114 77 59MB
English Pages 354 [355] Year 2021

Lecture Notes in Electrical Engineering 763
Miguel Botto Tobar Henry Cruz Angela Díaz Cadena Editors
Recent Advances in Electrical Engineering, Electronics and Energy Proceedings of the CIT 2020 Volume 2
Lecture Notes in Electrical Engineering Volume 763
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 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 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 leontina. [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 http://www.springer.com/series/7818
Miguel Botto Tobar Henry Cruz Angela Díaz Cadena •
•
Editors
Recent Advances in Electrical Engineering, Electronics and Energy Proceedings of the CIT 2020 Volume 2
123
Editors Miguel Botto Tobar Eindhoven University of Technology Eindhoven, The Netherlands
Henry Cruz Universidad de las Fuerzas Armadas ESPE Sangolquí, Ecuador
Angela Díaz Cadena Universitat de Valencia Valencia, Spain
ISSN 1876-1100 ISSN 1876-1119 (electronic) Lecture Notes in Electrical Engineering ISBN 978-3-030-72211-1 ISBN 978-3-030-72212-8 (eBook) https://doi.org/10.1007/978-3-030-72212-8 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 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 Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
Performance Analysis of a Solar Water Heater for Space Heating in Residential and Commercial Buildings . . . . . . . . . . . . . . . . . . . . . . . . David Alejandro Arguello Bravo, Javier Martínez-Gómez, Esteban Francisco Urresta Suárez, David Rodger Salazar Loor, and Gonzalo Guerrón Permeability Characterization of a Composite Reinforced Material with Fiberglass and Cabuya by VARTM Process. Case Hybrid Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diana Belén Peralta-Zurita, Diego Jimenez-Pereira, Jaime Vinicio Molina-Osejos, and Gustavo Adolfo Moreno-Jiménez Effect of Conventional and Ecological Dielectric on the Wire Electrical Discharge Machining WEDM Process on AISI-D3 Steel . . . . Cristian Pérez-Salinas, Diego Molina-Molina, and Leónidas Ramirez-Gangotena Corrosion Analysis in Different Cookware Materials . . . . . . . . . . . . . . . Javier Martìnez-Gomez, Marco A. Orozco-Salcedo, Augusto Riofrio, Gonzalo Guerrón, and Ricardo A. Narváez C.
1
16
31
43
Corrosion Rate Comparison Between a ZnCrO4 Coating and a Mixture of Epoxy Plus PU Coating on HSLA ASTM a 1011 Gr50 Steel Exposed to a Saline Spray Corrosion Chamber . . . . . . . . . . Cristian Guilcaso, Augusto Coque, Xavier Vaca, Leonidas Ramírez, Diego Molina, and Isaac Simbaña
53
Weibull Reliability Analysis in Hydraulic Jet Pumps, Case Study Block 56 – Ecuador . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diego Ayala, Lenin Pozo, and Wilson Padilla
67
v
vi
Contents
Structural Analysis Method for Aeronautical Modifications in the Integration of Electro-Optical Systems in Helicopters for the Implementation of Intelligence, Surveillance and Reconnaissance (ISR) Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . Roberto Narváez Aguilar, Danny Flor Mancheno, Diego Paredes Sánchez, and Flor Garcés Mancero Impact Analysis of Migration from Súper Gasoline to Others of Lower Octane Number in Ecuador . . . . . . . . . . . . . . . . . . . . . . . . . . Carlos Francisco Terneus Páez, Absalón Guillermo Cabrera Mera, and Rubén Darío Grandes Villamarín
79
95
Experimentation of Adaptive Strategies in High-Speed Machining (HSM) for Rough Milling Process Using Prodax Aluminum . . . . . . . . . 109 Francisco Infante Castillo and Borys Culqui Culqui Optimization of the Setup of Workpiece Zero Point in a Numerical Control Machine with an Artificial Vision System . . . . . . . . . . . . . . . . . 123 Andrea Robalino Pinango and Borys Culqui Culqui Tribological Characterization of Erosive Wear Resistance as a Criteria of Material Selection for Fabrication of Construction Equipment and Machinery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Juan Angel Barella, Juan Manuel Victorio Vallaro, Mercedes Lozano Rus, Eldo José Lucioni, and Huber Gabriel Fernández Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Johanna Tobar, Álan Prócel, Andrea López, Bladimir Bacca, and Eduardo Caicedo Trends in Technological Advances in Food Dehydration, Identifying the Potential Extrapolated to Cocoa Drying: A Bibliometric Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 A. D. Rincón-Quintero, L. A. Del Portillo-Valdés, A. Meneses-Jácome, C. L. Sandoval-Rodríguez, W. L. Rondón-Romero, and J. G. Ascanio-Villabona Analysis of the Energy Potential of a Tangential Microturbine for Application in a Passivhaus Environment . . . . . . . . . . . . . . . . . . . . . 181 J. G. Ascanio-Villabona, L. A. Del Portillo-Valdés, O. Lengerke-Pérez, B. E. Tarazona Romero, A. D. Rincón-Quintero, and M. A. Durán-Sarmiento
Contents
vii
Development of a Fresnel Artisanal System for the Production of Hot Water or Steam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 B. E. Tarazona-Romero, A. Campos-Celador, Y. A. Muñoz-Maldonado, J. G. Ascanio-Villabona, M. A. Duran-Sarmiento, and A. D. Rincón-Quintero Portable Robotic Modular Kit for Teaching Gestures in Children with ASD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Johanna Tobar, Joffre Delgado, Brandon Muñoz, Bladimir Bacca, and Eduardo Caicedo Descriptive Study of a Rotary Machine Affected by Misalignment and Imbalance Applying the Wavelet Transform . . . . . . . . . . . . . . . . . . 226 Camilo Leonardo Sandoval-Rodriguez, Brayan Eduardo Tarazona-Romero, Omar Lengerke-Perez, Carlos Gerardo Cárdenas-Arias, Diana Carolina Dulcey Diaz, and Oscar Arnulfo Acosta Cárdenas Flexible Manufacturing Systems Optimization with Meta-heuristic Algorithm Using Open Source Software . . . . . . . . . . . . . . . . . . . . . . . . . 243 Fabian Izquierdo, Edwin Garcia, Byron Cortez, and Luis Escobar Estimation of the Energy Consumption of an Electric Utility Vehicle: A Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Gianina Garrido-Silva, Jessica Gissella Maradey-Lazaro, Arly Dario Rincón-Quintero, Omar Lengerke-Pérez, Camilo Leonardo Sandoval-Rodriguez, and Carlos Gerardo Cardenas-Arias Artistic Creations Supplied by Renewable Energy Located in the Most Attractive Mountains of Azuay. Case Study: Cultural Heritage of Quingeo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Daniel Icaza, Santiago Pulla Galindo, Carlos Flores-Vázquez, and Fabián Sangurima Paute Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 F. Endara, C. Pérez, J. Rodriguez, D. Ortiz-Villalba, and J. Llanos Blackberry (Rubus Glaucus) Natural-Fiber Reinforced Polymeric Composites: An Overview of Mechanical Characteristics . . . . . . . . . . . . 300 Enrique Mauricio Barreno-Avila, Morayma De Los Ángeles Balladares-Pazmiño, Alex Francisco Barreno-Avila, and Segundo Manuel Espín-Lagos Design and Construction of a Passive Control System for Seismic Isolation of Flexible Element Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 Lenin Abatta-Jácome, Carlos Vega-Rivas, Roberto Villagran-Mayorga, and Edison E. Haro
viii
Contents
Microstructure Damages and Changes on Mechanical Properties of the Heat-Affected Zone on Welded Joints of High-Strength Low-Alloy Steel Due to Multiple Repairs . . . . . . . . . . . . . . . . . . . . . . . . 330 Carlos Naranjo-Guatemala and John Cruz-Aldaz Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
Performance Analysis of a Solar Water Heater for Space Heating in Residential and Commercial Buildings David Alejandro Arguello Bravo1 , Javier Martínez-Gómez1,2(B) , Esteban Francisco Urresta Suárez2 , David Rodger Salazar Loor3 , and Gonzalo Guerrón2,4 1 Universidad Internacional SEK, Quito 170134, Ecuador
[email protected] 2 Instituto de Investigación Geológico y Energético (IIGE), Quito 170518, Ecuador 3 Facultad de Ciencias de la Ingeniería, Universidad Técnica Estatal de Quevedo, Quevedo
120301, Ecuador [email protected] 4 Universidad UTE, Rumipamba y, Bourgeois, Quito 170147, Ecuador
Abstract. This research aims to analyze the performance of a solar water heater for space heating in residential and commercial buildings in the city of Quito in Ecuador. The collector system is environmentally friendly since during its operation no greenhouse gases are generated, being these the main causes of global warming produced by the burning of fossil fuels. One of the advantages of this innovation is that it is easy to install and transport. It can be installed outside apartment buildings and homes as if they were windows in the building itself, and its location allows the collector to be kept clean, improving the heating performance of the heat transfer fluid and offering a versatile sustainable building concept that reduces energy consumption. Carrying out tests under different meteorological conditions, the efficiency calculation of the solar air collector was developed, obtaining an average value of 59.09% with an acceptable collector performance, achieving an average useful heat of 3807.65 W and an output temperature of around 34.6 to 94 °C. These important data were obtained through methodologies of: thermal analysis in the collector, energy balances in each section and definition of mathematical models. This efficiency achieved offers good perspectives in the application of this type of technologies that contribute to the use of alternative or renewable energies, since the solar resource can be used to cover the energy demand at low costs. Keywords: Solar air collector · Efficiency · Thermal · Sustainable · Saving · Energy
1 Introduction Solar thermal energy has relevance in industry, whether used domestically or scientifically it presents wide feasibility and technical viability. It is considered an important © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 1–15, 2021. https://doi.org/10.1007/978-3-030-72212-8_1
2
D. A. Arguello Bravo et al.
renewable energy source for the reduction of global climate change, and technologies that promote its use are currently being widely accepted in research and development [1]. Solar energy as an inexhaustible energy resource that promotes the development of energy policies, such as: stopping the massive devastation of ecosystems, minimizing activities that pollute the air and water, deforestation and excessive CO2 generation [2]. One of the solar technologies currently used are solar collectors. This equipment allow the collection, absorption and transfer of solar energy, taking advantage of the thermal power generated and using it as an energy source in different applications as shown in Fig. 1. The essential objective of the air collector is to absorb as much direct and diffuse radiation as possible in order to convert it into thermal energy [3].
Fig. 1. Diagram of the convective-loop solar collector and its operation.
Solar collectors have various uses, among the most common are water heating, space heating and industrial processes. One of their most interesting applications is the heating of buildings, proving to be an alternative with a lot of potential, since it reduces the high energy consumption, reduces costs, and provides thermal comfort through air conditioning in homes and commercial buildings. Between 45%–55% of the energy consumed in a building corresponds to space heating and cooling, and it is essential to improve the energy efficiency of thermal conditioning equipment and building envelopes [4]. It is a challenge to replace traditional energy sources that produce greenhouse gases with alternative energy sources, thus avoiding high air pollution levels. In solar collectors, different factors must be considered such as: the optimal angle of inclination of the equipment, the cloudiness, albedo, and the different objects or buildings that can cause shadows on the collectors, since they are variables that directly affect the system performance. In order to analyze the influence of the inclination angle, it is necessary to use transport equations that differ fundamentally in the expressions used to decompose solar radiation into direct and diffuse, since in general there are measured or estimated global irradiation data on the horizontal plane [5].
Performance Analysis of a Solar Water Heater
3
Thermal insulation in solar heating systems becomes indispensable for good system performance. This variable must also be considered in buildings to reduce heat losses [6]. When the thermal comfort of a dwelling or building is not adequate, it can cause health problems for users. Therefore, measures must be taken in aspects of air temperature regulation, average radiant temperature, air speed and air humidity. The cost analysis in thermal efficiency influences the implementation of solar collection systems, which involves the evaluation of various criteria in the configurations and setup of collectors. Studies have been carried out to achieve efficiency levels of more than 75%, using a double countercurrent flow with an absorber plate in the middle of the cover and bottom of the collector [7], and driving turbulence through baffles attached to the absorber plate to increase heat transfer [8, 9]. In the aspect of collector construction, the cost can be reduced according to studies [10, 11] that show that the transparent cover can be avoided. In the comparison between a single-pass air-heating collector with a double-pass collector [12], the thermal efficiency of a double-pass solar air-heating collector is 10% higher; while in a double-pass collector system with an absorber matrix it is 25% higher than that of a collector without a matrix [13]. It should be considered the type of material used to obtain the best thermal performance from each element of the collector, such as having a transparent cover which must have a high transmissibility to short wavelength radiation [14]. There are different designs and models, which makes it easier to choose according to the required needs. The most used solar collector is the flat type, because of its average efficiency that goes from 40% to 60% and because of the temperatures that it can deliver at the output of between 50 °C–200 °C. Figure 2 shows the air diagram in the flat collector.
Fig. 2. Air flow diagram in the flat collector.
The modification of a liquid solar collector for its dual function as an air and water heater was studied by Zhang et al. [15]. The device has three modes of operation, modes A and B for air and water heating, respectively, and mode C for simultaneous heating of air and water. The experimental results show that the average efficiency of modes A and B can reach a value of around 51%, while the average efficiency in mode C can be significantly higher (73.4%). Othman et al. [16] developed and tested a hybrid solar collector for electricity generation and the simultaneous heating of air and water, reaching a maximum thermal efficiency of 76%. Nematollahi et al. [17] investigated the performance of a hybrid solar system for dual heating of air and water. The device
4
D. A. Arguello Bravo et al.
consisted of a flat plate solar collector connected to a water storage tank. The results indicate that this prototype has an efficiency between 3 and 5% higher than a single purpose system. This article investigated the technical feasibility of using solar water heating collectors in air heating applications for space conditioning. This study is relevant in Ecuador, where most of the solar collectors are employed for domestic water heating, with the possibility of using or recycling this equipment in air heating systems. Unlike previous investigations, the present study does not seek dual or single operation of a solar collector by modifying an existing design or developing a new one, but rather the direct use of liquid solar collectors for air heating.
2 Materials, Methods and Calculations This study proposes to develop solar thermal collectors with the optimization of resources and obtaining the greatest possible benefit without generating solid waste, thus increasing industrial, manufacturing and environmental eco-efficiency. The solar thermal collector used is of the type of flat surface in the shape of a rectangular prism that is normally part of a solar water heater, similar as shown in Fig. 3. The aim of this investigation is to use this equipment to optimize energy resources, obtaining valuable results for future applications in air heating and ventilation.
Fig. 3. Flat-surface front face absorbing solar energy and internal insulation material of a solar thermal collector.
2.1 Parts of the Solar Thermal Collector The main parts of a solar collector are the absorber plate, the thermal insulation and the transparent cover as shown in Fig. 4. The correct position of these elements provides the shape and structure of the solar collector, having several alternatives of air circulation such as: over and under flow, intermediate plate with double glass, and without glass. The correct distribution in the collectors influences the reduction of energy losses and provides a better efficiency to the heat transfer process.
Performance Analysis of a Solar Water Heater
5
Fig. 4. Components of a flat surface solar collector
Transparent Cover – Glass. The transparent cover of the solar thermal collector is made of glass with a rectangular shape and a flat surface, as this allows an increase in transmittance, i.e. the passage of an amount of energy through a transparent surface in a given time. Absorber Plate – Metal. The absorber plate of the solar thermal collector is a black rectangular shaped metal plate and on its surface it has a design of vertical and horizontal tubes that are part of the metal plate, distributed in such a way that the fluid moves inside it when entering and leaving it, the sum of all these geometric configurations make a single piece with a single area of heat transfer. This plate acts as a black panel by having the property of absorptance in its space; this means how much a material can absorb incident radiation in its area or plane and basically depends on the finish and color of the materials. Thermal Insulation – Glass Wool. The solar thermal collector used for this research is hermetic and factory sealed in all its parts which makes it impossible to disassemble the set because if this were carried out, it would increase the energy losses in the same apart from those it already has and even worse to visualize the thermal insulator with its thickness used for this application, according to references the most suitable in thermal insulators for this type of collectors is the glass wool for its good thermal properties. 2.2 Analysis of Solar Thermal Energy in the Collector Figure 5 shows a diagram of the analyzed flat-plate solar collector, whose principle of operation is based on the fact that the absorber plate has high solar absorptivity and low emissivity, being capable of taking the greatest amount of this energy from the absorber and that the cover is capable of prevents radiative and convective losses. When analyzing the energy source, the mechanisms of heat transfer such as: conduction, convection and radiation, the resistances that oppose the process, the configuration ˙ u [18]. of the collector and the final absorber heat Q ˙ u = Acol .F[S − U (Ti − Ta )] Q
(1)
Figure 6 shows the schematic diagram of the variables considered within the analysis of the thermal capture system.
6
D. A. Arguello Bravo et al.
Fig. 5. Diagram of a solar collector.
Fig. 6. Energy behavior of the solar thermal collector.
Energy Balance in the Collector Cover. The cover absorbs energy by two fundamental mechanisms: the heat radiated from the absorber plate and the heat gained by convection from the hot fluid circulating inside the collector [19]. BEC
dTc = h4 Acv Tp − Tc + h3 Acv (Tai − Tc ) − h1 Acv (Tc − Taex ) − h4 Acv (Tc − Taex ). dt (2)
Energy Balance on the Collector’s Absorber Plate. The energy balance in the absorber plate has 3 components, the heat delivered to the indoor air, the heat that is lost to the environment through the insulation of the collector bottom, and the heat delivered by radiation directly to the collector cover [19]. dTp = τvαp It Ap h5 Ap Tp − Tai + Ub Ap Tp − Taes + h4 Acv Tp − Tc (3) dt Air Balance Inside the Collector. The air is heated with the energy absorbed by the absorber plate and in turn the absorber plate loses energy to the collector cover [19]. BEP
BAiC
dTai = mC ˙ p Ti + h5 Ap Tp − Tai − h7 Acv (Tai − Tc ) − mC ˙ p To dt
(4)
Performance Analysis of a Solar Water Heater
7
Heat Removal Factor. The heat removal factor is defined as the ratio of the heat removed by the fluid in the tubes to the useful energy if the entire collector were at the fluid inlet temperature [20]. Fr =
m ˙ ∗ Cp ∗ (To − Ti ) Ac [S − UL (Ti − Ta )]
(5)
Important Equivalences. The parameters of the collector cover are optimized by the selectivity assessment according to the analysis performed [21]. α = 0.92 − 0.95
(6)
τc = 0.15 − 0.25
(7)
Fr = 0.8 − 0.9
(8)
Overall Thermal Loss Coefficient of the Collector. The overall heat transfer coefficient can vary between 10 W/m2 K and 7 W/m2 K, depending on whether the collector is handmade or built in an industrialized way. Collector Efficiency Factor. This factor is used when the air mass flow is not known in the interior of the collector based on its geometry and specific design, this value has less accuracy and reliability since it is an estimated amount and varies from reality under strict conditions [22]. In this study it is not necessary to determine it since the air flow must be considered. m ˙ ∗ Cp Fr ∗ UL ∗ Ac (9) ln 1 − F UL = Ac m ˙ ∗ Cp Mean Absorber Plate Temperature. The mean absorber plate temperature will always be higher than the average fluid temperature. This temperature difference is generally small for liquid systems and significantly higher for air systems. Collector Efficiency. The performance of the collector is determined by its efficiency, which is defined as the ratio of the useful gain over some specified time period to the incident solar energy over the same time period. Tpm − Ta η = Fr τcv αp − Fr ∗ UL ∗ (11) It Energy Losses from the Collector. The air mass flow that circulates through the collector is fundamental to obtain the useful heat in each test, from the flow rate of the equipment used (2 HP blower), which supplies air constantly inside the collector and evaluates two variables: the speed and the area through which the fluid enters, with heat. The calculated area values are reflected in Table 1.
8
D. A. Arguello Bravo et al.
Table 1. Calculation of the solar thermal collector surface, depending on the surface of each collector component. Geometrical shape
Area (m2 )
Cover-glass
Rectangle
1,61634
Absorption plate
Vertical tubes Horizontal tubes Rectangle
2,424792 0,12516 1,61634
Collector
Rectangular Prism cover Absorption plate
3,57107 1,61634 4,166292
Components
The velocity value is verified in the equipment’s technical sheet and it is checked that it has a value of V = 0.5 m3 /s in order to apply the mass flow rate formula, considering the constant air density, obtaining a value of ν = 0.06125 kg/s. m ˙ = ρ.V
(12)
The fluid passing through the interior of a circular pipe has a Reynolds number given by the following equation. In which the mass flow rate of 0.06125 kg/s, the pipe diameter of 0.024 m, the area of the section through which the fluid passes is 0.0004524 m2 , and the value of dynamic viscosity being 0.0000174 Pa · s. The amount achieved in the Reynolds number meets the following condition Re ≥ 4000 with a value of 186748. R =
m ˙ ·D As · u
(13)
Since it is a turbulent fluid that passes through the interior of the circular pipe, it presents a fundamental characteristic that favors this application. This characteristic is called mixing processes, which means that there is a notable increase in the exchange of heat, mass and amount of movement in the fluid mass due to the effects of transverse velocity factors due to the turbulence presented. Figure 7 shows the behavior of the ambient temperature with respect to solar irradiation during the periods of time, when exceeding one hour established within the test, it is noticeable that there is a very variable behavior. Due to the fact that the month of December was mostly cloudy and rainy, the maximum and minimum ambient temperature were 17 and 14 °C, respectively, as well as solar irradiance was 810 W/m2 K maximum and 298 W/m2 K minimum.
