative fuel cell during mode switching from WE to FC. In summary, it has been a direct and reliable method to use HFR to characterize the distribution of water content and guide the gas purging for fuel cells. Unfortunately, the application of HFR test during the switching mode of URFC has not been tried. In this study, according to the correlation between water content and HFR, we explored the change of HFR at different purging speeds when the stack was converted from WE to FC mode through experiments, and divided the purging into three stages, and an efficient and stable dynamic purging strategy is designed, which increases the purging efficiency by 62.8% and reduces the gas consumption by 40.4%.

Dynamic Gas Control Strategy for Mode Switching …

159

2 Experimental Setup To verify the change law of HFR under different purge gas quantities, groups of experiments were carried out. Schematic diagram of the test station system is shown in Fig. 1.

2.1 URFC Design A kind of URFC was selected for purging verification. The cell consisted of end plates, current collecting plates, stamped bipolar plates and membrane electrode assemblies. The cell had three joints with the reactive area of 180 cm2 . There were three pairs of cavities in the cell, namely the inlet and outlet of oxygen cavity, the inlet and outlet of hydrogen cavity and the inlet and outlet of cooling water, with the cathode as the purging side.

2.2 Experimental Device The flow controller of Alicat Scientific Inc. was selected, whose accuracy is 1SCCM and can be used to accurately control the gas inlet flow. A constant temperature water bath was applied to control the water temperature. The water bath was equipped with water pump, and its pipeline was covered with insulation layer. The water bath can introduce heated water into the cell cooling water

Fig. 1 Schematic diagram of test station system

160

M. Guo et al.

cavity to keep the cell temperature at 60 °C. A thermocouple was inserted into the cell to measure the temperature of it. The accuracy of the HFR meter was 0.001 mῼ and the frequency was 1000 Hz. The measuring position was set on the current collecting board, so the total HFR value of the whole cell can be measured.

2.3 Reactants Supply Nitrogen was selected as the purge gas, whose flow rate was controlled by the flow controller. Only the cathode side was filled with purge gas. Deionized water at 60 °C was used to wet the cathode side, simulating the state of the cell at the end of the WE mode. Deionized water was supplied by another same constant temperature water bath.

2.4 HFR Test The experimental set up of the test system is shown in Fig. 2. Insulated deionized water was introduced into the cell cooling cavity to maintain the cell temperature at 60 °C. Before purging, 60 °C deionized water was inlet into the cathode side, and was discharged after 2 min to simulate the wet state of the membrane when electrolysis was completed. Nitrogen gas was then inlet into the cathode side to continuously purge until the HFR value of the cell reached 4 mῼ (HFR of a single battery is 240 mῼ · cm2 ). Then the gas supply was stopped and

Fig. 2 Experimental device setup scene

Dynamic Gas Control Strategy for Mode Switching …

161

the change of HFR during and after the gas supply was stopped was recorded. The recording frequency was every 20 s.

2.5 Setting of Different Gas Volume According to general purging strategy, the volume flow of the reaction gas at the current density of 0.5 A/cm2 was used as the standard purging volume. Following this standard, 10 test points in the range of 0.2–1.6 A/cm2 were set up based on the current density (Table 1). Based on Faraday’s law, the volumetric flow rate of cathode (O2 ) can be calculated. q˙o2 =

m˙ o2 I · M˙ o2 = ρo2 4F · ρo2

where, q˙o2 : Volume flow of oxygen (cm3 /s) m˙ o2 : Mass flow of oxygen (g/s) ρo2 : Density of oxygen (g/cm3 ) M˙ o2 : Relative molecular mass of oxygen I : Current (A) F: Faraday constant (C/mol) The flows were set as follows. Table 1 Setting of different gas volume Current density (A/cm2 )

Gas volume flow (L/min)

0.2

0.68

0.3

1.02

0.4

1.35

0.5

1.69

0.6

2.03

0.7

2.37

0.8

2.71

1

3.38

1.2

4.06

1.4

4.74

1.6

5.41

(1)

162

M. Guo et al.

Fig. 3 The change of HFR in the whole process

3 Results and Discussion 3.1 Variation Law of HFR Under Standard Gas Purge At 60 °C, the HFR of the URFC in the fully wetted state is about 90 mῼ · cm2 . After shutdown, in the dry state after completing gas purge, the HFR is about 200 mῼ · cm2 . And then the first purging is conducted: purging by using the standard gas volume is carried out when the cell is in a fully wetted state- on the cathode side, the standard gas volume is defined as the volume corresponding to the current density of 0.5 A/cm2 , that is, when the flow rate is 1.69 L/min. During the purging process, the HFR will increase. After the HFR rises to 240 mῼ · cm2 , the purging will be stopped. The HFR change of the whole process will be recorded as shown in Fig. 3. It can be seen that the whole purge process lasted for 9 min, and the rate of HFR rise grew faster and faster over the time period. After the purge was stopped, the HFR slowly decreased, falling to 180 mῼ · cm2 after 10 min and was basically stable. The HFR value at this time is slightly less than the HFR value in the shutdown state, so it can be considered that the water content of the membrane electrode assembly is much smaller than that of the water content in the electrolysis end state.

3.2 Changes of HFR Under Different Gas Volume Purges Ten groups of purge volumes were divided according to the current density, and the HFR change curve corresponding to each volume is shown in Fig. 4. Figure 5 shows that the rate of HFR increases faster and faster at each gas volume. The larger the gas volume, the faster the HFR growth rate, the greater the growth rate changes.

Dynamic Gas Control Strategy for Mode Switching …

163

Fig. 4 The change of HFR under different gas volume

Fig. 5 Relationship between growth rate of HFR and gas flow

The average growth rate of HFR is defined as the ratio of the increasing amount of the HFR from the initial to the termination condition of purge (when HFR reaches about 240 mῼ · cm2 ) to the correspondingly consuming time. For example, a volumetric flow of 0.68 L/min takes about 28.7 min to raise the impedance value from 90 to 240 mῼ · cm2 , with an average HFR change rate of 5.23 mῼ · cm2 /min. Comparing the average HFR change rate and the gas volume flow (Fig. 5), it shows that the gas volume flow and the average HFR change rate are not completely linear. When the gas volume flow is small (