3 Results and Discussion The experiment was developed at the Miguel de Cervantes campus of SEK University, located in the city of Quito. Due to time availability, the solar collector tests were
Performance Analysis of a Solar Water Heater 900 17 800
16 15
810 712
18
16
16
14
700 664
810 762
15
16
16 15
14
761
16 14
712
14
761 664
600
614
12 10 8
517
500
9
6 400 It (W/m2) 300
Ta (°C)
298
200 > 2 pm > 8 am > 8 am > 2 pm > 2 pm > 2 pm > 8 am > 2 pm > 2 pm > 8 am > 2 pm > 2 pm
4 2 0
Fig. 7. Ambient temperature vs. average values of solar radiation on the solar collector, during the 12 tests.
carried out only for 12 days. Input, output and ambient air temperatures Ti, To and Ta, respectively, of the solar thermal collector were recorded, as well as the total irradiation It presented, being measured daily at different times of the day, between 8:00 and 12:00 on weekends, and between 14:00 and 18:00 during weekdays (one measurement every 15 min, with a total of 204 measurements), variables that allow the calculation of useful heat and collector efficiency. Data such as air density were considered ρ = 1.225 kg/m3 , air flow volume V = 0.05 m3 /s, air specific heat of 1.0052 kJ/kgK, total area of solar thermal collector Ac = 9.353702 m2 , coefficient of thermal loss in the collector Ul = 7 W/m2 K, average temperature of the absorber plate is Tpm = 30,88 °C, mass flow is m ˙ = 0.06125 kg/s and a dimensionless parameter of average collector performance is UL (Tpm − Ta )/It = 0.16, all these values are used to obtain the efficiency of the solar thermal collector. As shown in Table 2, an average useful heat value of 3807.05 W, a maximum of 4740.44 W and a minimum of 1022.04 W was determined, as well as a daily efficiency with a maximum of 63% and a minimum of 37%, and a total solar thermal collector efficiency of 59.09%. This value is higher than the average efficiency of 51% achieved by the dual device developed by Zhang et al. (2016) [15] in the air heating mode. The trend of inlet air temperatures is observed in Fig. 8 are even presenting variations in 1 °C at most, due to the forced convection that exists by the instrument used “2 hp blower” to provide air to the interior, when this activity happens the air hits the blades of the dynamic machine of the instrument creating friction in the same between the fluid and the internal walls of the equipment so its temperature increase. While the output temperature varies considerably in a ratio of 5 to 1 unlike the other temperatures when passing through the solar thermal collector. The curves shown in Fig. 9 detail the variation over time of the output temperature with respect to the solar irradiance presenting a directly proportional and homogeneous distribution in each location according to the range of the established hours, there is a
10
D. A. Arguello Bravo et al. Table 2. Average results obtained in each test Time
Ta (°C)
Ti (°C)
To (°C)
It (W/m2 )
Qu (W)
η (%)
1
>2 pm
17
18
34,6
298
1022,04
0,37
2
>8 am
15
16
82,6
712
4100,46
0,62
3
>8 am
14
15
77
664
3817,25
0,62
4
>2 pm
16
17
94
810
4740,77
0,63
5
> 2 pm
16
17
88,4
762
4395,99
0,62
6
>2 pm
16
17
94
810
4740,77
0,63
7
>8 am
15
16
82,6
712
4100,46
0,62
8
>2 pm
16
17
60
517
2647,45
0,55
9
>2 pm
16
17
88,33
761
4391,68
0,62
10
>8 am
15
16
71,3
614
3404,74
0,59
11
>2 pm
14
15
77
664
3817,25
0,62
12
>2 pm
14
15
88,3
761
4512,97
0,63
Average
>8 am, 2 pm
15,33
16,33
78,18
674
3807,65
59,09
Test
100
94
90
82,6
Temperature (°C)
80
88,4
94
88,3
82,6
77
88,3 71,3
70
77
60
60 50
Ta (°C)
Ti (°C)
To (°C)
40 30 20 10 0
34,6 18 17
16
15
17
17
17
15
14
16
16
16
16 15
17 16
17 16
16 15
15 14
15 14
> 2 pm > 8 am > 8 am > 2 pm > 2 pm > 2 pm > 8 am > 2 pm > 2 pm > 8 am > 2 pm > 2 pm
Fig. 8. Outlet, inlet and ambient temperature during the 12 tests
maximum output temperature of 94 °C and irradiance of 810 W/m2 , and a minimum output temperature of 34.6 °C and irradiance of 298 °C. These values can be better appreciated in Fig. 10, which presents a simple set of linear efficiency data with respect to the UL (Tpm − Ta )/It. The dispersion points presented are linear, so a linear trend is applied to show the efficiency increase at a constant fluid rate; the value of R square is 0.9989 which explains the ideal fit of the data to the line because of its good proximity to 1. There are other similar studies applied to space heating by solar thermal collectors. In [4] with a counter-current double flow collector, an efficiency of 45% was obtained, with
Performance Analysis of a Solar Water Heater 94 82,6
Solar Irradiance (w/m2)
800
77
810
700 712 600
762
100
94
88,4
88,3 82,6
810 712
88,3 71,3 77
80 761
761
60
664
664 614
500
90 70 60 50 40
517
30
400 34,6 It (W/m2)
To (°C)
300
20
Outlet temperature (°C)
900
11
10
298 200
0 > 2 pm> 8 am> 8 am> 2 pm> 2 pm> 2 pm> 8 am> 2 pm> 2 pm> 8 am> 2 pm> 2 pm
Efficiency
Fig. 9. Output temperature vs. solar irradiation in the solar collector, during the 12 tests
0.70 0.65 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.10
y = 5.1253x - 0.2235 R² = 0.9989
0.11
0.12
0.13
0.14
0.15
0.16
0.17
0.18
Ul(Tpm-Ta)/It Fig. 10. Efficiency curve of the solar collector working by forced convection for the 12 tests.
an average of 38%. In [24] there is an average efficiency of 20%, with values between 7 and 28% with a convective loop type of collector. In [23], an average efficiency of 58.5% and a daily efficiency of 60% with a perforated plate collector. In the first mentioned study, it was determined that the use of a membrane-type absorber plate composed of circular pipe in branches and metal plate, adhered by welding and with a single convective body at the passage, presents the same thermal properties throughout its flat surface, thus creating more heat transfer area, so that the air supplied by the interior of the absorber plate passes through. This effect is very important and must be considered when manufacturing a solar collector, since this has a significant influence on its efficiency. Thermal energy losses occur when the manufacturing is done by hand and no materials are used to guarantee
12
D. A. Arguello Bravo et al.
the performance of the heating system, so when it is industrialized it has different manufacturing processes that provide hermetic confinement in the set of parts and pieces of the solar thermal collector. This type of solar collector, having this advantage provides that its forced convective flow can work in optimal conditions; this means that it raises the air mass flow inside the pipe and reaches a turbulent fluid that allows the exchange of heat, mass and amount of movement in the fluid mass by effects of transverse velocity factors. This theory is also manifested in the study [22] with a type of double-deck flat-plate collector which states that the increase in efficiency with increased air mass flow is also more evident for turbulent flow because by increasing the aspect ratio, the cross-sectional area of the duct where the air flows is reduced. Thus, it can be concluded that if the heat transfer rates are higher, they will have more impact on the efficiency of the collector. To achieve this condition, it is necessary to have a turbulent flow that circulates inside the collector, avoiding the laminar flow that decreases the collector performance. Therefore the collector has a temperature range of 34.6 to 94 °C at its output, with the highest being 94 °C, with an input temperature of 15 to 18 °C, ambient temperature of 14 to 17 °C and irradiance of 298 to 810 W/m2, with an average of 674 W/m2. As can be seen, the useful heat generated has values that fluctuate from 1022.04 to 4740.77 W, with an average of 3807.65 W, the temperature in the convective flow solar collector by 77 °C and with a mass airflow of 0.06125 kg/s. Comparing with several studies already carried out for this type of application, it is found that in [20], it achieves an output temperature of 63 °C, with an input temperature of 34 °C. In this way, the air temperature in a counter-current double-pass solar air heater increases by 29 °C, with a mass flow rate of 0.024 kg/s and with an irradiance of approximately 700 W/m2 depending on the climate presented in that city. In [13], according to the analysis carried out, it is shown that the air entering the dual-flow solar air collector develops a temperature increase of 31 °C, reaching 78 °C at solar noon and with a minimum of 55 °C in the first and last hours of the day, with an irradiance of 543.5 W/m2 and a maximum collector efficiency of 48%. On the other hand, in [14], it argues that the air temperature in the collector of inclined flat plates is lower in the first and last hours of the day because of the low values of irradiance presented in those hours, however a maximum output temperature of 54.85 °C and a minimum of 24.85 °C [19], provides important information about what happens if we work with a constant irradiance of 900 W/m2 , the air outlet temperature increases if the heater length is fixed, since the air thickness decreases, that is, it is directly proportional if the distance from the heater increases, so does the outlet air temperature. Carrying out a comparative analysis between the data of the solar thermal collector, with forced convective flow, with respect to the investigations found, it is determined that this collector is by far the best option for indoor heating, achieving performances that stand out in adverse weather conditions, optimizing resources and generating environmentally friendly energy savings.
Performance Analysis of a Solar Water Heater
13
4 Conclusions In this study, the registered data of the solar incidence were evaluated, determining values of average efficiency of the flat plate solar collector was 59.09% with an acceptable performance of the collector achieving an average useful heat of 3807.65 W, this allowed to reach outlet air temperatures between 34.6 °C to 94 °C, existing a directly proportional relation between the climate conditions and the performance of the collector by its repercussion in the final result when directly influencing the flat surface. The efficiency reached and the outlet air temperature achieved by the geometry and the 9.35 m2 area of the solar thermal collector, indicate that having results of these characteristics promise in the future the applicability of space heating in various places such as schools, medical centers, offices, workshops and industrial warehouses by avoiding excessive energy consumption and reducing greenhouse gases harmful to the environment, developing renewable energy technologies. This solar thermal collector was born from the idea of applying several important concepts today, such as eco-efficiency and resource optimization. These terms deal with how to make better use of what is available to create savings and be able to benefit from them. When these conditions are met, it is possible to provide quality of life to people by meeting their needs and reducing environmental impacts. The solar collector meets all these requirements by generating heating, ventilation and dehumidification in the internal areas of a house or building. Acknowledgements. The authors wish to thank the Faculty of Mechanics of SEK International University with the project P041819 Parque de Energias Renovables, for providing the necessary conditions for the experimental process. The Instituto de Investigación Geológico y Energético (IIGE) and the Japan International Cooperation Agency (JICA) are also thanked for the donation of the solar collector.
Nomenclature Acol Acv Ap Fr BEC dTc dt h1 h2 h3 h4 h5 h6 m ˙ Q1 Q2
Collector area Glass Cover Area Plate area Heat Removal Factor Energy that is accumulated in the solar thermal collector Heat Transfer Coefficient Heat exchange coefficient Heat transfer coefficient between the indoor air and cover Heat transfer coefficient for radiation between plate and cover (glass) Heat transfer coefficient by natural convection between the indoor air and the plate Convection heat transfer coefficient between the insulation and the outside air Mass flow Heat radiated by the absorber plate Hot air convection heat
14
Q3 Q4 Q5 Q6 Q7 Qu S Tpm Ta To Ti Tm Tai Taex Tsky Taes τcv UL Fr η
D. A. Arguello Bravo et al.
Convection heat to the outside air Q4 Heat from radiation into outer space Q5 Heat delivered to indoor air Heat from radiation into outer space Heat delivered to indoor air Heat that is lost to the environment through insulation of the collector bottom Heat delivered by radiation to the collector cover Useful heat Solar radiation absorbed per unit area Average plate temperature Room temperature Outlet fluid temperature Fluid temperature at inlet Average temperature between Tp and Ti Indoor air temperature Outside air temperature Sky Dome Temperature Outside air temperature through insulation Glass Transmittance Overall thermal loss coefficient of the collector Heat Removal Factor Optical efficiency of the collector
References 1. Ferrer, J.M.I., Salas, J.M.: Dimensionado de un sistema térmico solar mediante simulación y su validación energética. Ingeniería Energética, vol. XXXIV, no. 1, pp. 56–65, January 2013 2. Victoria Cardozo, J.D.F.F.: Diseño y construcción de un calentador solar didáctico. Revista de la Sociedad Colombiana de Física, vol. 37, no. 2, pp. 338–348 (2005) 3. Nandwani, S.S.: Energía Solar - Conceptos básicos y su utilización. Universidad Nacional, Heredia, Costa Rica, p. 26, June 2005 4. Quiñonez, J., Hernández, A., Larsen, S.F.:Evaluación termoenergética de un colector solar calentador de aire de doble paso en contracorriente diseñado para la calefacción de edificios. Avances en Energías Renovables y Medio Ambiente, vol. 16, pp. 55–62 (2012) 5. Gallegos, H.G., Righini, R.: Ángulo Óptimo para planos colectores de Energía Solar integrados a Edificios. Avances en Energías Renovables y Medio Ambiente, vol. 16, no. 31, p. 7, October 2012 6. Giraldo, J., Arango, J.P.: Proceso de optimización en el diseño de sistema de calefacción solar pasivo, Revista Técnica “energía”, vol. 16, no. II, pp. 100–110, Enero 2020 7. Mohamad, A.: High efficiency solar air heater. Solar Energy 60(2), 71–76 (1997) 8. Pottler, K., Sippel, C.M., Beck, A., Fricke, J.: Optimized finned absorber geometries for solar air heating collectors. Solar Energy 67(1–3), 35–52 (1999) 9. Ammari, H.D.: A mathematical model of thermal performance of a solar air heater with slats. Renew. Energy 28(10), 1597–1615 (2003) 10. Kutscher, C.F., Christensen, C.B., Barker, G.M.: Unglazed transpired solar collectors: heat loss theory. J. Solar Energy Eng. 115(3), 182–188 (1993)
Performance Analysis of a Solar Water Heater
15
11. Kutscher, C.F.: Heat exchange effectiveness and pressure drop for air flow through perforated plates with and without crosswind. J. Heat Transfer 116(2), 391–399 (1994) 12. Ramani, B., Gupta, A.V.S., Kumar, R.: Performance of a double pass solar air collector. Solar Energy 84(11), 1929–1937 (2010) 13. González, S.M., Larsen, S.F., Hernandez, A.: Simulación del comportamiento térmico de un colector solar de aire de doble flujo mediante el software SIMUSOL. In: Comunicaciones del XXXV Congreso de ASADES, Rosario, Santa Fe, October 2019 14. Lammardo, A., Baritto, M.: Modelo matemático del comportamiento térmico de un colector solar de placas planas inclinadas para calentamiento de aire, Revista Ingeniería UC, vol. 17, no. 3, pp. 19–27, December 2010 15. Zhang, D., Li, J., Gao, Z., Wang, L., Nan, J.: Thermal performance investigation of modified flat plate solar collector with dual-function. Appl. Therm. Eng. 108, 1126–1135 (2016) 16. Othman, M.Y., Hamid, S.A., Tabook, M.A.S., Sopian, K., Roslan, M.H., Ibarahim, Z.: Performance analysis of PV/T Combi with water and air heating system: an experimental study. Renew. Energy 86, 716–722 (2016) 17. Nematollahi, O., Alamdari, P., Assari, M.R.: Experimental investigation of a dual purpose solar heating system. Energy Convers. Manag. 78, 359–366 (2014) 18. Fauroux, L.E., Jagër, M.: Diseño y Análisis de Colectores Solares Planos. In: Memorias del COINI 2013 UTN FRSR (2013) 19. Koulibaly, A., Bayón, J.J.G.: Modelación de un colector solar para calentamiento de aire. Ingeniería Energética 36(3), 292–302 (2015) 20. Quiñonez, J., Hernández, A.: Evaluación y Simulación Computacional de un Modelo físicomatemático el Colector Solar Calentador de Aire de Doble paso en contracorriente diseñado para la calefaccion de Edificios. XXXVI Reunión de Trabajo de la Asociación Argentina de Energías Renovables y Medio, vol. 1, pp. 123–130, October 2013 21. Fauroux, L.E., Diaz, D.O., Blanco, G.E., Degaetani, O.J.: Modelado, y análisis económico de colectores solares planos,” Revista Digital de Departamento de Ingeniería e Investigaciones Tecnológicas de la Universidad Nacional de la Matanza, vol. 1, no. 1, p. 14, May 2016 22. Gómez, A.E.Á., Fandiño, J.M.M., Sarmiento, J.F.B.: Evaluación energética de un colector solar de placa plana de doble cubierta, Ingeniería y Desarrollo, no. 27, pp. 93– 112, January 2010 23. Hernández, A., Salvo, N., Fernández, C., Suloy, H.: Diseño y evaluación térmica de un colector solar calentador de aire de placa perforada para calefacción de edificios. Avances en Energías Renovables y Medio Ambiente, vol. 12, p. 133 (2008) 24. Hernández, A., Fernández, C., Salvo, N., Suligoy, H.: Diseño, construcción y primeros ensayos de un colector solar calentador de aire de tipo loop convectivo para el calentamiento de edificios. Avances en Energías Renovables y Medio Ambiente, vol. 11, pp. 75–82 (2007)
Permeability Characterization of a Composite Reinforced Material with Fiberglass and Cabuya by VARTM Process. Case Hybrid Material Diana Belén Peralta-Zurita1(B) , Diego Jimenez-Pereira2 , Jaime Vinicio Molina-Osejos1 , and Gustavo Adolfo Moreno-Jiménez1 1 Universidad Internacional SEK del Ecuador, Quito, Ecuador
{diana.peralta,jaime.molina,gustavo.moreno}@uisek.edu.ec 2 Instituto Superior Tecnológico Loja, Loja, Ecuador [email protected], [email protected]
Abstract. In this study, the influence of the use of synthetic and natural fiber in the characterization of permeability in composite materials was analyzed. The Vacuum Assisted Resin Transfer Process (VARTM) was applied to glass fiber samples Chopped Strand Mat and Fourcroia mercadilla, known as “cabuya”, to observe the advance of the epoxy resin flow front IN2. Additionally, a sandwich-type hybrid reinforcement with the aforementioned fibers was used and its incidence on the permeability of the compound was measured. The cabuya fiber achieves a reduction of 4. 38% at infusion time compared to fiberglass. In addition, the use of cabuya natural fiber within the compound decreases the infusion time in 7.40% with respect to the 12.14% presented by fiberglass. To determine the permeability of the different fibers, the experimental procedure was used through Darcy’s Law. The calculated permeability was; 7.3628 × 10−11 m2 , 8.5765 × 10−11 m2 , 1.0065 × 10−10 m2 for fiberglass, woven cabuya and hybrid material respectively. Keywords: VARTM · Hybrid compounds · Hybrid material · Resin flow · Porosity · Fiber · Characterization
1 Introduction Composite Polymers Reinforced with Fibers (FCRP) consist of synthetic or natural fibers embedded in a polymeric matrix, where the fibers are responsible for giving the structural properties to the composite material [1]. Within the synthetic fibers, there are those of glass, carbon, boron among others; while in natural fibers are those of aramid, linen, sisal, etc. When using a combination of natural and synthetic fibers together in a polymeric matrix the composite material is called a hybrid. By containing a natural fiber in its composition, the material becomes biodegradable [2], contributing to the preservation of the environment and the use of renewable raw materials [3, 4]. Liquid compound molding [5] (LCM) is used to make components made of FCRP [7]. One of the LCM techniques is called Vacuum Assisted Resin Transfer (VARTM). This process consists of placing the material used as reinforcement (fibers) in an open © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 16–30, 2021. https://doi.org/10.1007/978-3-030-72212-8_2
Permeability Characterization of a Composite Reinforced Material
17
mold and enclosing it in a flexible bag. When applying the vacuum, the pressure inside the bag decreases, causing a reduction of the air and allowing the passage of the resin to impregnate the fibers through pipes arranged in the mold [8]. One of the advantages of VARTM [9]. Is the use of low-cost tools to produce high quality composite parts, which makes it the preferred manufacturing technique in different industries [10, 11]. In order to obtain solid, homogeneous and excellent quality pieces with the VARTM process [12, 13] the permeability of the fibers used as reinforcements must be investigated [14, 16]. Permeability is a measure that indicates the ease with which a fluid flow through a porous material (fibers) [17]. If fiber permeability is high the fluid flows more quickly, whereas if the permeability is low the fluid has more resistance to flow. For molding simulation by liquid compounds (LCM), the permeability [18], is an important factor to predict the time of filling and the possible dry zones of the obtained part [19, 20, 21]. Today there is no standardized method for characterizing the permeability of fiber in composites. Due to this reason, many researchers use experimental approaches based on Darcy’s Law [23, 24]. To determine variables such as: filling time, permeability, viscosity, pressure, speed among others that are involved in the VARTM process or resin transfer molding (RTM). Next, it is detailed the method to determine the permeability of synthetic and natural fiber, [25–27] used by some researchers. In 2006, [18] designed a device to study the behavior of fiber compaction and permeability in the VARTM process. In the case of permeability, [28, 29] tests were carried out with 1, 5, 1 0 and 20 layers of carbon fiber (HTS- 12K-Aero) and Cycom890 epoxy resin, where they used linear injection [30]. With the help of a sensor, located above the vacuum bag, they extracted the data from the position of the resin flow front according to the infusion time. These data were used to determine the permeability [31, 32] from the relationship between the position of the flow front and the filling time using the integral form of Darcy’s Law [33] determined the permeability in the fiber plane, Matt’s fiberglass and Roving. For which, they performed an experimental procedure based on the observation of the VARTM process. They placed two digital video recorders, one on top of the fibers and the other on the bottom; to observe the progression of the resin flow pattern and record the instantaneous position of the resin flow front. Finally, these data were placed in Darcy’s Law to determine permeability. Compared [34] the permeability between ramie fiber and a composite of ramie fiber with jute. The vegetable fibers were placed on the mold and sealed with a vacuum bag. To record the flow time and the position of the flow front, they used a chronometer and a marker respectively. Subsequently, [35] these data were introduced in Darcy’s Law to determine fiber permeability. In this study, a hybrid composite is presented, whose matrix is epoxy resin reinforced with fiberglass and cabuya [36, 37]. In the absence of a permeability database for this type of material [38, 39] the main objective in this research is to characterize the permeability of the hybrid compound [40]. Additionally, the influence of the use of synthetic fiber (fiberglass) and natural fibers (cabuya) on permeability is analyzed.
2 Design For the characterization of the permeability of the hybrid material (Resin + Fiberglass and cabuya), infusion tests were carried out on each of the fibers, and the radial flow front
18
D. B. Peralta-Zurita et al.
movement was recorded. The combination of Glass + Cabuya + Glass was selected due to its excellent mechanical properties as recommended [41, 42]. Woven cabuya fiber and fiberglass type Chopped Strand Mat was used. The samples have a flat rectangular shape of 275 mm × 230 mm. The flow front radial advance data was measured by recording the process, these; allow to determine the approximate value of permeability using Eq. (1) and Eq. (2), a variant of Darcy’s Law for radial flows [43]. uε E= (1) (A) 4tP rf − 1 + ro2 (2) A = rf2 2 ln ro Where: rf = Flow front radius, ro = Resin entry radius μ = Viscosity, ε = Porosity, t = Time, P = Injection pressure. The experiment was repeated 20 times to lower error due to the experimental uncertainty [18], Table 1, shows the samples, orientation and number of layers for the fiber. Table 1. Fibers used in the radial measurement of permeability |Material
Type
Number of layers
Orientation
Fiberglass
Chopped
1
Normal
Cabuya Fiber
Woven
1
0°
Glass + Cabuya + Glass
Hybrid
3
Normal, 0°, Normal
The fibers used in the infusion experiments (see Fig. 1a, 1b and 1c):
Fig. 1. Types of fibers: Fig. 1(a) Chopped Strand Mat fiberglass, Fig. 1(b) Knitted yarn fiber and Fig. 1(c) Hybrid
2.1 Resin Selection Epoxy resin IN2 was used with its respective hardener AT 30 SLOW. Tables 2 and 3 show the characteristics of the same as the hardener.
Permeability Characterization of a Composite Reinforced Material
19
Table 2. Properties of epoxy resin IN2 and hardener Property
Unity
Appearance −
Resin
Hardener
Combined
Clear liquid Clear liquid Clear liquid
Viscosity (25°)
Mpa.s 500–800
10–20
200–450
Density
g/cm3
1.07–1.13
1.12–1.18
1.08–1.18
Table 3. Curing properties of the hardener Trade name
Preservation time Gelling time to to 25 °C 25 °C
Demolition time
Curing time to 25 °C
At 30 slow
80–100 min
18–24 h
24 h
8–11 h
2.2 Equipment and Assembly For the permeability characterization, the following materials and equipment were used: 1.5 cfm vacuum pump, mold structure, vacuum bag Aero film VB200 of 50 a thickness, sealing tape, filming machine, resin container, transparent hose 6 mm diameter, silicone connectors, 3 mm inner diameter spiral hose, reservoir with pressure gauge. In addition, the mold consists of a glass plate with a rectangular shape of 323 mm × 295 mm which allows an optimal assembly of the fiber samples and vacuum extraction hoses. Figure 2 shows the arrangement of the sealing tape, silicone connectors, spiral hose and fiberglass on the mold (see Fig. 2):
Fig. 2. Arrangement of materials on the mold for characterization
The assembly of all the equipment for the characterization of the permeability in experimental form is indicated in Fig. 3, where a) Vacuum pump, b) reservoir with pressure gauge, c) structure of the mold, d) video recorder, e) resin container. (see Fig. 3):
20
D. B. Peralta-Zurita et al.
Fig. 3. Assembly of the equipment for characterization
2.3 Infusion Process The fraction of fiber volume used is 60%. Table 4 shows the conditions of injection pressure, temperature and resin viscosity used for the characterization of permeability, while Table 5 indicates the amount of resin used according to the weight of the reinforcement used on the different experiments. Table 4. Conditions used for the characterization of permeability Injection pressure (Pa) Temperature (°c) Resin viscosity (Pa. s) 8 000
20–25
0.65
Table 5. Amount of resin used for each fiber sample Specimen Reinforcement weight (kg)
Amount of resin (kg)
1
0.023
0.015
2
0.017
0.011
3
0.063
0.042
2.4 Data Collection A recording process was used for data collection. The resin flow front position was measured for times 20, 40, 80, 160, 320, 640 and 840 s on the three fibers samples, and the average was determined. Figure 4 shows the resin flow front average position value using Wolfram Mathematica analysis software. Then, the term A Eq. (2) is graphed from Eq. (1) the term with respect to the times indicated above.
Permeability Characterization of a Composite Reinforced Material
21
2.5 Validation and Numerical Simulation The simulation of the VARTM process is performed on the specimens using the experimental permeability data. Being, this data validated by the simulation of the VARTM process in the Autodesk Moldflow software. Figure 4 shows the comparison between the VARTM process experimentation and the simulation for the Chopped Strand Mat fiberglass. Infusion times of 20, 80, 320 and 640 s respectively. A circular flow front is obtained for each time, since the type of injection used in the simulation is radial, with a tetrahedral mesh of 344 187 tetrahedra, 63 223 number of nodes and average aspect ratio of 33.88, as observed in the experimentation. a) rf= 39.67mm
rf= 34.15mm
b) rf= 63.67mm
rf= 55.57mm
c) rf= 95 mm
rf= 90.35mm
d)rf= 121 mm
rf= 126.31mm
Fig. 4. Comparison between the simulation and experimentation of the VARTM process for fiberglass. a) t = 20 s, b) t = 80 s, c) t = 320 s and d) t = 640 s.
22
D. B. Peralta-Zurita et al.
For the cabuya fiber, the shape of the flow front in the different times of infusion and its respective simulation is more homogeneous in comparison to the fiberglass. (See in Fig. 5a, 5b, 5c and 5d) show the flow front for 20, 80, 320 and 640 s. Shape irregularities can be observed due to different flow speeds on certain regions. a) rf= 34 mm
rf= 29.88mm
b)rf= 57.33 mm
rf= 52.62 mm
c) rf=
rf= 92.88 mm
97.33mm
d) rf= 127 mm
rf= 130.6 mm
Fig. 5. Comparison between the simulation and experimentation of the VARTM process for the Cabuya Fiber. a) t = 20 s, b) t = 80 s, c) t = 320 s and d) 640 s
Figure 6(a) shows certain irregularities at the beginning of the infusion process for the hybrid material, these irregularities differ from the data obtained in the simulation affecting resin flow front measurement. However, as infusion time progresses, Fig. 6 (b), the simulation is consistent with the experimentation process. In Fig. 6(c), when
Permeability Characterization of a Composite Reinforced Material
23
the infusion time is 320 s; the flow front position measurement is 103.33 mm in the experimentation, while in the simulation is 105.41 mm. Also, in Fig. 6(d), it is observed that the resin flow reaches the upper and lower edge when the time is 640 s both in the experimentation and the simulation. a) rf= 34 mm
rf= 37.45 mm
b) rf= 60.67mm
rf= 63.27 mm
c) rf= 103.33mm
rf= 105.41 mm
d) rf= 134 mm
rf= 135.25 mm
Fig. 6. Comparison between the simulation and experimentation of the VARTM process for the hybrid material: a) t = 20 s, b) t = 80 s, c) t = 320 s and d) t = 640 s
24
D. B. Peralta-Zurita et al.
2.6 Discussion of Findings Table 6 shows the resin flow front average position obtained through the experimentation, using VARTM process, at different times of measurement on the specimens of Chopped Strand Mat glass fibers, Cabuya and the composite material used. Table 6. Flow front position for Glass Fiber Type Chopped Strand Mat, Cabuya and Hybrid Material Time (S)
N. experiments
RF (MM) average Glass chopped strand mat
Cabuya
Hybrid
20
20
39.67
34
34
40
20
50
43.67
46
80
20
63.67
57.33
60.67
160
20
78.33
76.33
79
320
20
95
97.33
103.33
640
20
121
127
134
840
20
127.33
135
145
As observed in Fig. 7, at the beginning of the infusion the resin flow front position is greater in the glass fiber. On the other hand, at the end of the infusion, that is when t = 840 s, the position is higher on the Hybrid material, which determines that the resin flow rate and permeability are higher in this material, in the Glass and Cabuya fibers analyzed individually (Table 7).
Fig. 7. Comparison of the position of the advance of the resin front between fiberglass, cabuya and hybrid material
The term A of Eq. (1), are plotted for each of the fibers, as well as for the hybrid material. This term allows to perform a linear adjustment as shown in Fig. 8 and thus obtain the equations of the straight lines to determine the permeability.
Permeability Characterization of a Composite Reinforced Material
25
Table 7. Constant used in the characterization of the permeability of fiberglass, cabuya and hybrid material Time (S) R0 (M) Glass chopped strand mat
Cabuya
Hybrid
rf (m)
A (m2) rf (m)
A (m2)
20
0.006
0.0397
0.0044 0.0340
0.00289 0.0340 0.00289
rf (m)
A (m2)
40
0.006
0.050
0,0081 0.0436
0.00570 0.0460 0.00654
80
0.006
0.0637
0,0151 0.0573
0.01158 0.0606 0.0134
160
0.006
0.0783
0,0254 0.0763
0.0238
0.0790 0.0259
320
0.006
0.0950
0,0408 0.0973
0.0434
0.1033 0.0501
640
0.006
0.1210
0,0733 0.1270
0.0824
0.1340 0.0936
840
0.006
0.1293
0,0860 0.1350
0.0953
0.1450 0.1129
Fig. 8. Linear adjustment for determining the slope of the straight line in the fibers of: a) Glass Chopped Strand Mat, b) Cabuya, c) Hybrid
Table 8 shows the values of the determined slopes, for each linear adjustment made of the fibers used as reinforcement. In addition, the permeability values (K) calculated by Eq. 1 are detailed. Of the three types of fibers analyzed in this study, the hybrid has the highest value of K, therefore, the resin flows more easily.Table 8. Table 8. Permeability calculated for Fiberglass, Cabuya and Hybrid Material Material
Straight slope
u(Pa. s)
ε
P(Pa)
K (m2 )
Fiber of glass Chopped Strand
0.0000996815
0.65
0.4
88000
7.3628 × 10−11
Mat Woven Cabuya fiber
0.000116112
0.65
0.4
88000
8.5765 × 10−11
Hybrid material
0.000136275
0.65
0.4
88000
1.0065 × 10−10
3 Discussion It may seem contradictory that the resin flows more quickly on the hybrid material, since it has a greater number of layers and there should be greater resistance to flow, but as
26
D. B. Peralta-Zurita et al.
presented [39] the addition of jute and ramie fabric (vegetable fiber) reduces the infusion times between 13% and 30%. In this research, when combining the Fiber of Cabuya with the Glass fiber, in a ratio of 20% and 40% respectively, a reduction of the infusion time of 12.14% for the Fiberglass and 7.40% for the Cabuya fiber. In addition, in Fig. 7 it was observed that the advance of the stream flow is greater for the Cabuya fiber than for the glass fiber. For this reason, the permeability is also greater, where K = 8.5765 × 10−11 m2 for the woven Cabuya fiber and K = 7.3628 × 10−11 m2 for fiberglass Chopped Strand Mat. His data also agrees with the study performed [40, 45] which exhibit a high permeability of natural fiber on glass fiber, due to the multi-scale structure of the vegetable fiber and high porosity providing more [46] resin flow channels. Another phenomenon observed at the time of the characterization of the permeability, are the irregular shapes of the resin flow front for one layer of Chopped Strand Mat Fiber glass, while, for the hybrid material, the shapes were more symmetrical. In the research presented [47]. It is explained that this phenomenon is due to an inconsistent distribution of Fiber mass in one layer of Chopped Strand Mat fiberglass, giving rise to irregular and asymmetric flow fronts, which can affect the measurement of the position of the front of the resin flow and therefore permeability. By increasing the number of layers, the flow fronts soften, achieving a greater isotropy of the fibers. This behavior explained by the aforementioned authors, agrees with the results obtained in this investigation [41, 48, 49]. Figure 9 shows the flow front shapes obtained in the study [1, 37, 42].
Fig. 9. Characterization of the permeability on Chopped Strand Mat glass fiber for different number of layers. a) 1, b) 2, c) 3, d) 4 Comas-Cardona et al. (2011).
While, in Fig. 10(a), the irregular flow front obtained in this study is observed, when using one layer of fiberglass; on the other hand, in Fig. 10(b) a more homogeneous and circular resin flow front is seen for the hybrid material (3 layers).
Permeability Characterization of a Composite Reinforced Material
27
Fig. 10. Effect of the number of layers on the shape of the resin flow front. a) 1 layer of Chopped Strand Mat fiberglass, b) Hybrid material (3 layers).
4 Conclusions The permeability of the different fiber depends on several aspects such as: injection pressure, porosity of the material, type and viscosity of the resin, which is why there is a difference of the values determined between different authors [56–58]. From the analysis, it is emphasized that the use of vegetable fibers in combination with synthetic fibers (hybrid material), decrease the infusion times between 7.40% and 12.14%. In hybrid materials, the flow advance of resin is more regular compared to the use of fibers individually, due to a more consistent distribution of the fiber masses. The technique of the radial flow, used in the experimentation, presents the advantage that allows to measure the position of the resin flow front in 2 directions with a single experiment. The VARTM process is applied in multiple automotive, aerospace, military and commercial applications. This process has many critical aspects [59, 60], among which it stands out, to ensure that there are no air leaks during the infusion of resin to avoid defects in the piece and to ensure that the resin completely impregnates the fibers used as reinforcement. In the Chopped Strand Mat fiberglass, it was complicated to obtain circular resin flow fronts, which is why there was great dispersion in the measurements of the position of the resin flow front, influencing the calculation of the permeability.
Abbreviations FCRP Fiber Reforced Composite Polymers LCM Liquid Composite Molding VARTM Vacuum Assisted Resin Transfer Molding
References 1. Yingguang Li, Q.C.: Composite Structures. Compos. Struct. 212(3), 83–93 (2012). https:// doi.org/10.1016/j.compstruct.2019.01.027
28
D. B. Peralta-Zurita et al.
2. Nabil Bouhfid, M.R.-O.: Modelling of Damage Processes in Biocomposites. Fibre-Reinforced Composites and Hybrid Composites, vol. 5 (2019). https://doi.org/10.1016/B978-0-08-102 289-4.00005-9 3. Margarethe Hofmann, A.H.: Critical raw materials: a perspective from the materials science community. Sustain. Mater. Technol. 17(9), e00074 (2018). https://doi.org/10.1016/j.susmat. 2018.e00074 4. Li, Y.X.L.: Permeability and mechanical properties of plant fiber reinforced hybrid. Mater. Des. 86(5), 313–320 (2015). https://doi.org/10.1016/j.matdes.2015.06.164 5. Alen Thomas, P.K.: A review on transition in the manufacturing of mechanical components from conventional techniques to rapid casting using rapid prototyping. Materials today Proc. 5(5), 11990–12002 (2018). https://doi.org/10.1016/j.matpr.2018.02.173 6. Park, J., Lee, H., Lee, S., Kyung, Y., Kim, J.H., Lee, K., Paik, K.-W.: Fabrication and characterization of epoxy molding films (EMFs) for wafer-level and panel-level fan out packages. In: 2018 IEEE 68th Electronic Components and Technology Conference (ECTC), vol. 8, pp. 712–717 (2018). https://doi.org/10.1109/ECTC.2018.00111 7. Lai Jiang, D.W.: Bioresin infused then cured mycelium-based sandwich-structure biocomposites: Resin transfer molding (RTM) process, flexural properties, and simulation. J. Clean. Prod. 207(1), 123–135 (2019). https://doi.org/10.1016/j.jclepro.2018.09.255 8. James, M.: Production of a class 8 truck trailer bed using c-PBT thermoplastic prepreg & vacuum bag processing. In: 2010, 10th Annual Automotive Composites Conference and Exhibition 2010 (pág. 595), Michigan (2010). www.proceedings.com 9. Milad Tajdini, M.H.: An experimental investigation on effect of adding natural and synthetic fibres on mechanical and behavioural parameters of soil-cement materials. Int. J. Civ. Eng. 16(4), 353–370 (2018). https://doi.org/10.1007/s40999-016-0118-y 10. Wafa Ouarhim, N.Z.: Mechanical and Physical Testing of Biocomposites, Fibre-Reinforced Composites and Hybrid Composites. En W. P. Engineering, 3 - Mechanical performance of natural fibers–based thermosetting composites, pp. 43–60 (2019). https://doi.org/10.1016/ B978-0-08-102292-4.00003-5 11. ASTM International. Standard Test Method for Tensile Properties of Plastics, vol. 08.01, Conshohocken (2014). https://doi.org/10.1520/D0638-14 12. Carmen, R.R.: Novel ABS-based binary and ternary polymer blends for material extrusion 3D printing. Mater. Sci. Addit. Manuf. 29(17), 1859–1866 (2014). https://doi.org/10.1557/ jmr.2014.158 13. Kao, C.-T., Lin, C.-C., Chen, Y.-W., Yeh, C.-H., Fang, H.-Y., Shie, M.-Y.: Poly(dopamine) coating of 3D printed poly(lactic acid) scaffolds for bone tissue engineering. Mater. Sci. Eng. C 56, 165–173 (2015).https://doi.org/10.1016/j.msec.2015.06.028 14. Christian Lubombo, M.H.: Effect of infill patterns on the mechanical performance of lightweight 3D-printed cellular PLA parts. MaterialsToday COMMUNICATIONS (2018, in Press). https://doi.org/10.1016/j.mtcomm.2018.09.017 15. Christian Polzin, S.S.: Characterization and evaluation of a PMMA-based 3D printing process. Rapid Prototyp. J. 19, 37–43 (2013). https://doi.org/10.1108/13552541311292718 16. Verleye, S.L.B.: Modelling of the permeability of non-crimp fabrics for composites. En W. P. Engineering, Non-Crimp Fabric Composites, Manufacturing, Properties and Applications, 260e, 242–259 (2011) 17. Mahsa Nouraddini, M.E.: Development and characterization of edible films based on eggplant flour and corn starch. . Int. J. Biol. Macromol. 120 Part B(12), 1639–1645 (2018). https:// doi.org/10.1016/j.ijbiomac.2018.09.126 18. Dominik Bendera, J.S.: Flow rate control during vacuum-assisted resin transfer molding (VARTM) processing. Compos. Sci. Technol. 66(13), 2265–2271 (2006). https://doi.org/10. 1016/j.compscitech.2005.12.008
Permeability Characterization of a Composite Reinforced Material
29
19. Verrey, J.: Dynamic capillary effects in liquid composite moulding with non-crimp fabrics. Compos. Part A Appl. Sci. Manuf. 37(1), 92–102 (2006). https://doi.org/10.1016/j.compos itesa.2005.04.011 20. Christopher, L., Benson, G.T.: Is There a Moore’s Law for 3D Printing? 3D Printing and Additive Manufacturing, 5(1) (2018). https://doi.org/10.1089/3dp.2017.0041 21. Cortés, C.S.: Rapid casting and new technologies in investment casting. Ingeniería e Investigación 26 (2006) 22. DS Ponce, V.G.: Propiedade Mecánicas de compuestos biodegradables elaborados a base de ácido poliláctico reforzado con fibras de abacá’. Revista EPN 33(2) (2014) 23. Edgar Torres, J.L.: Sistema de posicionamiento aplicado a la técnica de impresión 3D modelado por deposición fundida. Revista de Investigación, Desarrollo e Innovación RIDI 3(1) (2012) 24. Yahya Ta¸sgın, N.K.: Production of Kevlar-Carbon-Aramid Composite Materials by using Vacuum Assisted Resin Infusion Molding Method. Ciencia e Tecnica Vitivinicola 33(2), 13–24 (2018) 25. Dinny Harnany, I.M.: The effect of sisal fiber content in the biocomposite product from injection molding process on its mechanical properties. AIP Conf. Proc. 1, 1983 (2018). https://doi.org/10.1063/1.5046268 26. Truong, G.T., Son, M.-K., Choi, K.-K.: Strength and durability characteristics of latexmodified jute/macro-synthetic hybrid fibre-reinforced concrete. J. Adv. Concr. Technol. 17(2), 79–92 (2019). https://doi.org/10.3151/jact.17.79 27. Lee, Y.J.: A prediction method on in-plane permeability of mat/roving fibers laminates in vacuum assisted resin transfer molding. Polymer Compos. 27(6), 665–670 (2015). https:// doi.org/10.1002/pc.20259 28. Fressoli , M.: Impresiones 3D y fabricación digital: una nueva revolución tecnológica? Integr. Trade J. 39, 116–129 (2016) 29. Garima Mittal, K.Y.-S.: Reinforcements in multi-scale polymer composites: processing, properties, and applications. Compos. Part B Eng. 138(4), 122–139 (2018). https://doi.org/10. 1016/j.compositesb.2017.11.028 30. León-Cabezas, M.A., Martínez-García, A., Varela-Gandía, F.J.: Innovative functionalized monofilaments for 3D printing using fused deposition modeling for the toy industry. Procedia Manuf. 13, 738–745https://doi.org/10.1016/j.promfg.2017.09.130 31. Liliana Serna, C., de Aída Rodríguez, S., Fred Albán, A.: Ácido Poliláctico (PLA): Propiedades y Aplicaciones. Ingeniería y Competitividad 5(1), 16–26 (2011). https://doi. org/10.25100/iyc.v5i1.2301 32. Malvika Nagratha, A.S.: Functionalized prosthetic interfaces using 3D printing: generating infection-neutralizing prosthesis in dentistry. Materialstoday Commun. 15, 114–119 (2018). https://doi.org/10.1016/j.mtcomm.2018.02.016 33. Naik, N.K., Sirisha, M., Inani, A.: Permeability characterization of polymer matrix composites by RTM/VARTM. Progress Aerosp. Sci. 65(2), 22–40 (2014). https://doi.org/10.1016/j.pae rosci.2013.09.002 34. Nguyen, A.K., Narayan, R.J., Shafiee, A.: 3D Printing in the Biomedical Field. Encycl. Biomed. Eng. 275–280 (2018). https://doi.org/10.1016/B978-0-12-801238-3.99875-1 35. Jagannathan Sundrababu, P.G.: Characterization and evaluation of mechanical properties of biodegradable reinforced composites material. Materials Today Proc. 5 (6(Part 2)), 14458– 14467 (2018). https://doi.org/10.1016/j.matpr.2018.03.032 36. Ali, M.A., Umer, R.: In-plane virtual permeability characterization of 3D woven fabrics using a hybrid experimental and numerical approach. Compos. Sci. Technol. 173(3), 99–109 (2019). https://doi.org/10.1016/j.compscitech.2019.01.030
30
D. B. Peralta-Zurita et al.
37. Comas-Cardona, S., Zhang, F.: Fiber reinforcements: correlating permeability and local spatial fibrous features. In: 18th International Conference on Composite Materials, Th465. Obtenido de (2011). https://www.iccm-central.org/Proceedings/ICCM18proceedings/ iccm_8.htm 38. Endruweit, A., Long, L.A.: Analysis of compressibility and permeability of selected 3D woven reinforcements. Sage J. Compos. Mater. 44(24), 2833–2862 (2010) 39. Mathilde Berchon, B.L.: La impresión 3D. Guía definitiva para makers, diseñadores, estudiantes, profesionales, artistas y manitas en general. Editorial Gustavo Gili, España (2016) 40. Araya-Calvo, M., López-Gómez, I., Chamberlain-Simon, N., León-Salazar, J.L., GuillénGirón, T., Corrales-Cordero, J.S., Sánchez-Brenes, O.: Evaluation of compressive and flexural properties of continuous fiber fabrication additive manufacturing technology. Addit. Manuf. 22, 157–164https://doi.org/10.1016/j.addma.2018.05.007 41. Xavier Punet, R.L.-T.: Polylactic acid organogel as versatile scaffolding technique. Polymer 113, 81–91 (2017). https://doi.org/10.1016/j.polymer.2017.02.056 42. Sui, X., Xie, X.M.: Creating super-tough and strong PA6/ABS blends using multi-phase compatibilizers. Chinese Chem. Lett. (2018). Chinese Chem. Lett.https://doi.org/10.1016/j. cclet.2018.04.035 43. CarosenaMeola, S.B.: Characterization of PLA/jute composites with infrared thermography. Conference: Frontiers on polymer science, 165–73 (2015). https://doi.org/10.1016/j.msec. 2015.06.028 44. Jin Zhu, J.Z.: Compatibilizer assistant SCF/ABS composites with improved mechanical properties prepared by fused deposition modeling. Polymer-Plast. Technol. Eng. 57(15), 1576–1584 (2017). https://doi.org/10.1080/03602559.2017.1410843 45. Rasero, M.A., Garcia, F.P., Amoros, J.E., Vidal, R.N.: Thermoplastic elastomer addition for recovering recycled plastics. Mechanical and Thermal Characterization. DYNA Ingeneiría e Industria, 87, 526–532 (2012). https://doi.org/10.6036/4674 46. Molina-Osejos, J.V., Ayabaca-Sarria, C., Peralta-Zurita, D.B., Gomez-Rosero, S., MorenoJimenez, G.A.: Mechanical capabilities of semi-rigid thermoplastics ABS and PLA from 3D printing. Int. J. Mater. Prod. Technol. 59(3), 253–269 (2019) 47. Qian Yan, H.D.: A review of 3D printing technology for medical applications. Engineering (2018). https://doi.org/10.1016/j.eng.2018.07.021 48. Rupinder Singh, H.S.: On the additive manufacturing of an energy storage device from recycled material. Compos. Part B Eng. 156, 259–265 (2018). https://doi.org/10.1016/j.compos itesb.2018.08.080 49. Soo-JinPark, F.-L.J.: Thermal properties of epoxy resin/filler hybrid composites. Polym. Degrad. Stab. 97(11), 2148–2153 (2012). https://doi.org/10.1016/j.polymdegradstab.2012. 08.015 50. Zenghao Zhu, Q.W.: On the examination of the Darcy permeability of soft fibrous porous media; new correlations. Chem. Eng. Sci. 173(12), 525–536 (2017). https://doi.org/10.1016/ j.ces.2017.08.021
Effect of Conventional and Ecological Dielectric on the Wire Electrical Discharge Machining WEDM Process on AISI-D3 Steel Cristian Pérez-Salinas1(B) , Diego Molina-Molina1 and Leónidas Ramirez-Gangotena2
,
1 Dirección de Investigación y Desarrollo DIDE, Universidad Técnica de
Ambato, Ambato, Ecuador [email protected] 2 Laboratorio de Materiales y Metrología, Universidad Politécnica Salesiana, Quito, Ecuador
Abstract. AISI-D3 steel is in high demand in the industry to manufacture special tools, dies and stamping tools. WEDM machining is an efficient alternative process for cutting steels that are difficult to machine by conventional processes. However, its parameters must be properly adjusted to achieve an adequate level of machining performance. This document analyzes the influence of a conventional and ecological dielectric on the cutting surface of an AISI-D3 steel workpiece under the controlled parameters of the WEDM process. A DOE was applied using a 24 full factorial design to evaluate the roughness and surface hardness HRC (response variables). The main factors and their significance that affects the response variables were identified through an experimental analysis and ANOVA. Applying SEM on the cut surface, the topography and diffusion caused by the cut were analyzed. The current, the feed rate and the dielectric are factors that affect the surface roughness, obtaining a minimum Ra of 3.86 µm. For HRC the significant factors were current and dielectric, where the ecological dielectric had a greater impact on surface hardness, consequently a benefit is obtained in steel wear. Keyword: Wire electrical discharge machining (WEDM) · Design of experiments (DOE) · ANOVA · Ecological dielectric · Roughness (Ra) · Rockwell hardness (HRC)
1 Introduction Electrical Discharge Machining (EDM) is a manufacturing process that implements thousands of small concentrated and controlled electric shocks per second to cut or shape a conductive material. This process is accomplished when current discharges or sparks occur between two electrodes [1, 2]. Today, Wire Electrical Discharge machining (WEDM) is an important machining tool for producing complex component shapes that would be difficult to machine by conventional processes. Its application reaches areas such as tool, mold and die manufacturing industries for the automotive, aerospace, nuclear, computer and electronic © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 31–42, 2021. https://doi.org/10.1007/978-3-030-72212-8_3
32
C. Pérez-Salinas et al.
fields [3]. WEDM uses a cable as an electrode (tool) which is wound between two reels producing a continuous cutting process (Fig. 1). A liquid dielectric medium is continuously supplied to release the eroded particles and provide a cooling effect. Among the main benefits of using a WEDM is that they can be precise and of an acceptable surface quality without exerting a strong cutting force and with negligible thermal damage in the cutting area. An important advantage is that it is used for cutting materials that are difficult to machine with conventional processes and there is no need for secondary heat treatments after machining [4]. The dielectric is a fundamental part of the process; but for many years, traditional dielectrics have shown negative effects on operators and the environment [5]. Dielectrics such as hydrocarbons affect the skin, respiratory and digestive systems. Its emissions can cause skin, respiratory and digestive disorders [6]. Inflammation of the skin and allergic rashes constitute 50 to 80% of occupational diseases according to [7]. Mold and bacteria resulting from the chemical or biological breakdown of these types of dielectrics contribute to the above problems [6]. In the study [7], it is mentioned that cancerous substances that are dangerous for the skin and the respiratory system include nitrosamines (a corrosion inhibiting substance), ethanolamine and chlorocarbons. Other dielectric products produced by EDM such as aromatic hydrocarbons, sulfur dioxide and monoxide can cause malignant tumors [8]. Researchers have experimented with the use of alternative dielectrics in order to improve process performance, improve operator safety, minimize fire risks and lessen environmental impact [9, 10]. Various types of fluids have been used, such as water, water-based solutions, mixed water-based dielectric chemicals [10, 11]. In recent years, efforts have been made to develop oil-in-water emulsions with a view to improving the environmental, health, and safety aspects of the EDM process. Electric discharge can occur in air; however, it is not stable within for roughing jobs, so its use is not recommendable for such applications. Stability and greater efficiency in machining is obtained with a dielectric fluid. Deionized water is used as a dielectric in WEDM due to its environmentally friendly impact [12]. The downside of deionized water is the corrosion it causes in some materials. This corrosion can be deep and undetectable by simple visual inspection. As this corrosion grows, it jeopardizes the integrity of the finished part. On the other hand, an environmentally friendly dielectric oil has a reduced environmental impact, unlike conventional oils, and avoids the corrosion problems caused by deionized water [13]. In general, water-based wire cutting provides better cutting performance with a small wire diameter (100000
25000
cfu/ml
pH
times higher in the conventional dielectric and about double in terms of conductivity. Although the total of solvents was slightly higher in the ecological dielectric, the total of suspended solids was 5 times lower. Therefore, greater physical chemical deterioration is evidenced in the conventional dielectric. Regarding the chosen values of the microbiological analysis applied to both the conventional and ecological fluids, it is revealed that the conventional emulsion is a medium for the excessive proliferation of bacteria. This is due to the chemical decomposition of its components during the electroerrosive process; therefore, it has been proven that this fluid has a high level of contamination, as is reported in other studies [9, 30]. Conversely, the ecological emulsion TL-ECO250-MG, presents a lower value of bacterial growth. The difference is significant (Fig. 5), since it presents a bacterial concentration below 25%. Finally, the obtained concentration of 25000 ufc/ml does not require the addition of bactericide, according to NTP 317.
Fig. 5. Dielectrics used and stored for microbiological analysis
40
C. Pérez-Salinas et al.
4 Conclusions After using the cutting fluids in experimentation and allowing them to rest for 6 days in storage, only the ecological cutting fluid meets the limits of concentration of bacteria without the need for bactericidal application based on NTP 317. The use of the ecological dielectric has a significant impact on both Ra and HRC, producing a minimal surface texture and increasing surface hardness in the area of the cut done by WEDM; therefore, it can replace the conventional dielectric and present advantages in both surface quality and environmental benefits. From the tests carried out, it can be concluded that the ecological dielectric makes it possible to obtain stability in the surface texture (Ra) despite changes in amperage in the cutting process, unlike the conventional dielectric, which increases the surface roughness as the amperage increases. Finally, the parameters with the greatest influence on surface quality (Ra) are the amperage and the feed rate, which is consistent with the findings of similar studies [4, 31, 32]. It is recommendable to use low values of the two parameters to obtain the lowest surface roughness. Acknowledgment. Thanks to the Technical University of Ambato, the Mechanical Engineering laboratory and the Steel Engineering company for their total support in carrying out this study.
References 1. Bisaria, H., Shandilya, P.: Experimental investigation on wire electric discharge machining (WEDM) of Nimonic C-263 superalloy. Mater. Manuf. Process. 34(1), 83–92 (2019) 2. Molera Sola, P.: Electromecanizado, electroerosión y mecanizado electroquímico. Marcombo (1989) 3. Garg, R., Singh, K., Sachdeva, A., Sharma, V.S., Ojha, K., Singh, S.: Review of research work in sinking EDM and WEDM on metal matrix composite materials. Int. J. Adv. Manuf. Technol. 50(5–8), 611–624 (2010) 4. Maity, K., Choubey, M.: A review on vibration-assisted EDM, micro-EDM and WEDM. Surf. Rev. Lett. 26(05), 1830008 (2019) 5. Kalyon, A.: Fatatit, and Application. The Environmental Impact of Electric Discharge Machining 3(3),123–129 6. Leppert, T.: A review on ecological and health impacts of electro discharge machining (EDM). In: AIP Conference Proceedings, vol. 2017, no. 1, p. 020014. AIP Publishing LLC (2018) 7. Frei, C., Hirt, C., Girardin, R., Dauw, D.F.: A new approach for contamination measurements for EDM dielectrics. CIRP Annals 36(1), 111–113 (1987)x 8. Valaki, J.B., Rathod, P.P., Khatri, B.C.: Environmental impact, personnel health and operational safety aspects of electric discharge machining: a review. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 229(9), 1481–1491 (2015/2014) 9. Srinivas Viswanth, V., Ramanujam, R., Rajyalakshmi, G.: A review of research scope on sustainable and eco-friendly electrical discharge machining (E-EDM). Materials Today Proc. 5(5), Part 2, 12525–12533 (2018) 10. Liu, Y., Ji, R., Zhang, Y., Zhang, H.: Investigation of emulsion for die sinking EDM. Int. J. Adv. Manuf. Technol. 47(1–4), 403–409 (2010)
Effect of Conventional and Ecological Dielectric
41
11. Valaki, J.B., Rathod, P.P.: Investigating feasibility through performance analysis of green dielectrics for sustainable electric discharge machining. Mater. Manuf. Process. 31(4), 541– 549 (2016) 12. Singh, S., Bhardwaj, M.: Characterization, and engineering, eeview to EDM by using water and powder-mixed dielectric fluid 10(02), 199 (2011) 13. Zhang, Y., Liu, Y., Ji, R., Cai, B., Shen, Y.: Sinking EDM in water-in-oil emulsion. Int. J. Adv. Manuf. Technol. 65(5), 705–716 (2013) 14. Kumar, V., Yadav, A.K., Singh, and Management. A review on current research trends in wire-electrical discharge machining (WEDM) 5(1), 103–112 (2016) 15. Klocke, F., Olivier, M., Degenhardt, U., Herrig, T., Tombul, U., Klink, A.: Investigation on wire-EDM finishing of titanium nitride doped silicon nitride in CH-based Dielectrics. Procedia CIRP 77, 650–653 (2018) 16. Mouralova, K., Kovar, J., Klakurkova, L., Bednar, J., Benes, L., Zahradnicek, R.: Analysis of surface morphology and topography of pure aluminium machined using WEDM. Measurement 114, 169–176 (2018) 17. Choudhuri, B., Sen, R., Ghosh, S.K., Saha, S.: Modelling of surface roughness and tool consumption of WEDM and optimization of process parameters based on fuzzy-PSO. Mater. Today Proc. 5(2), 7505–7514 (2018) 18. Majumder, H., Maity, K.: Prediction and optimization of surface roughness and microhardness using grnn and MOORA-fuzzy-a MCDM approach for nitinol in WEDM. Measurement 118, 1–3 (2018) 19. Lodhi, B.K., Agarwal, S.: Optimization of machining parameters in WEDM of AISI D3 steel using Taguchi technique. Procedia CIRP 14, 194–199 (2014) 20. Garg, M.P., Kumar, A., Sahu, C.K.: Mathematical modeling and analysis of WEDM machining parameters of nickel-based super alloy using response surface methodology. S¯adhan¯a 42(6), 981–1005 (2017) 21. Pérez Salinas, C.F., Moya, E., Coello, D.: Uso de un arreglo ortogonal para el análisis del proceso de electroerosión por penetración con electrodos de forma de grafito y cobre sobre micro-fundición de aluminio. Enfoque UTE 9(3), 67–79 (2018) 22. Singh, V.K., Singh, S.: Multi-objective optimization using Taguchi based Grey relational analysis for wire EDM of Inconel 625. J. Mater. Sci. Mech. Eng 2(11), 38–42 (2015) 23. Badii, M., Rodríguez, M.C., Wong, A., Villalpando, P.J.: Diseños experimentales e investigación científica 4(8) (2017) 24. Castaño, E., Domínguez, Q.: Diseño de experimentos: Estrategias y análisis en Ciencia y Tecnología (2016) 25. Kumar, R., Roy, S., Gunjan, P., Sahoo, A., Sarkar, D.D., Das, R.K.: Analysis of MRR and surface roughness in machining Ti-6Al-4V ELI titanium alloy using EDM process 20, 358– 364 (2018) 26. Alavi, F., Jahan, M.P.: Optimization of process parameters in micro-EDM of Ti-6Al-4V based on full factorial design 92(1–4), 167–187 (2017) 27. Prabhu, S., Vinayagam, B.K.: Optimization of carbon nanotube based electrical discharge machining parameters using full factorial design and genetic algorithm 14(3), 161–173 (2016) 28. Pérez-Salinas, C., Nuñez, R.V., Maiza, O.A., Zamora, L.F., Zumbana, J.P.: Regression models for the front grinding process on Grey Cast Iron block-engine 27(3), 510–521 (2019) 29. FORD ESPAÑA, S.: NTP 317: Fluidos de corte: criterios de control de riesgos higiénicos 30. Somashekaraiah, R., Gnanadhas, D.P., Kailas, S.V., Chakravortty, D.J.T.O.: Eco-friendly, nontoxic cutting fluid for sustainable manufacturing and machining processes 11(5), 556–567 (2016)
42
C. Pérez-Salinas et al.
31. Hastono Wijaya, F.G.: The effect of electric current and wire feed rate onWEDM towards the spur gear surface roughness. Int. J. Latest Eng. Res. Appl. (IJLERA) 02(05) (2017) 32. Han, F., Jiang, J., Yu, D.J.: Influence of machining parameters on surface roughness in finish cut of WEDM 34(5–6), 538–546 (2007)
Corrosion Analysis in Different Cookware Materials Javier Martìnez-Gomez1(B) , Marco A. Orozco-Salcedo2 , Augusto Riofrio3 , Gonzalo Guerrón4 , and Ricardo A. Narváez C.2 1 Universidad Internacional SEK, Quito Albert Einstein s/n and 5th, Quito, Ecuador
[email protected] 2 Instituto de Investigación Geológico y Energético (IIGE), Quito, Ecuador 3 Budapest University of Technology and Economics (BME), Magyar Tudósok Körútja 2,
Budapest 1117, Hungary 4 Universidad UTE, Rumipamba y, Bourgeois, Quito 170147, Ecuador
Abstract. Efficient use of energy is the main concern during the implementation of technology migration processes because several factors must be taken into account for this purpose. In this paper, the migration process from cookers based on liquefied petroleum gas (LPG) to electric induction stoves is presented. To accomplish this study, several tests have been performed in three kinds of pots made from different materials: stainless steel, enameled cast iron, and aluminum. The purpose of these tests is to see how different materials for cookware would be affected by introducing them into a salt spray chamber. Indeed, the tests try to reach similar conditions that cooking with salt. The achieved results try to emphasize the selection, in terms of corrosion, of the best material for suitable cookware production applied in induction stoves. The methods established in the standard ASTM B895 – 05 have allowed to accurately evaluate the corrosion property. A wrong choice of materials can lead to an inadequate life cycle assessment of pots. After completing this research, it has been found that enameled cast iron and stainless-steel show higher corrosion resistances. Keywords: Corrosion · Cookware · Induction cookware · Pots · Induction pots · ASTM B895 – 05
1 Introduction Achieving a sustainable energy management system requires maximizing efficient use of energy resources, coupled with the preferential use of renewable energy sources. It is particularly necessary to introduce improvements in the policies of renewable energy and energy efficiency in households [1, 2]. Within this context, The General Assembly of the United Nations (UN) has declared 2014–2024 the “Decade of Sustainable Energy for All”. This resolution tries to improve the statistics about 1.3 billion people worldwide who still live without electricity, and that more than 2.6 billion people rely on traditional biomass for cooking and heating (Smith, 2014). In order to achieve the UN’s goal, the Ecuadorian government is intended © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 43–52, 2021. https://doi.org/10.1007/978-3-030-72212-8_4
44
J. Martìnez-Gomez et al.
to provide universal access to electricity and electricity and clean cooking facilities using electricity to all the habitants of the country [3, 4]. Several studies have been performed on clean cooking facilities using electricity. In this regard, it has been observed that electric coil stoves have low efficiency and high power consumption, for this reason they have not been considered are not a good policy alternative option for an efficient clean cooking facility, due to their low efficiency and high power consumption [1, 3]. In the case of sense, induction stoves seem to be a better option, because they present to improve cooking with higher energy efficiency based on electricity at home/and local energy production. Induction stoves were selected because of their several advantages in comparison to traditional stoves. The most important advantages are: i) energy efficiency, ii) enhanced safety, iii) no waste of energy when cookware is removed from the hob, iv) automatic cut–off function in case of overheating, and v) no emission of harmful flue gases [5, 6]. Furthermore, it has been observed that in areas where power supply is available, special electrical appliances such as electric kettles, rice stoves, ovens and microwaves are also used and lead to a reduction of biomass used for cooking in an effective way [5–7]. Recently, India (Himachal Pradesh) started to develop a program on “access to clean cooking alternatives in rural India”. In nearly 4000 rural household’s induction stoves were introduced to improve clean cooking facilities, as part of this program [6–8] Although, it is not developing a country-wide program, the availability of inexpensive portable induction cooking stoves is shifting the balance towards cooking with electricity. This program is mainly happening in cities because of electricity costs and power availability, but these constraints are changing as electrification expands and prices for induction stoves fall [5–9]. In the case of Ecuador, the National Efficient Cooking Program (NECP) is being projected across the country and the first step has been achieved, improving the electric grid [6–8]. The main impact of the Energy mix change on the Ecuadorian society is the development of the NECP. The NECP objective is to minimize fossil fuels uses and to reduce LPG subsidy. Until now the government of Ecuador subsidizes the LPG for the population. In Ecuador 15 kg of bottled LPG costs 1,60 us$ for the user, while in neighbor countries this price is on average thirteen times higher (17 us$ and 23 us$) [7, 8]. This means that the total cost of this subsidy for the country is about 690 Mus$ per year, and the total cost caused by smuggling lays around 20%. In addition, about 78% of the domestic demand for bottled LPG is imported, which creates a major dependency and therefore a significant outflow of national funds abroad, that considerably affects the balance of trade in Ecuador [8, 9]. So the government’s aim is to increment local electricity production through new hydropower infrastructure, to partly sustain the national cooking energy consumption and displace in this manner imported and subsidized LPG consumption. The NECP is also related with the idea of clean cooking, related with the reduction of harmful gases emissions in the kitchen. Several programs have been developed in order to improve clean cooking in China [10] Africa, [11], India [12] and Burkina Faso [13]. The reason for the implementation of these programs has been mainly because people stopped using stoves based on solid fuels with poor combustion, low energy efficiency and that also produces harmful smoke
Corrosion Analysis in Different Cookware Materials
45
that mainly affects the female and children population, which are most of the time with their mothers during their first years. Another reason to develop these programs is that about 25% of outdoor particle pollution emissions, and significant contributions to CO2 and shorter-lived greenhouse pollutants, can be accounted to the incomplete combustion, and poor energy efficiency, characterizing biomass combustion [10–12]. With 3.9 million premature deaths annually, traditional uses of biomass cooking fuels are now understood to be the largest single environmental health threat in the world, although they only affect about 40% of the world’s population (Smith et al. 2014). Related with this idea, several investigations have been performed, in order to evaluate the reduction of greenhouse gases and concentration of carcinogenic compound emissions in different types of stoves during cooking [11–13]. Indoor levels of particles in overdeveloped countries are much lower than in developing countries, and this is generally attributable to the advancement in technology for general household cooking activities, as the use of cleaner fuels (such as LPG, electricity and natural gas), for cooking and heating [10] instead of traditional biomass. A recent study estimates that in India, more than one million premature deaths occur annually consequently from household air pollution due to the use of biomass for cooking activities [11, 12]. An induction cooker has advantages when compared with a traditional cooker. There are two major advantages of the induction cooker, namely, energy saving and safety enhancement. Also, the induction cooker is provided with different types of built safety functions to reduce potential fire hazard in contrast with electrical and LPG based cookers [11, 13]. Currently, Ecuador is running the migration from liquefied petroleum gas (LPG) based cookers to electric induction cookers plan for changing the productive matrix of the country from LPG and petroleum to electrical energy based on hydroelectric plants. A cookware manufacturing project for induction cookers is necessary to accomplish these policies. It is expected to fabricate and use between 2 to 3 million of induction cookware sets between 2014 and 2016. They would be composed of three pots with bottom diameters of 140, 160 and 180 mm respectively and a frying pan with bottom diameter of 180 mm. Within this policy is really important to choose the material in terms of corrosion resistance. This program is world pioneer campaign called “efficient cooking plan”, and they have been working in several modifications of the electrical grid as well as industry [13, 14]. One of the main global challenges for the twenty-first century is to have enough energy to ensure a reasonable standard of living, clean water to drink and clean air to breathe. Within this context, the ability to manage corrosion is an important aspect of using materials effectively and efficiently in order to meet these challenges. Currently, materials reliability is becoming more important. Otherwise, when reliability is not assured, safety is compromised, and failure occurs [15, 16]. Perhaps the most striking feature of corrosion is the immense variety of conditions in which occurs and the large number of forms in which appears. Numerous handbooks of corrosion data have been compiled that list the corrosion effects of specific material/environment combinations; still, the data cover only a small fraction of the possible
46
J. Martìnez-Gomez et al.
situations and only for specific values of the study involved [17–19]. To prevent corrosion, to interpret corrosion phenomena, or to predict the outcome of a corrosion situation for conditions other than those for which an exact description can be found, the engineer must be able to apply the knowledge of corrosion fundamentals [20–22]. Laboratory corrosion tests are used to predict corrosion behaviors when service history is lacking and time or budget constraints prohibit simulated service (field) testing. They can also be used as screening tests prior to simulated service testing. Laboratory tests are particularly useful for quality control, materials selection, materials and environmental comparisons, and the study of corrosion mechanisms. In this sense, “Standard Test Methods for Evaluating the Corrosion Resistance” of metals parts/specimens by immersion in a sodium chloride solution were employed. This standard is used to measure the corrosion of several home instruments [22–24]. ASTM B895−05 test methods cover a procedure for evaluating the ability of sintered Powder Metallurgy stainless steel and aluminum parts/specimens to resist corrosion when immersed in a sodium chloride (NaCl) solution. The ability of metals for resisting corrosion when immersed in sodium chloride solution is important to their end use. Causes of unacceptable corrosion may be incorrect alloy, contamination of the parts by iron or some other corrosion-promoting material or improper sintering of the parts. For example, undesirable carbide and nitride formations caused by poor lubricant burn off or improper sintering atmosphere. The purpose of this test is to see how different materials for cookware would be affected when introduced into a salt spray chamber. This test tries to reach conditions similar to cooking with salt. A wrong choice of the material can lead to an inadequate life cycle assessment of the pot.
2 Materials and Methods 2.1 Pots Material Corrosion Resistance in a Salt Spray Chamber The following test was conducted using the observational method applied to three kinds of induction pots. Each one is made of different materials. These test methods cover a procedure for evaluating the ability of sintered P/M stainfewer steel parts/specimens to resist corrosion when immersed in a sodium chloride (NaCl) solution. Corrosion resistance is evaluated by one of two methods. In Method 1, the stainlesssteel parts/specimens are examined periodically and the time to the first appearance of staining or rust is used to indicate the end point. In Method 2, continued exposure to the sodium chloride solution is used to monitor the extent of corrosion as a function of time. This standard does not purport to address the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use
Corrosion Analysis in Different Cookware Materials
47
The first one is made of AISI 304 stainless steel in its body and AISI 430 stainless steel in its bottom, the second one is made of enameled cast iron in its body and its bottom, after being applied a vitrification treatment and the last one made of aluminum in its body and stainless steel in its bottom, as shown in Table 1. The pots are immersed in a sodium chloride solution. Then they are examined periodically and the time of the first appearance of staining or rust are used to indicate the end point. In order to perform this test, the following apparatus were required: Table 1. Specifications of the three tested pots. Nº
Body Material Bottom Material
Diameter of the bottom [cm]
Diameter Thickness of of the top the body [cm] [mm]
Thickness of the bottom [mm]
1
AISI 304 Stainless steel
20.00
20.00
0.5
1.8
2
Enameled iron Enameled iron 20.00
20.00
0.7
0.7
3
Aluminum
20.50
2.0
2.5
AISI 430 Stainless steel Aluminum and AISI 430 Stainless steel
20.50
• A salt spray chamber • Sealable Glass or Plastic Jars, of suitable capacity for specimens to be completely covered by the NaCl solution. • Glass Beads (4 mm is recommended). • Glass Stirring Rods. • Tongs (Stainless steel or plastic, nonmetallic plated) In order to perform this test, the following reagents were required: • A sodium chloride solution consisting of 5 6 0.1% (by mass) NaCl shall be prepared using distilled or deionized water conforming to Specification D 1193 (Type 4) and ACS reagent grade NaCl solution. The 5% NaCl solution shall be prepared no less than 16 h before beginning the corrosion testing. • Concentrated HCl. • Distilled or deionized water. In order to perform this test, the following test specimen were required: • Usually test parts are sintered parts, but they may also be standard transverse rupture bars as defined in Test Method B 528. A minimum of five parts/specimens shall be used for each test. • The density of the parts or specimens as well as any post sintering treatments, (that is, coining, repressing, machining, etc.) shall be stated. Parts or specimens shall be free
48
J. Martìnez-Gomez et al.
of oil, dirt, grease and fingerprints. If they have been cleaned, the cleaning method shall be stated. Refer to Practices A 380 and G 1 for recommended cleaning practices. • The use of tongs or gloves, or both, to prevent contamination in handling is suggested. The procedure was the following: • 2 L of water mixed with 104,6 grams of salt were added into the salt spray chamber, as can be seen in Fig. 1 a).
Fig. 1. a) Image of the salt spray chamber for material corrosion test. b) Image of two pots suspended in the salt spray chamber.
• Place one part or specimen per jar on top of the glass beads. Add the NaCl solution to each jar so that the volume of solution, in milliliters, is at least five times the mass of
Corrosion Analysis in Different Cookware Materials
• •
• • •
49
the specimen in grams. The distance from the surface of the part/specimen to the top of the solution should be at least 25 mm. The ratio of the volume of air to the volume of solution in a jar is recommended to be about 1:2 to 1:3. The pots were placed in a suspended position inside the salt spray chamber, as can be seen in Fig. 1 b). The salt spray chamber was covered with a lid, and it was kept the temperature of 30 °C with heater. Examine the parts/specimens after 1/2, 1, 2, 4, and 8 h and at 24 h intervals from the onset of the test. Thereafter, the interval may be lengthened as time progresses until the first appearance of rust or stain. The corrosion life of a part or specimen is the time of the previous examination, that is, the last examination taken before the observation of stain or rust. Do not include corrosion which appears at the interface between a part/specimen and the glass beads This temperature was maintained for 24 h. Some pictures of pots were taken at half, 1, 2, 4, 8, 12 and 24 h of the test. Finally, pots were removed and immediately examined to verify if they presented corrosion or bubbles.
Once the data was obtained from mentioned tests, their respective analysis was made using graphics and tables based on the standard test methods ASTM B895−05 Standard Test Methods for Performance of Range Tops.
3 Results Figure 2 shows pictures of body stainless steel AISI 304 and bottom stainless steel AISI 430 pot, enameled cast iron pot and body aluminum and bottom stainless steel AISI 430 pot. These photos were taken before and after corrosion resistance test in a salt spray chamber. After 8 h of test, aluminum body pots showed a different color and the first appearance of stain or rust. After finishing the test, the corrosion percentage of cookware was analyzed by ASTM B895−05 Standard Test Methods, the results are showed in Table 2. In these results, the area of stain or rust after 24 h is compared to the total surface area of every cookware. After analyzing the obtained results, it was possible to determine that it is not good when pots present a minimum percentage of stain or rust. The aluminum pot showed a little of oxidation due to wrong material choice. It is not suitable for food. While, enameled iron and stainless-steel pots did not present a minimum percentage of stain or rust.
50
J. Martìnez-Gomez et al.
a)
b)
c)
d)
e)
f)
Fig. 2. a) and b) Images of stainless steel cookware before and after corrosion resistance in a salt spray chamber test; c) and d) Images of cast iron pot before and after corrosion resistance in a salt spray chamber test; e) and f) Images of aluminum body pot before and after corrosion resistance in a salt spray chamber test.
Corrosion Analysis in Different Cookware Materials
51
Table 2. Results to perform the test of corrosion of material in saline fog camera. Nº
Body Material
Bottom Material
Corrosion percentage [%]
1
AISI 304 Stainless steel
AISI 430 Stainless steel
1 the Weibull distribution assumes a form similar to the normal distribution (specifically between 2.6 < β < 5.3 it approximates the normal distribution). The scale parameter (ï) represents the characteristic life or the age at which 63.2% of the population has failed. The location parameter G serves to identify the beginning of the distribution along the x-axis and it is related to the minimum life of the product.
Fig. 1. β and η parameter, how affect the distribution [11].
1.3 Reliability Weibull Function The reliability function is the ratio between the distribution n function and the reliability function given by: R(T) = e
β − T−γ η
(2)
70
D. Ayala et al.
2 Analysis and Results 2.1 Parameter Estimation of Weibull Distribution The objective of this research is to determine the parameters of the Weibull function (β, ï and U), which give the flexible behavior to the model. Estimation of reliability parameters is essential to model repairable systems and determine maintenance policies. However, this estimation becomes more difficult when the system failure times are not identically or independently distributed [12]. To determine β, ï and U three methods will be used. 2.2 Maximum-Likelihood Estimation (MLE) It delivers a better approximation of the parameters and consists of solving a system of equations that contains the parameters ï and β implicitly. The solution is based on numerical iteration. To obtain this system of equations, it starts from the hypothesis that the sample (simple random) comes from a Weibull distribution of parameters ï and β. Therefore, its probability density function corresponds to f (t). If the MLE function is applied to this probability density, we obtain, L(η, β) =
n
f(ti)
(3)
i=1
Substituting Eq. (3) into the probability density function f (t), t β ∂F(t) β β−1 exp f(t) = = βt ∂t η η One obtains,
n n βn β−1 −β β ti exp −η ti L(η, β) = ηβ η i=1
(4)
(5)
i=1
We take logarithms on both sides of the equation and look for the parameters ï and β as the values that maximize the likelihood function,
n n β 1 1 i=1 ti Ln(ti) − − Ln(ti) = 0 (6)
n β β n i=1 ti i=1 n 1 1 β β ti η= (7) n i=1
Since the system of equations has no explicit solution, a calculation algorithm or a computer program must be used to solve it [13]. This work uses Microsoft Excel solver to generate iterations until the answer is converged. The algorithms ask for an initial value of calculation, it is convenient to obtain a first approximation of the parameters through some graphical methodology. Also, it is recommended to start by obtaining the parameter β and then calculate ï. Figure 2 shows Weibull probability diagram with the failure parameters and Table 1 shows the results.
Weibull Reliability Analysis in Hydraulic Jet Pumps
71
Table 1. Obtained parameters from MLE. Maximum Likelihood Method, MLE Shape parameter β
2.01
Scale parameter η
408.76
Location parameter U
1.00
Weibull Parameters Field 56 0.00 -1.00
Aprox 2 4.90
5.00
5.10
Aprox 1 5.20
5.30
Linear (Aprox 2) 5.40
-2.00
5.50
5.60
5.70
y = 3.4961x - 20.025 R² = 0.8059
-3.00 -4.00 -5.00
Fig. 2. Scale parameter η with MLE.
2.3 Implicit Method The method calculates the parameters from the mean, standard deviation and variance. It allows to determine ï and β easier than MLE; however, the approximation of the values is not the most adequate [13]. The calculation equations are as follows:
0.5772 (8) η = exp x + β π β= √ (9) S 6
n Ln(ti) (10) x = i=1 n 1 (Ln(ti) − x)2 n−1 n
s2 =
i=1
The results obtained by this method are presented in Table 2.
(11)
72
D. Ayala et al. Table 2. Weibull parameters obtained with the implicit method. Implicit method Shape parameter β
2.34
Scale parameter η
398.06
Location parameter U 1.00
2.4 Reliability Analytics Software Reliability Analytics software, offers the results of the Implicit, MLE, graph, least squares and other methods. The results are in Table 3: Table 3. Weibull parameters obtained with the software. Software Reliability Analytics Shape parameter β
2.01
Scale parameter η
409.96
Location parameter U
1
2.5 Analytical Application Two hypotheses are proposed that can easily be answered once the parameters of the function have been determined. Subsequently, the failures are characterized by analyzing 5 curves that define the failure behavior in HJP. Failure Analytical Application Determine the percentage of HJP that will fail after the first year of operation. For analytical calculations, the parameters obtained by the MLE method will be used. 54% of the HJP will have to fail after having passed their first year of work. The estimated time at which 10% will fail is 21.42 days. Graphical Analysis and Failure Characterization Probability Graph The selected distribution fits the data moderately, the points are located near the continuous distribution line and along it. Therefore, it can be assumed that the Normal distribution (β ≈2) is an appropriate option for the sample. The scale parameter or the characteristic life (ï), is the 63.2 percentile of the data, indicating that 63.2% of the jet pumps will fail in the first 414 days after the start time (t 0 = 0). In Fig. 3, the solid line indicates Weibull distribution maximum probability estimate of failure as a function of time. The estimated tenth percentile in the distribution
Weibull Reliability Analysis in Hydraulic Jet Pumps
73
Fig. 3. Distributions function.
Table 4. Interpretation of the probability graph. Variable
Field 56 hydraulic jet pump
Curve-Data
Moderate fit of the data with the curve for an exponential distribution
η= 414.53 characteristic life p63.2 B10 = p10
146
p90
649
74
D. Ayala et al.
up to the failure time (named as B10 life) is 146 days with a 95% confidence interval and the 90th percentile is estimated at 649 days (Table 4). Reliability Function Figure 4, corresponds to the reliability when a failure occurs, the curve indicates that the reliability decreases throughout the useful life of the pumps. The reliability at 414 days (MTFB) is 0.3678 (36.78%). Upon reaching 560 days, the function changes its slope, modifying the failure mode, in the equipment that exceeds this new limit it is evidenced that the reliability decreases at a lower rate, the interpretation of the curve is indicated in Table 5. Table 5. Parameter Field 56 hydraulic jet pump. Parameter
Field 56 hydraulic jet pump
Reliability at time MTFB
Reliability at 414 days = 36.78%
Change reliability behavior
day 560, reliability decreases at a lower rate
Fig. 4. Reliability plot.
Density Function The probability density function (Fig. 5) is the union of two functions: cumulative function and reliability function. This function shows the probability of failure during
Weibull Reliability Analysis in Hydraulic Jet Pumps
75
the life and survival of the pumps. Figure 5 illustrates the failure behavior that is caused by deterioration of parts of the equipment where the jet pumps work properly until a certain period of life prior to the survival stage. With the operating time, the behavior of the failure mode and the stages it takes according to the density function are displayed. In the first instance, the probability of failure in a certain period of time is found. In the case it was considered from the pump installation (t 0 = 0) until day 414 (MTFB). After exceeding the MTFB barrier, the jet pumps enter into the survival stage where reliability decrease.
Fig. 5. Density function for failure mode.
Hazard Plot The risk function (Fig. 6) for the 4 fields is analyzed in general terms. The results are summarized in Table 6. The risk that the pumps would fail increases over time. The failure function presents a typical behavior in the initial life stage, where a significant number of failures are observed in the first 3 months of operations (t < 97.2 days). Once the 97-day barrier has been overcome, the curve has almost no slope (the risk of failure remains constant) and the pump that reaches this stage less likely fail. Cumulative Failure Plot Figure 7 represents the cumulative distribution function (CDF). This plot describes the reliability of HJPs in terms of when they would fail. The cumulative failure graph shows
76
D. Ayala et al.
Fig. 6. Hazard plot.
Table 6. Interpretation of the hazard plot. Variable
Block 56, HJP
Majority presence of failures In t < 97.4 days. Initially the failure rate is increasing, after this time the slope decreases
the cumulative percentage of HJPs that fail at a particular time. The probability that HJPs would fail on day 414 is approximately 63%. Table 7 summarizes this interpretation. Table 7. Cumulative failure plot interpretation. Variable
Block 56, HJP
Cumulative distribution function 63% = 414.5 days, in general the probability of failure increases with time, however, from day 414 the growth rate decreases
Weibull Reliability Analysis in Hydraulic Jet Pumps
77
Fig. 7. Cumulative distribution function.
3 Conclusions • The use of three different methods for calculating Weibull parameters improves accuracy and, ultimately, the characteristic failure model. The present study uses the results of the MLE method to characterize the failures of the HJP.
Parameter
Implicit MLE
Software
Shape parameter β
2.34
2.01
2.01
Scale parameter η
398.06
408.76 409.96
Location parameter U 1
1
1
• The characteristic life (p63.2 = η) is 414 days, p10 and p90 in the probability graph gave a value of 146 and 649 days respectively. • Analyzed curves show reliability decrease with time. This represents normal behavior due to HJP wear, but there is a significant number of failures associated with the first 97 days. This is the period of time with the highest failure rate, indicating apparent problems associated with the installation of the pumps. It is not ruled out that shutdowns be derived from problems due to poor quality components, design errors or
78
D. Ayala et al.
unforeseen conditions. The failure behavior is modified once the characteristic life (η = 414 days) is reached.From this point on, the failure rate does not increase with the same magnitude, noting that there are fewer problems in the operational life of the HJP.
References 1. Samad, A., Nizamuddin, M.: Flow Analyses Inside Jet Pupms Used for Oil Wells. Madras: Indian Institute of Technology Madras (2013) 2. Hesham, A., Mikhail, S., Mohsen, A.: Jet Pump Performance with Secondary Fluids Differ in Density and Viscosity from Primary Fluid. International Petroleum Exhibition and Conference SPE, Abu Dhabi (2006) 3. Sánchez, J.: Estudio del Sistema de Bombeo Hidraúlico Tipo Jet en el Pozo Libertador 123 del Campo Libertador para la Producción de Petróleo en el Período 2010. UTE, Quito (2011) 4. Coronado, D.: Selección de una Bomba Hidráulica Tipo Jet del Pozo Guanta 3 del Campo Guanta. UTE, Quito (2010) 5. Hatzlavramidis, D.: Modelling and Design of Jet Pump. Technical Conference and Exhibition SPE, San Antonio (1989) 6. Coterville, J., Ferschneider, G., Hoffmann, F., Valentin, E.: Research on Jet Pumps for Single and Multiphase of Pumping of Crudes. Annual Technical Conference and Exhibition of SPE, Dallas (1987) 7. Mallela, R., Chatterjee, D.: Numerical Investigations of the Effect of Geometry on the Performance of Jet Pump. Journal of Mechanical Engineering Science, Dallas (2011) 8. Manjit, K., Dhruva, P., Aditya, K., Mihir, J., Rohit, T.: Large Scale Jet Pump Performance Optimization in a Viscous Oil Field. SPE 166077- MS, Mangala (2013) 9. Salazar, R., Fitz, E., López, I., Rojano, A.: Confiabilidad y Análisis de Fallas Utilizando la Distribución Weibull. UNAM, México, D.F. (2017) 10. Abbasi, B., Rabelo, L., Hosseinkouchack, M.: Estimating Parameters of the Three-Parameter Weibull Distribution Using a Neural Network. European Journal of Industrial Engineering, Florida (2008) 11. Stack Exchange. (01 de febrero de 2017). Stack Exchange. https://stats.stackexchange.com/ questions/259625/meaning-of-weibull-scale-and-shape-from-flexsurvreg. 12. Nasr, A., Soufiane, G., Mounir, S.: Estimation of the Parameters for a Complex Repairable System with Preventive and Corrective Maintenance. International Conference on Electrical Engineering and Software Aplications, ICEESA, Tunisia (2013) 13. Evans, J., Kretschmann, D., Green, D.: Procedures for Estimation of Weibull Parameters. USDA - University of Wisconsin, Washington, D.C. (2019)
Structural Analysis Method for Aeronautical Modifications in the Integration of Electro-Optical Systems in Helicopters for the Implementation of Intelligence, Surveillance and Reconnaissance (ISR) Capability Roberto Narváez Aguilar1,2(B) , Danny Flor Mancheno2,3 , Diego Paredes Sánchez1,2 , and Flor Garcés Mancero1 1 Universidad de las Fuerzas Armadas, Sangolquí, Ecuador {rcnarvaez,dfparedes1,fegarces}@espe.edu.ec 2 Fuerza Aérea Ecuatoriana, Quito, Ecuador 3 Dirección de la Industria Aeronáutica, Quito, Ecuador
Abstract. Persistent surveillance and collection of airborne intelligence, surveillance and reconnaissance (ISR) information is critical in today’s warfare against terrorism. High resolution imagery in visible and infrared bands provides valuable detection capabilities based on target shapes and temperatures. The design of the installation of the Wescam MX-15i electro-optical (EO) system in helicopter Bell 430, must structurally adapt turret MX-15i system, allowing the operation of the aircraft and system in all capacities. The technical characteristics of the Bell 430 helicopter and the MX-15i EO system are analyzed, in addition to determining alternatives for the location of the system turret, a methodology is established for evaluating dimensional, operational, integration and security aspects; Each aspect is analyzed and it is established that the most suitable location is on the right side of the helicopter at the height of the aircraft floor. An aerodynamic analysis of the installation is carried out establishing the preliminary geometry to be modeled, then the aerodynamic loads to which the installation is exposed are determined using software, and the translational and rotational loads to which the installation is subjected are analyzed. Structural design is presented for the installation of the turret on the aircraft, this design allows disassembly of the system if required for other types of helicopter operation. Design verification subjected to loads takes efforts and displacements, deformations and the safety factor is derived structural installation. It also performs frequency analysis then be compared with the frequency exciters, a fatigue analysis shows structural stability and long life of the installation. Keywords: Aeronautics · Alteration/modification · Bell 430 · Helicopter · MX-15i · Electro-optical system · Intelligence · Surveillance and reconnaissance (ISR) capability
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 79–94, 2021. https://doi.org/10.1007/978-3-030-72212-8_7
80
R. N. Aguilar et al.
1 Introduction The importance of intelligence, surveillance and reconnaissance (ISR) tasks in any security or military operation has been confirmed over time and is even more relevant in missions in the 21st century. When we talk about ISR we are referring to a long list of space, air, land and maritime systems, surveillance and reconnaissance vehicles ranging from satellites to iconic manned aircraft such as U-2 and UAVs, all widely used in armed conflicts, with the main objective of providing intelligence and reconnaissance data through optical images, radar mapping, infrared or electronic signals. Effective data that an ISR platform specialized in any of its functions can provide as an early warning of enemy threats, as well as allowing security and military forces to increase efficiency, coordination and lethality, as well as directly contributing to maximize the resources of the forces of action, which is why today they have become deeply necessary tools for today’s battle scenarios, especially to combat non regular forces or organizations. A clear example of this is the installation of the MX-15i electro-optical system in Bell 430 helicopters and for the design of this installation, important aspects such as aerodynamic, structural, electrical, and electronic compatibility must be considered. Placing a significant weight, in this case 130 lb of the camera, on a small aircraft can represent many modifications in its performance, due to the movement of its longitudinal and lateral center of gravity (CG). The Bell 430 is a certified light-medium twin-engine helicopter according to the requirements in the transport helicopter category (FAR Part 29, Class B) (see Fig. 1).
Fig. 1. The Bell 430 is a certified light-medium twin-engine helicopter FAR Part 29
Figure 2 shows the Wescam MX-15i equipment which is an airborne image capture turret idealized for application in military operations, urban security and surveillance; the system has an optimized infrared image processor which improves dynamic range, reduces noise, corrects damaged pixels, and performs electronic zoom and other image quality improvement functions.
Structural Analysis Method for Aeronautical Modifications
81
Fig. 2. Electro-optical system wescam MX-15i EO/IR
1.1 MX-15i Electro-optical System Components The MX-15i System is equipped with the following basic components specifically required for Intelligence, Surveillance and Reconnaissance (ISR) missions [5] (Fig. 3): • • • • •
Turret with 5 sensors or load detectors Master Control Unit (MCU) in this case, included in the turret. Wiring GPS antenna Control console – – – –
Hand controller P/N: 42228 Displays Recorders Operator Control Unit (OCU)
Fig. 3. Wescam MX-15i system operation diagram
82
R. N. Aguilar et al.
2 Location Analysis 2.1 Analysis of Alternative Locations of the MX-15i Turret in the Bell 430 helicopter One of the most important factors in the installation of a system of these characteristics is the analysis of the location of the turret, operational and technical factors that allow the most optimal performance of the electro-optical system must be considered, restrictions have been identified that affect the operation of the components of the MX15i electrooptical system in the Bell 430 helicopter which are: • • • •
Dimensional restrictions Operational restrictions Integration restrictions Security restrictions
Dimensional restrictions limit turret installation under the helicopter and the available height is 11.1 in. and the turret is 18.95 in. high. Regarding operational restrictions, a relevant aspect is that the aircraft has a high degree of visibility, minimizing the “non-visible areas”, which are the areas where the system does not have visibility in view of the fact that the helicopter structure obstructs it. In addition to making the use of the laser indicator impossible, this is achieved by analyzing the “cutout diagram” of the system’s operation. Taking these considerations into account, the possibility of installing the system at the rear of the helicopter fuselage is ruled out since in this position its “non-visible areas” will be approximately 23% of the system’s visibility capacity, considering only the option of a side installation. The visibility (cutout) diagram is prepared using the methodology recommended by Wescam in which the aircraft is modeled and the turret is placed in the area that needs to be analyzed, later using SolidWorks® software, making vertical axis in the turret, cuts are made every 10° in the azimuth direction, and the resulting geometry is determined in each cut, thus determining the elevation angles that the system will have visibility in a certain location. For electrical integration, the electrical requirements of the MX15i electro-optical system are: • 28 V DC, for power supply to the system. • 10 amperes average, for current consumption in the equipment. • 32.5 peak amps with a duration of at least 0.35 s, for system startup. Given that for the operation of the MX-15i electro-optical system in the Bell 430, not only must the specific electrical requirements of the helicopter be considered, but it must also include the additional components that are required for its operation (monitors, lights, video recorders, etc.). 2.2 Preliminary Location The analysis determined that the most suitable location will be the right side of the helicopter’s fuselage between stations STA 178.50 and STA 196.50, so the visibility
Structural Analysis Method for Aeronautical Modifications
83
diagram of the MX-15i electro-optical system in said location shows that its non-visible area it is 13.7% (Fig. 4).
Fig. 4. Right side of the aircraft coutout diagram
3 Aerodynamic Analysis 3.1 Loads Determination For the design of the installation, three types of loads that can occur during the operation of the MX-15i electro-optical system in the Bell 430 helicopter have been considered, these loads can be defined as: • Aerodynamic Loads • Translational Loads • Rotational Loads Aerodynamic loads are defined as the forces generated when a body moves within a fluid medium, in this case air, the loads have been considered in each of the Cartesian axes: longitudinal Z axis (drag), X axis transverse and vertical Y axis [3]. The operation characteristics of the aircraft have been determined in order to be able to enter these conditions as data in order to calculate the values of the aerodynamic loads in each of the axles using Flow Simulation software: • Analysis speed 150 Kt. • Relative wind movement: Negative direction of the Z axis
84
• • • • • • • • • • • •
R. N. Aguilar et al.
Air fluid medium in conditions at sea level. Type of analysis: External Fluid reference axis: Z axis Type of fluid: Air Flow type: Laminar and turbulent Thermal conditions: Adiabatic wall Roughness: 0.1 μm Fluid pressure: 101325 Pa Temperature: 293.2 °K Speed in X: 0 m/s Speed in Y: 0 m/s Speed in Z: −78 m/s
With these data the results were obtained by performing the simulation in the software; a safety factor of 1.5 is applied to these values according to the FAR in its part 29.303 (Safety factor) that quotes: “Unless otherwise indicated, a safety factor of 1.5 must be used. This factor applies to the exterior and inertia loads unless its application to the resulting internal stresses is more conservative.” It is necessary to indicate the methodology that was applied to determine the aerodynamic loads: it was subjected to the modeling of the installation, that is, both the aircraft, turret and support, to a fluid medium in order to establish the total generated aerodynamic loads using the CFD software, later the same effect was simulated but only with the aircraft, establishing that the difference between the two simulations determines the additional aerodynamic loads that the aircraft will support. This methodology allows establishing the drag generated by the turret in addition to the interference drag produced by the interaction between solids in a fluid medium, a characteristic that cannot be evaluated if only the turret is simulated with its support (Table 1). Table 1. Aerodynamic loads generated by the installation Force
Value
Value included safety factor
Fx (transversal) −20.43 N
−30.65 N
Fy (vertical)
−89.94 N
−134.9 N
Fz (drag)
−466.87 N −700.3 N
Translational loads are defined as the loads affected by the G forces to which the aircraft is subjected, in the Helicopter Flight Manual BHT-430-FM-1 Rev. 18 in its chapter 1–9 (Maneuverability) does not indicate the limits of load factor to which the aircraft may be subjected, only the maneuvers prohibited from being carried out, by virtue of this, FAR 29 part 29.337 (Limit maneuvering load factor) was used indicates that: “the range of load factor during maneuvers range from +3.5 to −1.0” [1].
Structural Analysis Method for Aeronautical Modifications
85
This information allows us to establish that the load on the vertical axis is given by the equation: F = m ∗ 3.5 g Where: m: Sum of the masses of the MX-15i electro-optical system turret and its support. g: Acceleration of gravity (9.8 m/s2 ) F = 62.455 kg ∗ 3.5 (9.8 m/s2 ) F = 2142.2 N According to FAR part 29.303 (Safety factor) this value is also multiplied by the safety factor of 1.5 and to have a conservative design criterion, this value will be applied in each of the axes. When speaking of rotational loads, the loads due to the rotational roll movement of the helicopter carried out in its longitudinal axis have been determined. These loads originate from the imbalance and rotational inertia of the MX-15i electro-optical system turret and its support with respect to the longitudinal axis of the Bell 430 helicopter. To determine the loads, the distance in the vertical and transverse axes between the center of gravity (CG) of the aircraft and the CG of the turret with its support must be initially established, since the action of the loads can be represented in the turret applied directly to your CG; These distances will be considered as turning radio of the turret with respect to the longitudinal axis. With the values of distances from the CG of the MX-15i turret and its support to the required axes as indicated in Figs. 5 and 6, the equation that relates the angular speed, the mass and the turning radius is used in order to determine the rotational force on the longitudinal axis; these values also consider the FAR in its part 29.303 (Safety factor) with a safety factor of 1.5, the values obtained are absolute values of rotational force, however for calculation purposes the most critical conditions have been considered in addition to the sense of the axes for simulating loads (Table 2). FR = m ∗ ω2 ∗ r Where: FR : Rotational force m: Mass of the MX-15i electo-optical system turret and its support ω: Angular velocity r: Radius of gyration, distance to CG It can be established that aerodynamic loads are greater than rotational loads and less than translational loads, this is because the geometry of the turret is designed to generate the least possible aerodynamic resistance, however the mass of the turret multiplied by the gravities and the safety factor generates translational loads. Regarding rotational loads, it is necessary to consider that the roll rotation at 12º/s, that is 0.21 rad/s, was taken into account, establishing the dynamic relationships between the mass that increases to the aircraft and its angular movement.
86
R. N. Aguilar et al. Table 2. Rotational loads generated by the installation
Turn
Turret MX-15i & supportmass (kg)
(rad/s)
Rotational force in X axis (N)
Rotational force in Y axis (N)
Roll
62,45
0.21
−3.17
−2.75
Fig. 5. Three-dimensional distance between the CG of the aircraft and the turret
4 Design of Installation Components Operational and maintenance requirements require that the MX-15i system installed on the Bell 430 helicopter be modular and detachable, that is, allow the turret of the system to be installed on other aircraft in a suitable way. For this, it has been decided to design an installation attached to the fuselage in the lower right lateral area between stations STA 171.60 and 196.50, which will allow maintaining the required level of visualization and fastening the turret support to main members of the fuselage structure. The bracket design will consist of a 7075 T651 0.375” aluminum plate in which 6 holes will be drilled in order to distribute the 6 NAS 1351-4 bolts (recommended by the system manufacturer) that initially support the turret. The aluminum plate will be held by means of “L” profiles, which will provide the rigidity and resistance required in case of plate flexion and will be bolted to a reinforcement attached to the skin of the aircraft which will be riveted with rivets MS20470AD42 which has a diameter of 1/8, the plate will be fastened on its upper part by means of two diverging tubes that will be attached to lugs that will be bolted to both the horizontal plate and the vertical reinforcement of the aircraft. The tubes must be diverging from the axis of the CG of the camera to the ears, managing to form a structural “triad” which allows to resist the loads in the three coordinate axes X, Y, Z [2].
Structural Analysis Method for Aeronautical Modifications
87
Fig. 6. Proposed external structural design
The lower and upper ears will be manufactured by a CNC machine in 7075 T6 aluminum according, its design by means of the inclined ears allows to direct the line of action of the loads applied in the CG of the turret to the three axes X, Y, Z in addition of transferring said efforts to the upper ears and these in turn to the reinforcements installed inside the aircraft. The exterior reinforcement of the skin of the helicopter will consist of a patch made of 2024 T3 aluminum of 0.040 thickness which will give rigidity and stability to the components, in addition to protecting the skin of the aircraft and will be riveted to it by means of rivets MS20470AD4-2. The most critical structural modifications for the installation of the MX-15i electro-optical system in the Bell 430 helicopter will be those made within the aircraft in order to structurally compensate the efforts generated by both aerodynamic, translational and rotational loads and that will be transmitted to the helicopter. To meet this objective, structural reinforcements have been designed in the sections where both the ears and the “L” profiles will be bolted, these sections are between WL 30 and WL 55, the modification consists of five reinforcements made of 7025 T6 aluminum of 0.125 thickness, the use of filler sheets that level the thickness of the internal skin of the helicopter with the edge of the frames, that is to say 0.071 , is also considered; The design also contemplates an internal padding for the internal reinforcements of 0.125 which will allow full contact between the reinforcements that have straight faces with the curvature of the skin of the aircraft (Fig. 7).
88
R. N. Aguilar et al.
Fig. 7. Internal structural reinforcements
5 Modeling and Verification of Design 5.1 Loads and Restrictions The loads applied to the structural modification have already been obtained, these loads were added and their result was applied to the center of mass of the MX-15i electrooptical system turret, the results of the applied loads are (Table 3): Table 3. Summary of applied loads Load type (Newtons)
Axis of application X Axis (Cross)
Y Axis (Vertical)
Z Axis (Longitudinal)
Aerodynamic
−20.43
−89.94
−466.87
Translational
−2142.20
−2142.20
−2142.20
Rotational
−4.75
−4.13
0
Total
−2167.38
−2236.26
−2609.07
The maximum total values of loads applied in each one of the axes of the aircraft are obtained, this configuration of loads would be the most critical that could occur during the flight operations and operation of the MX-15i electro-optical system. The restrictions to be applied in the simulation were determined. The type of restriction is “fixed geometry”, being applied to the external surfaces of the frames, trying to reflect the effect that the structure of the helicopter has with the modification made, there are internal constraints given within the assembly that are referred to as mates and
Structural Analysis Method for Aeronautical Modifications
89
that determine the movement limitations or the degrees of freedom of the components joining.. 5.2 Static Analysis Result The stress plot under the Von Mises criterion indicates a clear stress distribution very concentrated in the area of the structural modification, it is observed that the maximum value is 115.6 Mpa being located in the upper “L” profile (Fig. 8).
Fig. 8. Stress plot
The displacement plot shows what is intuitively predicted, the maximum displacements will occur at the front and rear ends of the semicircular plate, although the values are extremely low being in the order of 0.85 mm displacement in the most critical condition, that is, with maximum severity, translational speed and rotational speed, an unlikely but not ruled out situation. The unit deformation is defined as the deformation per unit length of a material subjected to a load, it is dimensionless in the case of the model it is determined that the maximum unit deformation occurs in the “L” profile precisely where the maximum stresses were presented, its value is 1,027 × 10−3 . The plot of safety factor allows to visualize the distribution of the safety factor on the model and to locate the areas that have minimum safety values and less than 1.5 in the case of aviation, the calculation uses the criterion of Maximum voltage of Von Mises in view that all the components are ductile, it can be seen that the minimum safety factor of the structural modification is 4.1, that is to say that its safety margin is 310%, that is, it has good structural stability of the installation, the graph shows in red the areas of the components that have the lowest safety factors.
90
R. N. Aguilar et al.
The simulation also determines that the load on the most critical bolt, which is located in the front of the upper profile, is 383 lbf at tension and 123.2 lbf at shear, with its efforts being 42.22 Ksi and 13.58 Ksi respectively, obtaining safety margins of 2.7 tensile and 3.4 shear, considering that the NAS 1351-4 bolt has a creep resistance of 160 Ksi and a shear resistance of 60 Ksi (Fig. 9).
Fig. 9. Safety factor plot
5.3 Frequency Analysis Strictly, the loads, except for the own weight, are never static in a pure form since they must be applied in some way to the supporting structure, and said application of load presupposes a variation in time. However, if the application is slow enough, loading will not produce dynamic effects and, therefore, may be treated as static. To determine if a load varies “slowly”, its own period must be compared with the “natural period” of the structure on which it is being applied. The natural period is the time it takes for the structure to go through a cycle of free vibration; vibration that occurs after the variable load has been removed or has stopped varying. The most important dynamic characteristics of a structure are: natural periods and damping, there being a difference between them in the order of importance: the natural period is always fundamental in the dynamic phenomenon and influences all cases of dynamic loads, while damping in others may not be relevant (Table 4). According to MIL-STD-810F Annex C, figure 514.5C-11 indicates the different areas that are affected by vibrations in a helicopter, observing that the area where the MX15i electro-optical system installation was designed is affected due to the vibrations caused by the main rotor of the aircraft, according to the information obtained from
Structural Analysis Method for Aeronautical Modifications
91
Table 4. Structural natural frequency Mode Frequency (Hertz) 1
311.08
2
329.43
3
597.59
4
625.46
5
755.81
the manufacturer through the “Bell Costumers Service Engineers” it is established that in the Bell 430 helicopter, the vibration due to the rotation of the main rotor is 5.8 Hz and multiplied by the number of main rotor blades (4), there would be 23 Hz of exciter frequency (Fig. 10).
Fig. 10. Vibration zones in Bell 430 helicopter
It can be seen that the excitation frequencies are very low with respect to the vibrational frequencies of the structural modification, so there will be no resonance problems. 5.4 Fatigue Analysis Fatigue is a type of break that those parts or components that are subject to cyclical or variable loads can experience, these loads can be rotation, bending or vibration; the cyclical stresses to which the pieces are subjected cause the breakage to lower stresses than those established in the static analysis. “The parts affected by fatigue have fracture surfaces with two zones that are easily distinguishable with the naked eye, the first zone has a matte appearance, with grooved surfaces like a shell and that is generated at the point or surface defect of the part; In this area there are beach marks or ridges in a semicircular shape that indicate how the
92
R. N. Aguilar et al.
crack has advanced with the number of cycles. The second zone of rupture has a shiny appearance and a fine grain is observed because it is the part that undergoes a fragile final rupture”. “It is necessary to consider that fatigue is the first cause of breakage of metallic materials, 90% of these fail due to fatigue and in aeronautics 18% of the causes of accidents are due to material fatigue” [4]. Fatigue analysis is performed of both aerodynamic, translational and rotational loads applied in the structural modification in order to determine the cycles that the design will operate without failure due to fatigue aspects (Fig. 11).
Fig. 11. 7075 T6 Alloy S-N Curve
For this, the S-N curves of the materials involved in the installation were identified, these being aluminum alloy 2024 T3, 7075 T6 and AISI 4130 steel alloys. Then, the event to be analyzed was created, which was the one shown in the static study that was initially carried out and a stress ratio R = 0 was considered, since an intermittent cycle occurs in which a state of maximum stress is presented when the turret is installed and the flight is made and when the turret is disassembled. The structural modification for the installation of the MX-15i electro-optical system in the Bell 430 helicopter shows a failure at 2.18 × 108 charge and discharge cycles in the area indicated in the graphic, which corresponds to the upper external angle, the The rest of the components need a greater number of fatigue failure cycles, the certain degree of uncertainty must be considered because the analysis does not consider the load changes of smaller magnitude but of greater frequency that occur in flight, so it is necessary to apply a preventive maintenance measure to compensate for such uncertainty, this measure will be to carry out NDI inspections (Visual every 20 h of flight and through penetrating dyes every 200 h of flight). A load factor graph is also presented, which represents the safety factor, that is, it indicates the relationship between the value of the stress that causes the fatigue failure with the stress, the structure breaks when the loads applied to the model are multiplied by
Structural Analysis Method for Aeronautical Modifications
93
the result of the indicated minimum safety factor, as previously indicated, the structure will support 2.18 × 108 loading and unloading cycles, considering that this would allow the system to operate properly for the expected life cycle, in addition to ensuring the monitoring of the structure through NDI inspections (Fig. 12).
Fig. 12. Modification life plot
6 Conclusions • The integration of Intelligence, Surveillance & Reconnaissance (ISR) capability in aircrafts is feasible to transform civil aircraft into useful platforms for gathering information for use in security and defense missions. • It is determined that once the design method and structural analysis for the integration of Wescam MX-15i electro-optical system have been applied in the Bell 430 helicopter, it is feasible to apply the aeronautical modification while maintaining the airworthiness of the aircraft. • A method is proposed that covers all aspects of structural integration in the modification of an aircraft, considering aerodynamic, rotational, translational loads required by aeronautical regulations, as well as dynamic, aeroelastic and fatigue aspects are also analyzed, ensuring an optimal design.
References 1. Federal Aviation Administration: Advisory Circular AC-43-13-2B “Acceptable Methods, Techniques and Practices - Aircraft Alterations”. AFS-300, Washington, DC 20591 USA
94
2.
3. 4. 5.
R. N. Aguilar et al. (2008Electronic Code of Federal Regulations: Federal Aviation Regulations 29, Title 14 "Aeronautics and Space", Chapter I "Federal Aviation Administration, Department of Transportation", Part 29 "Airworthiness Standards: Transport Category Rotorcraft", Washington DC , USA (1996) Fusell, F.: Aircraft video camera mount Airborne vehicle referenced recording device utilizing an electro-optical camera and an electronic alignment procedure. United States Patent 4,380,024 (1983) Roskam, F., Lan, S.: Airplane Aerodynamics and Performance. DARcorporation, Lawrence, Kansas 66044, USA (1997) Avilés Rafael, F.: Análisis de Fatiga en Máquinas. Paraninfo, Madrid España (2005) L3Harris Technologies: Wescam MX 15 Fully Digital. L3Harris, NASA Boulevard Melbourne, FL 32919, USA (2019)
Impact Analysis of Migration from Súper Gasoline to Others of Lower Octane Number in Ecuador Carlos Francisco Terneus Páez1,2(B) , Absalón Guillermo Cabrera Mera1 and Rubén Darío Grandes Villamarín3
,
1 Universidad de las Fuerzas Armadas ESPE, Av. General Rumiñahui, s/n, Sangolquí, Ecuador
{cfterneus,agcabrera}@espe.edu.ec 2 Escuela Politécnica Nacional, Av. Ladrón de Guevara, E11-253 Quito, Ecuador 3 Agencia de Regulación y Control de Energía y Recursos Naturales No Renovables,
Quito, Ecuador
Abstract. In 2018, the Ecuadorian government reduced the subsidy of the 90 octane gasoline through the Executive Order 490. This led to a decrease in the consumption of this fuel and a migration to a lower price and octane gasolines. The objective of this research is to analyze the energy and environmental impact of the migration from Súper 90 octane gasoline to a lower octane gasolines, known in Ecuador as Extra and Ecopaís. For this investigation, we carried out the annual gasoline consumption calculation of the transport sector, in order to study the energy and environmental implications of the issuance of this Executive Order. LEAP (Long - range Energy Alternatives Planning) software was used, which allows calculating energy consumption from data such as: vehicle fleet size, vehicles categories or types, average fuel consumption, annual route and others. Two analysis scenarios were proposed; the first SINDEC that refers to the gasoline consumption pattern before this measure was taken; and, the second, CONDEC, which considers the migration of gasoline consumption from Súper to Extra and Ecopaís and its effect on the variation of that pattern. One of the results due to this measure, is that, the country increased its fuel consumption by around 3%, with the corresponding increase in the emission of carbon dioxide, which is significantly higher than the reduction obtained by a strategic government project of alternative energy generation. Keyword: Gasoline · Energy consumption · Energy policy · Transport · Energy economics
1 Introduction In Ecuador, the growth of the economy in the decade 2007–2017, derived from high oil prices and a better tax collection, which turned into an increase in imports, mainly of motor vehicles. From 2008 to 2018, the vehicle fleet increased by 1.484.743 vehicles, reaching 2.403.651, an average annual growth rate of 10,7% [1]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 95–108, 2021. https://doi.org/10.1007/978-3-030-72212-8_8
96
C. F. Terneus Páez et al.
In 2018, the transport sector demanded 48% of all the consumed energy, 45% corresponds to land transport [2], a number significantly higher than the average for the Latin American and Caribbean region, where it does not exceed the 35% [3, 4]. The energy requirement of this sector increased by 276% in 2012–2013. Compared to the base period 1979–1980, it had an increase of almost three times in energy consumption [5]. In terms of environmental pollution, the transport sector in Ecuador, in 2010, emitted 45% of total greenhouse gas emissions [6]. Considering the demand for fuel in land transportation sector, gasoline represented 51% [2], which exceeded the national production. The Ecuador limited refining capacity [3] forced the country to increase the import of high-octane gasoline to improve the octane number of the Súper, Extra and Ecopaís gasoline, sold in Ecuador. In 2012, in order to improve the quality of fuels, the octane number was increased. The Extra and Ecopaís gasoline octane number increased from 81 to 87 and Súper from 90 to 92 [7]. The Ecopaís gasoline is the mixture of gasoline with anhydrous ethanol that come from sugar cane [8]. In 2014, due to the operational shutdowns at the Esmeraldas Refinery, the octane rating of gasoline was reduced by two points, reaching 85 and 90 octane for Extra and Súper, respectively [9]. The deadlines to return to the octane established in 2012 have been postponed; the last one was established at December 1st, 2020 [10]. In most countries of the American continent, the lowest gasoline classification is 87 octane [11]. In some high altitude areas in the United States (US), 85 octane gasoline has been used. However, some authors do not recommend its use in modern technology vehicles [12]. Splitter [13] makes a historical review since 1925, about the characteristics of the engine and the octane number of gasoline in the US. Chronologically, more efficient energy consumption has been possible by increasing the compression ratio and using higher octane gasoline, which was mainly achieved by two methods. The first, applied since 1920, involved the use of additives such as tetraethyl lead (TEL), that allowed high compression ratios and low fuel consumption. In 1970, the US Congress passed the Clean Air Act, the first national standard regulating exhaust gas emissions. One of the factors that triggered this law was a greater understanding of the effects of TEL, which led to its elimination and therefore a reduction in the octane number of gasoline. [13]. Currently, very few countries use it due to the effects of this component on human health. Due to the removal of TEL in gasoline engines, the efficiency decreased, leading to an increase in fuel consumption in the 1970s, compared to the 1960s. Due to the elimination of TEL in gasoline, engine efficiency decreased, and consequently fuel consumption increased in the 1970s, compared to the 1960s [13]. The second method, promoted since 1950, through the refining process by catalytic cracking, allows to decompose long chain molecules into short ones, which have a high octane number [14, 15]. Starting in 1970, advances in automotive electronic technology made it possible to increase the compression ratio of vehicles, without increasing the octane number of fuels, with the implementation of the knock sensor. This device provides feedback to the vehicle’s control unit, adjusting engine operating parameters, to reduce the tendency to detonation, although this implies some loss of efficiency, which is preferable to the engine operating with detonation [16, 17].
Impact Analysis of Migration from Súper Gasoline to Others
97
On the other hand, Ecuador maintained a pricing policy that implied high fuel subsidies. For instance, in 2014, the cost of these subsidies reached more than 5,000 million dollars. [18]. The percentage of this subsidy with respect to GDP, fluctuated between 2 and 4% [19]. Due to the economic impact that fuel subsidies have made in the fiscal box, President Lenin Moreno, through the Executive Order 490 [20] published in the Official Register Supplement 312 of August 23, 2018, reformed the Regulation of the Prices of Hydrocarbon Derivatives issued by Executive Order No. 338, published in the Official Registry No. 73 of August 2, 2005[21]. The reform determined the terminal sale price of the Súper gasoline for the automotive sector, to be calculated on a monthly basis by EP Petroecuador. However, in April 2019, its determination was transferred to the Agency for Regulation and Control of Hydrocarbons1 (ARCH)2 [22]. The aforementioned calculation is made based on the weighted average cost between the national production, imports, plus the costs of transportation, storage, marketing and an additional margin. The cost of imports fluctuates monthly according to the international price of WTI crude. The public price of the Súper gasoline was $2,98 USD in September, October and November 2018 [23]; while the Extra and Ecopaís gasoline maintained a fixed price of $1,48 USD [24]. Due to this price difference, around $1,50 USD/ gal, consumers migrated to cheaper fuels that are lower octane. This fact was collected and published by the press in various publications [25, 28]. Various studies have been carried out in order to measure the effects of the use of gasoline with different octane numbers on engine performance and in general the conclusion is that the higher the octane, the lower the fuel consumption [13]. Beck, Stevenson, & Ziman [17] study the relationships with four types of gasoline in Germany, in three European vehicles, using a chassis dynamometer. They conclude that with an octane higher than 90, there is a small improvement in fuel economy, and with an octane lower than this value, the fuel economy decreases significantly. Shuai, Wang, Li, Fu, & Xiao [29], carried out an investigation on the impact of octane variation in vehicles in China. They carried out fuel economy tests with octane numbers of 90, 93 and 97. They concluded that an increase in octane number by one unit produces an improvement of 1% on fuel economy. Rashid, Radzi, Racovitza, & Chiriac [30] conducted tests with three types of gasoline with different octane numbers of 95, 97 and 102. In this study, the Mitsubishi Campro CPS 1.6 L4 engine was used, with a compression ratio of 11:1. Under various load and speed conditions, the results of the study indicated that RON 97 fuel shows an improvement in Brake Specific Fuel Consumption (BSFC) of around 24% compared to RON 95 fuel. Binjuwair, Mohamad, Almaleki, Alkudsi, & Alshunaifi [31], carried out tests on a 4-stroke single-cylinder engine, with RON 91 and RON 95 grade gasoline. The experimental results showed the BSFC is lower with the higher octane gasoline. Guzmán et al. [32] carried out an investigation in Quito-Ecuador, using a dynamometer and carrying out tests on the road, which determines the performance of a vehicle, 1 Currently merged with other institutions to form the Agencia de Regulación y Control de Energía
y Recursos Naturales no Renovables. 2 Agencia de Regulación y Control Hidrocarburífero, by its initials in Spanish.
98
C. F. Terneus Páez et al.
using Super and Extra gasoline and their mix. They concluded that with Super a 15% improvement in fuel economy is obtained. Inga y Vidal [33] in Cuenca-Ecuador, used a chassis dynamometer to compare the consumption of a Chevrolet Aveo vehicle, using Super and Ecopaís gasoline. They found that with the first type of gasoline, the performance was 11% higher. Jácome y Villegas [34], in Quito-Ecuador, carried out a study on the variation of specific fuel consumption, using five mixing ratios between the Súper and Extra gasoline. They concluded that by using only the Súper gasoline in laboratory tests with a chassis dynamometer the specific consumption was reduced by 8,7%, while in road and track tests, the volumetric fuel consumption was reduced by 22,4% and 12,6% respectively. Regarding the use of energy and the impact on the environment, the Constitution of Ecuador (2008), in the Article 413 establishes that energy efficiency will be promoted [6]. This country has been part of the United Nations Framework Convention on Climate Change since 1994, that encourages search for energy efficiency in the relevant sectors of the economy of each country. In this context, it is interesting to study this massive decrease in the use of high octane gasoline in Ecuador, due to the decision of the government of this country to no longer subsidize it. Therefore, the present research aims to analyze the energy and environmental impact of the migration of Super gasoline to other lower octane gasoline in Ecuador. This document is organized as follows. After this introduction, Sect. 2 presents the methodology, uses the LEAP model which includes the algorithm, input data, and scenarios. In Sect. 3, there is a discussion of the results, and Sect. 4 presents conclusions and implications for energy policy.
2 Methodology In this research, the calculation of the annual energy consumption of the transport sector that uses gasoline in the different types of automobiles was carried out. For this purpose, the Long-range Energy Alternatives Planning Systems (LEAP) model was used. It was developed by the Stockholm Enviroment Institute (SEI-US), and it allows generating energy and environmental scenarios by entering technological specifications and obtaining a final consumption [35]. This model has been used to perform global energy consumption analysis in the transportation sector. When using LEAP, Shabbir [36] and Bitos [37] analyze the energy demand of transport in Pakistan and in Greece, respectively. On his side, Hong [38] analyzes the effectiveness of policies created by South Korea on the transport sector and examines the effect on energy consumption. The LEAP model for gasoline vehicles in Ecuador was divided in many branches such as city car, pick-up, motorcycle, taxi, van, and SUV. They can use different types of gasoline such as Extra/Ecopaís and Súper. As environmental load only carbon dioxide is considered (see Fig. 1).
Impact Analysis of Migration from Súper Gasoline to Others
99
Fig. 1. LEAP tree
2.1 The Algorithm of the LEAP Model Existing vehicles in a specific year are calculated based on historical sales data and a life cycle profile that describes the survival of vehicles as they age, is calculated as follows: Et,y,v = VE t,v ∗ St,y−v Et,y =
MA
Et,y,v
(1)
(2)
a−0
Where E is the number of existing vehicles in a specific year, t is the vehicle type, y is the calendar year, v is the vehicle age, V is the number of vehicles sold in a specific year, S is the fraction of vehicles that survive, that is entered as a life cycle, and MA is the maximum number of years a vehicle survives in operation. Specific consumption is the use of fuel for the distance traveled by the vehicle, is calculated as follows: CE t,y,y = CE t,y × DCE t,y−v
(3)
Where CE is the specific consumption of the vehicle in (L/100 km), DCE is the specific consumption reduction factor depending on how the vehicle ages, that is entered as a life cycle. The annual distance traveled by a vehicle is calculated as follows: Kt,y,y = Kt,y × DK t,y−v
(4)
Where K is the annual route of the vehicle in (Km), DK is the factor of change of route according to how the vehicle ages, that is entered as a life cycle.
100
C. F. Terneus Páez et al.
Energy consumption is the product between the vehicle’s stock, the specific consumption, and the route, is calculated as follows: CEnt,y,v = Et,y,v × CE t,y,v × Kt,y,v
(5)
Where CEn is the vehicle energy consumption. Emissions related to energy are the product between energy consumption, an emission factor and a degradation factor of the emissions produced, is calculated as follows: Emt,y,v,p = CE t,y,v × FEmt,v,p × DEmt,y−v,p
(6)
Where Em is the air pollutants emission, p is pollulant, FEm is the pollutant emission rate in (gr/km-veh), and DEm is the factor of change in the carbon dioxide emission factor as a function of vehicle age, that is entered as a life cycle. 2.2 Input Data Number of Vehicles in Ecuador. For this research, vehicles were divided into different types and are detailed in Table 1, along with the percentage of existence regarding the number of vehicles nationwide. These values have been obtained from the open database corresponding to the 2017 registration data collected by the National Transit Agency (ANT3 ), which is the official institution in charge of registering motorized vehicles that circulate in Ecuador.
Table 1. 2017 Ecuador vehicle registration. Vehicle Type
Units
City car
697.812
Stock share (%) 33
Pick-ups
363.889
17
Motorcycle
603.628
28
Taxis
68.497
3
Vans
21.405
1
SUV
385.295
18
Total
2.140.526
100
Reference: [1].
An additional 10% of the number of registered vehicles is considered as suggested by articles in the local newspapers [39, 40]. 3 Agencia Nacional de Tránsito, by its initials in Spanish.
Impact Analysis of Migration from Súper Gasoline to Others
101
Sales. In 2018, 293.695 vehicles were sold, according to the Yearbook of the Association of Automotive Companies of Ecuador (AEADE4 ) [36]. The sales percentages according to the type of vehicle are presented in Table 2.
Table 2. 2018 Vehicles sales share. Vehicle Type Sales share (%) City car
21
Pick-ups Motorcycle
3 54
Taxis
3
Vans
1
SUV
18
Reference: [41].
Fuel Consumption. The values obtained were based on information on fuel consumption by engine size, from the IEA [41] which graphs fuel economy as a function of the cylinder capacity. The median cylinder capacity of each type of vehicle was obtained from the Report of new vehicles acquired and reported in the period from January 2012 to June 2020 local purchase and direct import from the database5 of the Internal Revenue Service (SRI6 ) [42]. See Table 3. Table 3. Fuel Consumption of Extra/Ecopaís and Súper Vehicle Type
Median Engine displacement (cm3 )
Fuel Consumption Extra/ Ecopaís (L/100Km)
Fuel Consumption Súper (L/100Km)
City car
1.500
9,6
7,5
Pick-ups
2.400
11,6
9,0
Vans
2.700
12,3
–
Motorcycles
200
3,8
–
Taxis
1.600
9,8
–
SUV
2.000
10,7
8,3
Reference: IEA [43]. 4 Asociación de Empresas Automotrices del Ecuador, by its initials in Spanish. 5 https://www.sri.gob.ec/web/guest/matriculacion-vehiculos#estad%C3%ADsticas. 6 Servicios de Rentas Internas del Ecuador, by its initials in Spanish.
102
C. F. Terneus Páez et al.
A 22% more consumption of Extra / Ecopaís gasoline regarding the consumption of Súper gasoline was considered based on the research carried out by Guzmán [32], Inga [33] and Jácome [34], that performed their research in Ecuador. Traveled Distance. The route per year (KA) of new vehicles was calculated taking as a reference the odometer reading of the vehicle reviews of the Metropolitan Transit Agency (AMT7 ) of the Metropolitan District of Quito (DMQ8 ) carried out in the years 2017 and 2018 (See Table 4). Is calculated as follows: KA = 365 ×
KM 2018 − KM 2017 DD
(7)
Where KA is the distance per year in (km/yr), KM 2018 is the odometer reading in the 2018 revision in (Km), KM 2017 is the odometer reading in the 2017 revisions in (Km), and DD are days between the registration date 2018 and 2017. Table 4. Year Average mileage, new vehicles Vehicle Type KA (km/yr) City car
16.773
SUV
16.131
Pick-ups
29.195
Vans
29.805
Motorcycles Taxis
6.824 54.866
Reference: AMT [1].
Life Cycle Profiles. Describe vehicles distribution according to their years of age [35]. In this study, different profiles are used: efficiency degradation, existence of vehicles for different types, travel by age and vehicle survival, with 32 years as maximum. Fuel Consumption Degradation Profile. This profile indicates the fuel efficiency of a vehicle based on its age versus efficiency when it is new. The study carried out by Rajbahak, cited by Mutenyo, is taken as reference [44], which proposes the following exponential function (8). Fμ = F0 × e(−δA)
(8)
Where Fμ is the fuel efficiency of vehicles according to their age in (km/L), F0 is the fuel efficiency of new vehicles (km/L), δ is the degradation factor that varies according to vehicle age, and A is the vehicle age. 7 Agencia Metropolitana de Tránsito, by its initials in Spanish. 8 Distrito Metropolitano de Quito, by its initials in Spanish.
Impact Analysis of Migration from Súper Gasoline to Others
103
Existing Vehicles Profile. This profile indicates the percentage of the different types of vehicles existing in the base year by age [45]. These values were obtained from the 2017 ANT database. Travel Profile by Age. This profile indicates the percentage of vehicle mileage according to its age, compared to a new one [45]. This information was obtained from the AMT. There are no data on vehicle travel at the national level. In this research we assume that the travel profile in Ecuador is the same as in Quito. Vehicle Survival Profile. Indicates the vehicle percentage that continue to operate according to age. It was prepared taking as reference the data of the ANT [45]. Fuel Type Consumption. To estimate the amount of vehicles that consume Extra / Ecopaís or Súper gasoline, we used the statistics of the SRI website, which presents the value of the new vehicles acquired and reported in 2019. It is assumed that city car, pick-ups and SUV vehicles, valued over $30.000 USD consume the Súper gasoline before the Executive Order 490. The results are in Table 5.
Table 5. Fuel type consumption Vehicle Type
Extra/ Ecopaís fuel consumption (%)
Súper fuel consumption (%)
City car
95
5
Pick-ups
75
25
SUV
52
48
Reference: SRI [42].
2.3 Scenarios Two scenarios were used. The first one called SINDEC, considers there was no Executive Order 490, issued on August 23, 2018. For this scenario, the fuel dispatch of the Súper gasoline was 16%, and corresponds to two previous months, June and July 2018. The second scenario called CONDEC, after the Executive Order was issued, showed a fuel dispatch of the Súper gasoline of 10% in October and November 2018. These values were obtained from the local fuel dispatch data acquired from the ARCH (Fig. 2). In this analysis, the months of August and September 2018 were not considered as they presented an atypical behavior due to the issuance of the Executive order. As previously indicated, the increase in the price of the Súper gasoline caused a reduction of consumption of this fuel from 16% to 10% that is 38%. In this percentage, it is considered that the number of city car, pick-ups and SUVs that consume Súper gasoline is reduced in the CONDEC scenario, as shown in Table 6.
104
C. F. Terneus Páez et al.
Fig. 2. Fuel dispatch
Table 6. Fuel devices share (with the Executive Order 490) Vehicle Type
Extra/ Ecopaís fuel consumption (%)
Súper fuel consumption (%)
City car
92
3
Pick-ups
98
16
SUV
55
30
Reference: ARCH.
3 Discussion and Results The Executive Order 490 caused approximately 118.000 vehicles migrate from the Súper gasoline to a lower octane gasoline, which implies less efficiency. In the Coast region, the percentage of vehicles consuming Súper went from 17% to 12%, while in the Highland region it went from 14% to 9%. Due to the altitude of the Highland region, a lower fuel consumption of high octane gasoline is expected because it is harder for the detonation effect to occur. For instance, Quito is located at 2.850 m above sea level. However, Vertin et al. [46] indicates that the use of an 85 octane gasoline at high heights implies higher fuel consumption anyway. When a driver switches from Súper to Extra or Ecopaís gasoline for his vehicle, the possibility that the engine will detonate increases. However, current automotive technology prevents the engine from reaching a state of detonation by modifying its operating parameter [46]. This implies, however, a higher fuel consumption. This higher fuel consumption went unnoticed by the common driver due to the low price of Extra or Ecopaís gasoline since they preferred to consume more Extra or Ecopaís gasoline for $ 1.45 USD per gallon, to Súper gasoline for $ 2.98 USD per gallon. This is because, with this second option, refueling would be more expensive. However, for Ecuador, this behavior implied greater global fuel consumption. The variation in the demand for gasoline in the SINDEC and CONDEC scenarios indicates that in the second one, gasoline consumption is approximately 2% higher than in the first one, which can be seen in Fig. 3.
Impact Analysis of Migration from Súper Gasoline to Others
105
Fig. 3. Difference in global gasoline consumption
Comparing fuel dispatch of June - July and October - November 2018, before and after the Executive Order 490, there was an increase, determined with real data, of around 6 million gallons of gasoline, that is approximately 3%, see Table 7. This percentage is similar to 2% calculated with the LEAP software. Table 7. Fuel dispatch - gasoline Fuel
Months: June-July 2018 (Million gal. Gasoline)
Months: October-November 2018 (Million gal. Gasoline)
Difference (Million gal. Gasoline)
Súper
32,7
22,2
10,4
Extra/ Ecopaís
177,7
194,1
16,4
Total
210,4
216,4
6,0
Reference: ARCH.
In December 2018, through the Executive Order 619, the sale price of Extra and Ecopaís gasoline increased, which contracted the demand for fuels in Ecuador. So, it was not possible to verify the increase in fuel consumption due to the switch from Súper gasoline to those types of gasoline with lower octane in 2019 gasoline dispatch. The increase in gasoline consumption in Ecuador due to the issuance of Executive Order 490, implied a greater emission of greenhouse gases into the atmosphere. In this regard, if we calculate the increase carbon dioxide production associated with the increase in gasoline consumption, we get around 159.700 tons of CO2. This increase in carbon dioxide is four times greater than the reduction achieved in Villonaco wind power generation project. This was implemented in the province of Loja - Ecuador in 2013, with the purpose of reducing the local production of greenhouse gases in the generation of electricity [47].
106
C. F. Terneus Páez et al.
4 Conclusion The objective of this research was to analyze the impact of the migration of Súper gasoline to other types of gasoline of lower octane in Ecuador, using the calculation of global gasoline consumption as a methodology and considering two scenarios. The first scenario being SINDEC and the second one CONDEC, before and after the issuance of the Executive Order 490, which reduced the subsidy of the Súper gasoline. With the 90 octane Súper gasoline consumption, before and after the Executive Order, we concluded that this political decision led approximately 118,000 vehicle drivers to replace Súper gasoline with other types of gasoline of a lower price, but also a lower octane rating, which resulted in lower efficiency. The issuance of the Executive Order 490 increased the global consumption of gasoline in Ecuador by around 3%, with the respective increase in the emission of carbon dioxide, which is significantly higher than the reduction obtained by a strategic government project of alternative energy generation. The policies guide citizen behavior to achieve national objectives, however Executive Order 490, which increased the price of Súper gasoline, encouraged the use of other lower octane that are less efficient. This migration brought with it an increase in global gasoline consumption with a corresponding increase in carbon dioxide production, which occurs at a time when the world advocates a reduction in pollution produced by fossil fuels.
References 1. INEC, «Anuario de estadísticas de transporte 2018», October 2019. https://www.ecuadoren cifras.gob.ec/documentos/web-inec/Estadisticas_Economicas/Estadistica%20de%20Tran sporte/2018/2018_ANET_BOLETIN.pdf. Accessed 16 may 2020 2. IIGE: Balance-Energético-Nacional-2018.pdf (2019). [En línea]. Disponible en: https://bit. ly/2WruJKY. accedido 01 May 2020 3. Kreuzer, F.M., Wilmsmeier, G.: Eficiencia energética y movilidad en América Latina y el Caribe, p. 305 (2014) 4. Zaballa Romero, M., Jimenez Rivera, H., Franken, V.: Recomendaciones de política para el sector energético de América Latina y el Caribe en apoyo a la implementación de las NDCs. Quito: OLADE (2017) 5. Vallejo Carpio, C.: Análisis de Intensidad Energética para los Sectores de Uso Final de la Economía Ecuatoriana.», Cuest. Económicas, vol. 27 (2), 2019, [En línea]. Disponible en: https://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/59 6. Guayanlema, V.: Línea Base para investigación en eficiencia energética en el transporte, p. 13 Mar 2015 7. El Telégrafo: Gasolina de mejor calidad se comercializa en el país», El Telégrafo, abr. 02, 2012. https://www.eltelegrafo.com.ec/noticias/economia/4/combustible-menos-contam inante-se-comercializa-en-el-pais. 8. Mena Erazo,P.: Ecuador arranca venta de biocombustible», BBC, ene. 13, 2010. https://www. bbc.com/mundo/ciencia_tecnologia/2010/01/100112_0245_ecuador_gasolina_verde_jrg 9. El Comercio: Gasolina con menos octanaje en Quito y Cuenca», El Comercio, 16 July 2014. https://www.elcomercio.com/actualidad/gasolina-disminuye-octanaje-quito-cue nca-calidad.html. accedido 22 July 2020 10. El Comericio: Octanaje de las gasolinas extra y ecopaís no subirá tras el alza de precio», El Comercio, dic. 26 2018. https://www.elcomercio.com/actualidad/octanaje-gasolinas-extraEcopaís-mantienen.html. accedido 22 July 2020
Impact Analysis of Migration from Súper Gasoline to Others
107
11. Stratas advisors: Global Octane | Stratas Advisors», 17 November 2016. https://stratasadvis ors.com/Insights/111716-Global-Octane. accedido 08 ago 2020) 12. U.S department of energy, «Select the right octane fuel for your vehicle!», Fuel Economy. https://www.fueleconomy.gov/feg/octane.shtml. accedido 08 ago 2020 13. Splitter, D., Pawlowski, A., Wagner, R.: A Historical analysis of the co-evolution of gasoline octane number and spark-ignition engines. Front. Mech. Eng. 1, 16 (2016). https://doi.org/ 10.3389/fmech.2015.00016 14. Demirbas, A., Balubaid, M.A., Basahel, A.M., Ahmad, W., Sheikh, M.H.: Octane rating of gasoline and octane booster additives. Pet. Sci. Technol. 33(11), 1190–1197 (2015). https:// doi.org/10.1080/10916466.2015.1050506 15. EP Petroecuador: Principales daños en la unidad FCC de la Refinería Esmeraldas, EP Petroecuador. https://www.eppetroecuador.ec/?p=7067 16. Kaji, K., Matsushige, S., Kanamaru, M., Takahashi, J., Asano, S.: Development of Knock Sensor». SAE Trans. 95, 315–323 (1986) 17. Beck,C., Stevenson, P., Ziman, P.: The Impact of Gasoline Octane on Fuel Economy in Modern Vehicles, p. 10, October 2016. https://doi.org/10.4271/2006-01-3407 18. Delgado Noboa, D., Dávalos Villacrés, B.: Agenda Nacional de Energía 2016–2040. Ministerio Coordinador de Sectores Estratégicos (2016) 19. Mendoza, M.Á.: Panorama preliminar de los subsidios y los impuestos a las gasolinas y diésel en los países de América Latina. CEPAL, p. 72 (2014) 20. Decreto 490 - 2018 (2018) 21. Decreto Ejecutivo 338: Reglamento de regulación de precios de derivados de petróleo, 22 October 2015, Accedido: abr. 14, 2020. [En línea]. Disponible en: https://bit. ly/2WTfPMv 22. El Comercio: La fusión de tres agencias de control de Hidrocarburos, Minas y de Electricidad se cumplió», El Comercio, 01 July 2020. https://www.elcomercio.com/actualidad/fusion-tresagencias-control-hidrocarburos.html 23. El Universo: Nuevo precio de gasolina súper, en vigor desde hoy», El Universo, ago. 27, 2018. https://www.eluniverso.com/noticias/2018/08/27/nota/6925319/nuevo-precio-gas olina-super-vigor-hoy 24. ALDIA: La gasolina extra y ecopaís suben su precio de 1.48 a 1.85, ALDIA, dic. 18, 2018. https://www.aldia.com.ec/la-gasolina-extra-sube-su-precio-de-1-48-a-1-85/ 25. Coba, G.: Demanda de gasolina Súper cae 38,32% tras la eliminación del subsidio, Primicias.ec, ago. 19 2019. https://www.primicias.ec/noticias/economia/demanda-gasolina-supersubsidio-petroecuador/. 26. Pacheco, M.: El consumo de la gasolina súper en Ecuador bajó al mínimo este 2019 | El Comercio», El Comercio, 11 March 2019. https://www.elcomercio.com/actualidad/consumogasolina-super-minimo-transporte.html. accedido ago. 22 2020) 27. El Comercio, «El consumo de la gasolina súper se hundió en un 41% en enero del 2019», El Comercio, mar. 20, 2019. https://www.elcomercio.com/actualidad/despacho-gasolina-superreduccion-ecuador.html. accedido ago. 22 2020 28. El Comercio: El ahorro por baja de subsidios dejó USD 496 millones en un año | El Comercio», El Comercio, dic. 27, 2019. https://www.elcomercio.com/actualidad/subsidios-gas olina-super-ahorro-gobierno.html. accedido ago 22 2020 29. Shuai, S.-J., Wang, Y., Li, X., Fu, H., Xiao, J.: Impact of Octane Number on Fuel Efficiency of Modern Vehicles. SAE Int., p. 11, October 2013 30. Rashid, A.K., Mansor, M.R.A., Racovitza, A., Chiriac, R.: Combustion characteristics of various octane rating fuels for automotive thermal engines efficiency requirements. 157, 763– 772 (2019). https://doi.org/10.1016/j.egypro.2018.11.242
108
C. F. Terneus Páez et al.
31. Binjuwair,S., Mohamad, T.I., Almaleki, A., Alkudsi, A., Alshunaifi, I.: The effects of research octane number and fuel systems on the performance and emissions of a spark ignition engine: a study on Saudi Arabian RON91 and RON95 with port injection and direct injection systems», 158, 351–360 (2015). https://doi.org/10.1016/j.fuel.2015.05.041 32. Guzmán, A., Cueva, E., Peralvo, A., Revelo, M., Armas, A.: Estudio del rendimiento dinámico de un motor Otto utilizando mezclas de dos tipos de gasolinas “Extra y Súper. p. 13, dic. 2018 33. Inga, V., Vidal, J.: Evaluación del rendimiento de las gasolinas Súper y Ecopaísmediante un ciclo típico de conducción para para taxis del Azuay», Tesis de grado, Universidad del Azuay, Cuenca - Ecuador (2019) 34. Jácome Rivera,E.D., Villegas Mendoza, X.A.: Análisis del impacto de la mezcla de gasolina Súper y Extra en un vehículo Mazda modelo Allegro año 2008 con cilindraje de 1600 CC. en el consumo específico de combustible», Universidad de las Fuerzas Armadas - ESPE, Sangolquí - Ecuador (2020) 35. Di Sbroiavacca, N.: El Modelo LEAP, principales características y su aplicación en el diseño de Políticas Energéticas y Ambientales», Aquasec.org, 28 June 2013. https://aquasec.org/wp-content/uploads/2013/06/Di-Sbroiavacca-Presentacion-La-Ser ena-Long-range-Energy-Alternatives-Planning-LEAP.pdf 36. Shabbir,R., Ahmad, S.S.: Monitoring urban transport air pollution and energy demand in Rawalpindi and Islamabad using leap model. Energy 35(5), 2323–2332 (2010). https://doi. org/10.1016/j.energy.2010.02.025 37. Bitos, C.: Energy Demand Analysis and Energy Saving Potentials in the Greek Road Transport Sector, June 2015. [En línea]. Disponible en: https://bit.ly/2zxALAJ. Accedido: abr. 24, 2020 38. Hong,S., Chung, Y., Kim, J., Chun, D.: Analysis on the level of contribution to the national greenhouse gas reduction target in Korean transportation sector using LEAP model», Renew. Sustain. Energy Rev. 60, 549–559 (2016). https://doi.org/10.1016/j.rser.2015.12.164 39. El Diario.ec: 3.500 vehículos ruedan sin matrícula. El Diario.ec, ene. 24 2018. https://www. eldiario.ec/noticias-manabi-ecuador/461634-3500-vehiculos-ruedan-sin-matricula/ 40. El Comercio: 35 000 vehículos no cumplieron con el proceso de matriculación en el 2019. El Comercio, ene. 20, 2020. https://www.elcomercio.com/actualidad/vehiculos-usuarios-pro ceso-matriculacion-quito.html. accedido 22 July 2020 41. AEADE, Anuario 2018, Mauricio Montenegro / La Caracola Editores. Quito: Editorial Ecuador (2019) 42. SRI: Matriculación Vehícular, SRI. https://www.sri.gob.ec/web/guest/matriculacion-vehicu los#estad%C3%ADsticas. accedido 22 July 2020 43. IEA y ICCT: Fuel economy in major car markets». IEA, Paris, 2019. [En línea]. Disponible en: https://bit.ly/2WuVMoN. Accedido: 04 October 2019 44. Mutenyo, J., Banga, M., Matovu, F., Kimera, D.: Baseline Survey On Ugandas’s National Average Automotive Fuel Economy, p. 88, ago 2015 45. Stockholm Environment Institute: Leap_User_Guide_Spanish.pdf, ago. 2004. https://www. energycommunity.org/documents/Leap_User_Guide_Spanish.pdf. 46. Vertin, K., et al.: Gasoline anti-knock index effects on vehicle net power at high altitude. SAE Int. J. Fuels Lubr. 10(2), 2017–01–0801, March 2017. https://doi.org/10.4271/2017-01-0801 47. CELEC EP: Central Villonaco alcanzó niveles de producción históricos en los meses de julio y agosto de 2018. CELEC. https://www.celec.gob.ec/78-quienes-somos/511-central-villonacoalcanzo-niveles-de-produccion-historicos-en-los-meses-de-julio-y-agosto-de-2018.html.
Experimentation of Adaptive Strategies in High-Speed Machining (HSM) for Rough Milling Process Using Prodax Aluminum Francisco Infante Castillo and Borys Culqui Culqui(B) Universidad de las Fuerzas Armadas - ESPE, Sangolquí, Ecuador {fjinfante,bhculqui}@espe.edu.ec
Abstract. In this project, the rough milling mechanized process was carried out without cutting fluid to analyze their performance. It compared conventional and adaptive strategies in high-speed machining (HSM). The material tested was a 7075-76 aluminum alloy (Prodax aluminum) because of the excellent mechanical properties and the high scope to use in molds and dies. Machining time and tool temperature were measured by varying cutting parameters, such as cutting speed (Vc) and feed per tooth (fz). The conventional strategy was performed with a constant cutting depth of 2,188 mm and a cutting width of 64% in diameter. Similarly, the adaptive strategy was done with a constant cutting depth of 14 mm and a cutting width of 10% in diameter. The milling tool was an HSS straight with 4 flutes and with 25 mm in diameter. A Taguchi experimental methodology L32 (2 ˆ 1 4 ˆ 2) was applied to combine parameters and levels; therefore, 16 tests were carried out for each strategy. Furthermore, an ANOVA statistical analysis determined that the adaptive strategy has the lowest machining time in comparison with the conventional strategy. As a maximum success, a machining time reduction of 82.3% was reached. Keywords: HSM high speed machining · Adaptive milling · Conventional milling
1 Introduction Aluminum is one of the most common elements on earth after oxygen and silicon, this metal has high machinability, besides having a high electrical and thermal conductivity; at low temperatures, it does not present a transition from ductile to fragile. An important physical property of aluminum is that it has a resistance to oxidation and corrosion in addition to having a non-magnetic behavior. Because of its low melting point, aluminum does not work well at high temperatures [1]. In the aluminum alloy A7075-T6, the main alloy which provides an increase in hardness and strength is the zinc. The alloy has a T6 heat treatment that is characterized by to solubilize and age the alloy to increase its resistance. The alloy is hot rolled, subjected to a cold stretching operation for maximum stress relief. Prodax aluminum has excellent machinability, low weight, high thermal conductivity, high strength, good © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 109–122, 2021. https://doi.org/10.1007/978-3-030-72212-8_9
110
F. Infante Castillo and B. Culqui Culqui
stability, and corrosion resistance. Prodax aluminum is used in the mold and die industry. [2]. In machining processes, cutting parameters can be categorized as independent and dependent. By varying independent parameters such as cutting speed and feed rate, the most influential one can be determined in the high-speed milling process [3]. Researches developed by Dr. Solomon on high speeds machining (HSM) determined that, in non-ferrous materials, the temperature increases as the cutting speed increases up to approximately 200 m/min after that it decreases. In a study of [4], the author applies an adaptive strategy to perform complex parts, and he uses this strategy to rough and finish the blades of a turbine of a reaction engine using a carbide milling cutter with a speed of 8000 rpm. As for the trochoidal machining [4] performed by Abram Pelta proposes a saving of machining time by removing the trimmed tool path, to replace it with linear movements so that rapid interpolations are made outside the workpiece. The study analyzed the cutting force under three conditions: cutting speeds (25 to 50 m/min), feed per tooth (0.1, 0.2, 0.3 mm/tooth), and cutting depth (2, 5, 8 mm). The trials generated a 40% increase in the Fy component under the following specific conditions: cutting speed of 50 m/min, feed per tooth of 0.3 mm/tooth, and cutting depth of 5 mm, but was reached shorter machining times with a good finish surface. For this reason, statistical studies have developed with regression models and parameter optimization specified by the Taguchi and ANOVA methodology. In the present project, the reduction of machining time is between the conventional machining strategy and the adaptive machining strategy were compared. The trials were performed with high-speed machining (HSM) parameters. The parameters selected were cutting speed from 400 m/min to 700 m/min, feed rate from 0.05 mm/tooth to 0.2 mm/tooth. Besides, trials with similar power consumption were compared to determine reduction times.
2 Experimental Methods 2.1 Description of the Experiment The testing machine was a vertical machining center Fadal VMC 3016 with a Fanuc control unit model 18 iMB, main specs are 4-axis, maximum spindle speed of 10,000 rpm, 15 HP motor, and maximum rapid traverse speed of 22 m/min. The processes were performed with a combination of high-speed machining parameters, according to the curve proposed by Dr. Salomon. Besides, the tests were carried out without refrigerant fluid. The cutting parameters were the following: cutting speeds: 400, 500, 600 and 700 m/min; feed rates: 0.05, 0.1, 0.15 and 0.2 mm/tooth. In the conventional strategy, the constant of depth of 2.188 mm was used with a cutting width of 64% of the tool diameter. In the adaptive strategy, a constant cutting depth of 14 mm was used with a cutting width at 10% of the tool diameter. The configuration of the machining tests is shown in Fig. 1
Experimentation of Adaptive Strategies in High-Speed Machining (HSM)
111
Fig. 1. Machining configuration
2.2 Material Specification A 7075-T6 aluminum with a density of 2,810 g/cm3 with a length of 70 mm, width of 70 mm and a height of 20 mm was used as work piece. The chemical composition is shown in Table 1. Table 1. Al 7075-T6 chemical composition Chemical composition (% weight) Al
Zn
87–91.4
5.1–6.1 1.2–2 0.18–0.28 0.4 2.1–2.9 0.5 0.2
Cu
Cr
Si
Mg
Fe
Ti
2.3 Cutting Tool The tool used in the test is a 25 mm diameter milling cutter with four flutes. The specifications for this tool are the following: • • • •
Manufacturer: Somta Material: High-speed machining (HSS) Cutting depth: 1.25 mm Cutting Feed rate: 4800 mm/min.
2.4 Taguchi Methodology The Taguchi method has been used widely in quality and engineering design. Besides. The Taguchi method develops the procedures applying orthogonal arrays (OA) between
112
F. Infante Castillo and B. Culqui Culqui
parameters and levels in the experiment to obtain the best model with a reduction in the number of tests and minimizing the time and cost of experimentation [6]. It is important to identify the factors that affect the machining process, so factors must be varied within the design of the experiment to assess which factors have more impact on the process. For this experiment, the parameters and levels are shown in Table 2. Table 2. Parameters and levels of the experiment Item Parameter
Units
Level 1
2
3
4
A
Strategy
–
1 2 – Conventional Adaptive Machining Machining Strategy Strategy (AMS) (CMS)
–
B
Cutting Speed
[m/min]
400
500
600
700
C
Feed per tooth [mm/tooth] 0.05
0.10
0.15 0.20
In the methodology, the minimum number of experiments must be greater than or equal to DOF [7]. Total Degrees of Freedom (DOF) = (ni − 1) ∗ nf
(1)
Where: ni = number of parameters. nf = number of levels. According to Table 2, in this experiment ni = 3 and nf = 4, then: DOF = (4 − 1) ∗ 3 = 9 Then, the minimum number of experiments is calculated as follow: Minimum number of experiments = DOF + 1 = 9 + 1. Minimum number of experiments = 10. In this project, total combination between parameters and levels are 32. Therefore, all test was done. 2.5 Power Analysis The conventional machining strategy (CMS) uses cutting parameters opposed in comparison with the adaptive machining strategy (AMS). The conventional machining strategy
Experimentation of Adaptive Strategies in High-Speed Machining (HSM)
113
focuses on great cutting width (Ae) with a low cutting depth (Ap); meanwhile, the adaptive machining strategy focuses on low cutting width (Ae) with great cutting depth (Ap). Therefore, in this experimentation, the way to compare tests with CMS and AMS is the cutting power required for the cutting process. Tests with the same cutting power were compared. The cutting power for a cutting speed of 400 m/min and cutting depth of 1 to 5 mm are calculated as indicated in Table 3. Table 3. Cutting Power in conventional and adaptive machining strategies CMS (Ap [mm]) PC [kW]
AMS (Ap [mm])
Test
1
2
3
4
5
Test
14
A1B1C1
0.19
0.38
0.57
0.76
0.95
A2B1C1
0.41
A1B1C2
0.38
0.76
1.14
1.52
1.90
A2B1C2
0.83
A1B1C3
0.57
1.14
1.71
2.28
2.85
A2B1C3
1.25
A1B1C4
0.76
1.52
2.28
3.04
3.80
A2B1C4
1.66
The cutting depth for the same cutting power in conventional machining strategy was estimated by interpolation, obtaining the same cutting area between both strategies, as indicated in Table 4. From a cut width of 10% (2.5 mm) with a pass depth of 14 mm in adaptive machining, it is possible to compare with conventional machining with a cutting depth of 2,188 mm and a cutting width of 64% (16 mm). Similar facts occur with other combinations of parameters and levels. Table 4. Cutting area in conventional and adaptive machining Parameter
CMS AMS A1B1C1 A2B1C1
Cutting width Ae [mm] 16
2.5
Cutting depth Ap [mm] 2.188
14
Cutting area [mm2 ]
35
35
Cutting power [kW]
0.41
0.41
2.6 Comparison of Conventional Tool Path At a previous phase, the machining time for different tool paths in the conventional machining process were simulated in CAM software, and results indicated in Table 5. This task was done to determine the tool path with the minimum machining time in CMS.
114
F. Infante Castillo and B. Culqui Culqui
The follow periphery tool path was selected because of the shortest simulation time, in comparison to the other tool paths. This tool path makes 2 radial passes with a cutting width of 64% and a depth of 2.188 mm. Table 5. Time in conventional and adaptive machining Tool path
Parameters
Machining time [s]
Vc [m/min]
fz [mm/tooth]
Follow part
400
0.05
504
Follow Periphery
400
0.05
318
Profile
400
0.05
469
Trochoidal
400
0.05
726
Zig
400
0.05
1114
Zigzag
400
0.05
544
Zig with contour
400
0.05
953
2.7 Machining Strategies and Tool Path The tool path in conventional machining strategy (CMS) was follow periphery and the tool path in the adaptive machining strategy (AMS) was trochoidal. The machining tool paths were obtained by a CAD-CAM software as is shown in Fig. 2.
Fig. 2. a) Conventional tool path, b) Adaptive tool path
3 Results and Analysis 3.1 Tabulation of Experimental Data Machining time data were obtained through the VMC 3016 Fadal equipment. Simulation times were obtained by CAD-CAM software and real times were measured on the
Experimentation of Adaptive Strategies in High-Speed Machining (HSM)
115
Table 6. Experimental data Test
A B C Machining time [s] Tool temperature 1 CMS Vc [m/min] fz [mm/tooth] Simulation Real [ºC] 2 AMS
A1B1C1 1
400
0.05
A1B1C2 1
400
0.1
A1B1C3 1
400
0.15
A1B1C4 1
400
0.2
A1B2C1 1
500
A1B2C2 1
500
A1B2C3 1
318
390
72
275
285
68
237
240
47
218
224
44
0.05
343
358
49
0.1
252
282
55
500
0.15
222
227
49
A1B2C4 1
500
0.2
207
211
51
A1B3C1 1
600
0.05
313
325
48
A1B3C2 1
600
0.1
237
245
54
A1B3C3 1
600
0.15
212
213
44
A1B3C4 1
600
0.2
200
199
41
A1B4C1 1
700
0.05
291
305
43
A1B4C2 1
500
0.1
227
234
41
A1B4C3 1
700
0.15
205
210
39
A1B4C4 1
700
0.2
194
198
42
A2B1C1 2
400
0.05
198
223
43
A2B1C2 2
400
0.1
99
104
44
A2B1C3 2
400
0.15
66
70
43
A2B1C4 2
400
0.2
50
53
45
A2B2C1 2
500
0.05
158
165
48
A2B2C2 2
500
0.1
79
85
50
A2B2C3 2
500
0.15
53
57
51
A2B2C4 2
500
0.2
40
45
48
A2B3C1 2
600
0.05
132
135
47
A2B3C2 2
600
0.1
66
71
51
A2B3C3 2
600
0.15
44
49
48
A2B3C4 2
600
0.2
33
39
38
A2B4C1 2
700
0.05
113
117
39
A2B4C2 2
700
0.1
57
62
35
A2B4C3 2
700
0.15
38
44
36 (continued)
116
F. Infante Castillo and B. Culqui Culqui Table 6. (continued)
Test
A B C Machining time [s] Tool temperature 1 CMS Vc [m/min] fz [mm/tooth] Simulation Real [ºC] 2 AMS
A2B4C4 2
700
0.2
28
35
38
experimental tests. Similarly, temperatures were measured in tool flutes immediately after the machining process, data are shown in Table 6. 3.2 Machining Time Analysis When comparing data between the conventional machining strategy (CMS) versus the adaptive machining strategy (AMS) with similar cutting parameters, as indicated in Table 7, the real machining time for AMS is shorter than CMS. The curves have a decreasing behavior in the form of a saw-tooth, as is shown in Fig. 3. Table 7. Machining time and reduction percentage Test
Machining time CMS [s]
Machining time AMS [s]
% Reduction
A1B1C1 - A2B1C1
390
223
42.8
A1B1C2 - A2B1C2
285
104
63.5
A1B1C3 - A2B1C3
240
70
70.8
A1B1C4 - A2B1C4
224
53
76.3
A1B2C1 - A2B2C1
358
165
53.9
A1B2C2 - A2B2C2
282
85
69.9
A1B2C3 - A2B2C3
227
57
74.9
A1B2C4 - A2B2C4
211
45
78.7
A1B3C1 - A2B3C1
325
135
58.5
A1B3C2 - A2B3C2
245
71
71.0
A1B3C3 - A2B3C3
213
49
77.0
A1B3C4 - A2B3C4
199
39
80.4
A1B4C1 - A2B4C1
305
117
61.6
A1B4C2 - A2B4C2
234
62
73.5
A1B4C3 - A2B4C3
210
44
79.0
A1B4C4 - A2B4C4
198
35
82.3
In conventional machining strategy (CMS), test A1B1C1 presents a maximum machining time of 390 s and test A1B4C4 shows a minimum machining time of 198 s.
Experimentation of Adaptive Strategies in High-Speed Machining (HSM)
Machining time vs Cutting Speed
500
Time [s]
400
390
358 240
223
224
200 100 0
104
70
53
213 211
165 85
305
245
227
199
234 210 198
135 71
57
AdapƟve Strategy ConvenƟonal Strategy
325
282
285
300
117
117 49
39
62
35 44 400 400 400 400 500 500 500 500 600 600 600 600 700 700 700 700 45
Cutting Speed [m/min] Fig. 3. Machining time comparison between CMS and AMS
Similarly, in the adaptive machining strategy (AMS), test A2B1C1 presents a maximum time of 223 in s and test A2B4C4 shows a minimum time of 35 s. At the same cutting speed (B1 = 400 m/min), the percentage of machining time re-duction increases while the cutting feed rate increases, it increases from 42.8% to 76.3%. In the same way (B2 = 500 m/min), the percentage increases from 53.9% up to 78.7%, and so on. Therefore, tests with the lowest cutting parameters show the lowest time reduction percentage of 42.8%. On the opposite side, tests with the highest cutting parameters show the highest time reduction percentage of 82.3%. 3.3 Temperature Analysis Comparison between conventional machining strategy (CMS) versus adaptive machining strategy (AMS) respect to temperature, as indicated in Fig. 4. At these machining processes the cutting tool was cooled by convection, air was the cooling medium and no forced fluid was used. The temperature in conventional machining strategy (CMS) has an irregular decreasing behavior at a cutting speed of 400 m/min, it decreases strongly from 72 °C to 44 °C, but from 500 m/min, it tends to decrease more slowly. Fig. 4 shows a maximum temperature of 57 °C at 500 m/min and a minimum of 39 °C at 700 m/min. The temperature in adaptive machining strategy (AMS) shows more stable behavior at all range of cutting speed from 400 m/min to 700 m/min. Fig. 4 shows a maximum temperature of 51 °C at 500 m/min and a minimum of 39 °C at 700 m/min. Finally, the adaptive machining strategy shows more stable behavior and temperatures slightly lower than the conventional strategy.
118
F. Infante Castillo and B. Culqui Culqui
Temperature vs Cutting Speed
80
AdapƟve Strategy ConvenƟonal Strategy
Temperature [°C]
72 70
68
60
57
50 40
47 43
44
43
45 44
55 51 51 48 48
50
49 48
47
54 48 47 51
43 44
45 40
45 41
30
39
39
42 36
400 400 400 400 500 500 500 500 600 600 600 600 700 700 700 700
Cutting Speed [m/min] Fig. 4. Temperature comparison between CMS and AMS
3.4 Taguchi Analysis The operating time and temperature of the cutting tool were analyzed using the Taguchi experimental method to determine the influence of each parameter (A = strategy, B = Vc, C = fz) in the machining process as indicated Table 8, Fig. 5 and Fig. 6. Respect to machining time, the least values are reached with the following conditions: • Strategy (A2 = AMS) with 84.63 [s] • Cutting speed (B4 = 700 m/min) with 150.63 [s], and • Cutting feed per tooth (C4 = 0.20 mm/tooth) with 125.50 [s]. Respect to temperature, the least values are reached with the following conditions: • Strategy (A2 = AMS) with 44 [°C] • Cutting speed (B4 = 700 m/min) with 39.13 [°C], and • Cutting feed per tooth (C4 = 0.20 mm/tooth) with 41.75 [°C]. The parameters: strategy and cutting feed per tooth (in this order) have a great influence on machining time due to their large slopes shown in Table 5. On the other hand, the cutting speed has a great influence on the tool temperature as shown in Table 6, especially at high cutting speed this is in relation with Dr. Solomon researches that indicate temperature reduces when cutting speed increases. 3.5 ANOVA Method The ANOVA statistical analysis, known as the analysis of variance, is used to determine the variability of the data, obtaining the level of confidence of the experimental data. Analysis of variance establishes whether the population means are the same or different and determines the interrelationships between all the factors using in the test design, in addition to calculating the degrees of freedom, the sum of squares, F test, variance [7].
Experimentation of Adaptive Strategies in High-Speed Machining (HSM)
119
Table 8. Response for the means of operation time. Level
Control factors Machining time [s]
Temperature [°C]
A
B
C
A
B
C
1
259.13
198.63
252.25
49.19
50.75
49.88
2
84.63
178.75
171.00
44
50.13
50.13
3
159.50
138.75
46.38
44.63
4
150.63
125.50
39.13
41.75
Delta
174.50
48
126.75
5.19
11.63
8.38
Classification
1
3
2
3
1
2
Fig. 5. Main effects with respect to operation time
Analysis of the results of ANOVA concerning machining time is shown in Table 9. It was performed, with a significance level of 5% and a confidence level of 95%, the control of the ANOVA methodology was done by comparing the values of F and P. According to Table 9 was defined whether there is a statistical difference in the results. The statistical F with a 95% confidence for strategy (A) is: FA (0.05: 1: 24) = 1149.49, for the cutting speed (B) is: FA (0.05: 3: 24) = 21.20, for the feed per tooth (C), is: FA (0.05: 3: 34) = 121.10. These values are greater than the values tabulated in the Fisher (F) tables of 4.26 for strategy (A), for the cutting and feed rate per tooth (B and C) an F value of 3,009. Then, the null hypothesis where the population means are equal is rejected and we accept the alternative hypothesis where the means are different. Therefore, there is not a significant statistical difference, in the treatments, it is less than the maximum error that we are allowing P < 0.05. The P values obtained in Table 9 mean that the results are reliable complying statistically. The same analysis for temperature.
120
F. Infante Castillo and B. Culqui Culqui
Fig. 6. Main effects with respect to temperature
Table 9. Time and temperature variance analysis Time Source
DOF
Temperature
S
V
F
P
1198.35
0.0
S 215.28
V
F
P
215.28
6.08
0.021
A
1
243602
243602
B
3
10940
3646.75
17.94
0.0
684.6
228.20
6.45
0.002
C
3
77671
25890.16
127.36
0.0
404.6
134.86
3.81
0.023
Error
24
4879
203
–
–
849.3
35.39
–
–
Total
31
337092
–
–
–
2153.7
–
–
–
3.6 Regression Analysis for Time and Temperature Regression analysis is used for modeling and analysis that exists between one or more independent variables concerning the dependent variable. For this study, the independent variables are the machining strategy, the cutting speed (Vc) and the feed per tooth (fz). The prediction equations were obtained from the linear regression analysis as indicated in Eqs. (2) and (3). The regression equation of time and temperature will serve to relate the response parameter, concerning the experimental model parameters [7]. Time [s] = 626.5 − 0.1089 Vc − 825 fz − 174.5 Strategy
(2)
Temperature ◦ C = 79.36 − 5.38 Estrategy − 0.03350 Vc − 42 Fz
(3)
The Taguchi methodology ends with the representation of the values with a statistical parameter within a range, which is likely to fall into a level of confidence. Where CI is the confidence interval, Ve is the variance of the error found in Table 9. Ve = 35.458, Ve = 35.39 respectively, Neff is the effective number of repetitions, F is the 95% reliability factor, α It is the significance, faith is the degrees of freedom from
Experimentation of Adaptive Strategies in High-Speed Machining (HSM)
121
error, Tdof is the total factor of degrees of freedom, R is the number of repetitions. 1 N 1 ; Neff = (4) CI = Fα,1,fe Ve + nef R 1 + Tdof F0.05,1,24 = 4.226, Ve = 35.458, Ve = 35.39R = 3, Ttof = 7 Topt − CITemp < Texp < Topt + CITemp
(5)
[35 − 111.652] < Texp < [35 + 111.652] The optimum average temperature with a 95% confidence interval is: Topt − CITemp < Tempexp < Topt + CITemp
(6)
[36 − 9.349] < Tempexp < [36 + 9.349] Using Eq. 4 and 5, the confidence interval for time is CI = ±111.652 and for the temperature, CI = ±9.349, the values obtained in the experimental study were kept within the limits of the confidence interval for both time and the temperature. 3.7 Confirmation Experiments From the optimal parameters found in Table 10, three confirmation tests were performed according to the Taguchi methodology, the results obtained in the confirmatory tests, so that they are within the range of the confidence interval obtained in Eq. 5 and 6, thus corroborating the validity of the study regarding the machining time and temperature of the cutting tool. Table 10. Predictive values and confirmation of regression results. Level
Exp Pred
Error (%) Level
Exp Pred
Error (%)
Time [s]
Temperature [°C]
A1 B4 C4 (optimun) 35 36.27 3.50 A2 B2 C1 (Random) 358 356.3 0.477
A2 B4 C4 (optimun) 36 A1 B3 C3 (Random) 57
36.75 2.04
A2 B3 C1 (Random) 325 345.4 5.91
A2 B1 C4 (Random) 45
46.80 3.84
55.13 3.39
4 Conclusions • This research presents an optimization of the milling machining process performed on a Fadal VMC 3016 CNC machine tool. Machining time and tool temperature were
122
•
•
•
•
F. Infante Castillo and B. Culqui Culqui
analyzed against the following parameters: strategy, cutting speed, and cutting feed per tooth. All tests were performed with a Somta HSS milling tool cutter. In this project using a conventional machining strategy, the cutting depth was improved to 2.188 mm, increasing by 75% of the manufacturer’s recommendations. Similarly, the cutting feed speed was increased to 7130.12 mm/min, it is 32.67%. Another improvement was that the adaptive machining strategy did not show burrs on the workpiece or burns on the cutting tool. The strategy is a parameter that influences the process to obtain smaller times of operation. The adaptive machining strategy (AMS) shows better behavior compared to the conventional machining strategy (CMS). The adaptive machining strategy reduces the machining time from 42.8% (test A2B1C1) with the minimum machined parameters of cutting speed of 400 m/min and a cutting feed per tooth of 0.05 mm/tooth, up to a maximum reduction time of 82.3% (test A2B4C4) with maximum parameters of cutting speed of 700 m/min and a cutting feed per tooth of 0.20 mm/tooth. The parameters: strategy and cutting feed per tooth (in this order) have a great influence on machining time due to their large slopes shown in Table 5. On the other hand, the cutting speed has a great influence on the tool temperature as shown in Table 6, especially at high cutting speed this is in relation with Dr. Solomon researches that indicate temperature reduces when cutting speed increases From the regression Eq. (2) an error of 3.50% was obtained in parameter A2B4C4 besides in a random parameter A1B2C1 an error of 0.477% based on the experimental time and the prediction time of the equation. For the temperature using the combination of the parameters A2B4C4 an error of 2.04% and in a random value A2B1C4 an error 3.84% based on the mathematical Eq. (3) being both equations a good mathematical approximation of the design of experiments against real. Through the development of experiments, using the adaptive strategy (AMS) the type of chip is regularly elongated, with better finish and without excess material compared to the conventional strategy (CMS), the type of chip is irregular with burn in the tool cutting and excess material on the workpiece.
References 1. 2. 3. 4.
Askeland, D.R.: Ciencia e Ingeniería de lo materiales, pp. 68–73. Thomson, México (2004) Bohman, I.: Catálogo general de productos de Ivan Mohman, III edición (2018) Kalpakjian, S.: Manufactura, Ingeniería, tecnología, pp. 150–190. Prentice Hall, México (2008) Ashwin Polishetty, M.G.: A preliminary assessment of machinability of titanium alloy TI 6AL 4V during thin wall machining using trochoidal milling. Procedia Eng. 97, 357–364 (2014) 5. Samtas, G.: Optimisation of cutting parameters during the face milling of AA5083-H111 with coated and uncoated inserts using Taguchi method, Research Gate, pp. 120–134 (2015) 6. Roy, R.J.: A primer on the method Taguchi. SME, Society of Manufacturing Engineers, United States of America (2010) 7. Sunilkumar, S.: Optimization of Process parameters in milling operation by Taguchi’s technique using regression analysis. Int. J. Sci. Technol. Eng. (2016)
Optimization of the Setup of Workpiece Zero Point in a Numerical Control Machine with an Artificial Vision System Andrea Robalino Pinango and Borys Culqui Culqui(B) Universidad de las Fuerzas Armadas - ESPE, Sangolquí, Ecuador {ajpinango,bhculqui}@espe.edu.ec
Abstract. Nowadays, Computerized numerical control (CNC) machines are widely used to manufacture products. However, some tasks of the machining process are not completely automated; moreover, older machines does not have specific communication ports. This project proposes to optimize the setup of workpiece zero point, a very important task for referencing the workpiece. Artificial vision technology was used to automate micro-movements required to find the coordinates (x, y, z) of the workpiece origin (W) with respect to the absolute machine origin (M). The implemented device captures the image of the workpiece, and the interface (GUI) performs the computerized treatment of the image to obtain the coordinates of the workpiece zero point (W). Finally, a G code program with the coordinates is sent to the machine for registration at the offset setting function. The project is developed in a FADAL VMC 3016 machining center with a FANUC 18i-M control, model 2006, which was chosen for the case study. The results were a reduction of average setup time from 8.18 to 1.89 min and average absolute error on the accuracy of 0.495 mm. Keywords: Image recognition · Workpiece zero-point · Numerical control machine · Calibration of camera
1 Introduction The machining process in a CNC machine has the following general tasks: 1) Setup of the workpiece zero point, 2) loading and unloading of the workpiece, 3) machining operations, and 4) milling tool change. Despite the current level of technology, in slightly old machines like the FADAL VMC3016 (2006), the machining operations and milling tool change are automatically performed while the other tasks must be performed manually, especially in the setup of workpiece zero point [1]. Nowadays, certain companies are already working on artificial vision issues, such as DATRON company, which has developed a CNC machine that uses this technology to perform automatic measurements on a part [2].
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 123–136, 2021. https://doi.org/10.1007/978-3-030-72212-8_10
124
A. Robalino Pinango and B. Culqui Culqui
In the case of the research carried out by Klancnik and Senveter, reference is made to a system for the optical determination of the origin of the part in the CNC machine, being highly sophisticated and very expensive systems [3]. Therefore, the main objective was to optimize the setup of workpiece zero point by using an artificial vision technology system to automate micro-movements required to find the coordinates (x, y, z) of the workpiece origin (W) with respect to the absolute machine origin (M). An interface (GUI) was designed to interact with the CNC machine and send the data found to the offset setting function of FANUC 18i-M control.
2 Methodology The project implements an artificial vision system with hardware and software to control the setup of workpiece zero point in the machine. In addition, the calibration of the system and the tests with their results obtained are explained. The basic scheme of the devices is shown in Fig. 1.
Fig. 1. Project scheme.
2.1 Hardware System The machining process was performed on a FADAL VMC 3016 machining center with the FANUC 18i-M control that is in the manufacturing processes laboratory at the Armed Forces University – ESPE. In order to introduce the artificial vision system, a Logitech camera is used to determine the coordinates of the workpiece zero point in the X, Y axis. Z axis workpiece zero point is determined with the help of a distance sensor (See Fig. 2).
Optimization of the Setup of Workpiece Zero Point
Coordinate
detection
system in X-Y-axis
Coordinate
125
detection
system in Z-axis
Workpiece zero point
Fig. 2. Parts of the system in the FADAL VMC 3016 machining center.
2.2 Control Software For the development of the system, software is used to provide a controlled programming environment; it can be developed, installed, and run on Windows-based operating systems. It encompasses a C# programming language integrated development environment that is complemented by a graphical user interface (GUI) (See Fig. 3). 2.3 Communication Communication with the FANUC 18i-M control is a vital part of the system. For that, the FOCAS library (FANUC Open CNC API Specification) is used to interact with the CNC machine through ethernet communication. This library provides all necessary functions for the development of applications with the Windows operating system. 2.4 Artificial Vision Artificial vision was used to capture, make the computerized treatment of the image, and analyze the information required from the image, all in real-time [4]. Image Resolution The image resolution is required to select sensors. Equation (1) shows the calculus: 450 = 900 pixels (1) Image Resolution = 2 1 The 1024 × 768 sensor will work even though 900 pixels is larger than the smallest dimension which is 768.
126
A. Robalino Pinango and B. Culqui Culqui
Fig. 3. Diagram of the workpiece zero-point setup process.
Regarding obtaining the working distance, the location of the camera depends on the distance from the focus to the workpiece. Canny Algorithm To obtain edge enhancement, the canny algorithm is used to eliminate unnecessary parts of the background as shown in Fig. 4(a). With the help of the OpenCV library, the image of Fig. 4(b) is obtained.
Optimization of the Setup of Workpiece Zero Point
127
Fig. 4. Image Segmentation. a) Original image b) Image with the canny algorithm.
With the processed image, the limits of the workpiece are selected and the coordinates are obtained (See Fig. 5).
Fig. 5. Image processed with the coordinates of the possible zeros-part.
2.5 Graphical User Interface (GUI) The graphical user interface helps to control the process and visualize the image recognition. As the system works in Windows, it was implemented with the contribution of a compatible interface within the visual studio software developed in programming language C#. The general operations of the interface designed are supervision, monitoring, process control, and communication (See Fig. 6). The process begins establishing communication of the interface with the machine control unit, then the type of part is selected, either square or round. The image is captured, and digitization carries out the internal computerized treatment of the image to obtain the coordinates (X, Y, Z) of the corners or control points of the workpiece. The option “post process” in the user interface generates a G code program with the
128
A. Robalino Pinango and B. Culqui Culqui
A
B
D
C
Fig. 6. Calibration on the machining table with a 2D template.
coordinates of the workpiece zero point; finally, “setting work” button sends data to the machine for its registration. 2.6 Experimentation In this project, accuracy and setup time were analyzed to determine the grade of optimization in the setup of the workpiece zero point between automatic process with the artificial vision system versus manual process with the machine probe. Experiments for Accuracy The experiments for accuracy evaluated the proximity of the measurements to the reference values by evaluation of absolute errors. The tests were performed on two types of workpieces: square (A) and rectangular B), where four control points of the workpiece were located. These points designated as A, B, C, and D were located at the corners of the workpiece. Point A is placed at the left rear corner; the next points are located sequentially following the clockwise direction as indicated in Fig. 6. Then, the coordinates (X, Y, Z) for each control point were measured. The tests are listed as follows: • • • •
Manual Test A (square workpiece) Manual Test B (rectangular black color workpiece) Automatic Test A (square workpiece) Automatic Test B (rectangular black color workpiece).
Seven samples in Automatic Test A and five samples of Automatic Test B were taken on each of the control points (A, B, C, and D). Therefore, because each point has coordinates X and Y, 56 samples for Test A and 40 samples for Test B were performed.
Optimization of the Setup of Workpiece Zero Point
129
Experiment for Setup Time The experiment for setup time determined the percentage of reduction in setup time. The total time of the machining process of 15.45 min to manufacture a small piece is taken as a basis data, where the setup time is 8.18 min. The tests are listed as follows: • Manual Test • Automatic Test. For the manual test, four samples were taken, and for the automatic test, eleven samples were taken. 2.7 Calibration The camera calibration process is very important in the artificial vision system, which consists of determining the position and orientation of the camera with respect to a preestablished reference system. The coordinates of the reference point use a scaling factor determined by the base of the grid as is shown in Fig. 7.
Fig. 7. Calibration on the machining table with a 2D template.
Once the grid has been placed for the most precise and exact choice for the machine’s reference point, in this case M, it is indicated with a marker to avoid measurement errors of the point coordinates in the template because the same point is assumed for later use in the GUI. The found point is calibrated with various distances from camera to the part. However, to obtain a conversion factor, it is based on a distance at Z = −40 mm to avoid image distortion. In this way, the value of the Z axis will start at 62.09 as an initial reference to carry out the whole process. In the case of X, Y, the workpiece has to be centered to capture the image with the best conditions (See Fig. 5).
130
A. Robalino Pinango and B. Culqui Culqui
The conversion factor is computed with Eq. (2) as follows: conversion factor =
200 = 0.69444 288
(2)
Whereas, 200 is the value of one side of the workpiece, which is equivalent to the 288 pixels on the image. The accuracy in the Y-axis could be improved with Eq. (3): y = −4X 10−9 x4 + 9X 10−7 x3 − 0.0005x2 + 0.2546x − 15.154
(3)
3 Results The operation of the artificial vision system implemented in the FADAL VMC 3016 vertical machining center was evaluated with accuracy tests. The results for the tests are detailed in the following paragraphs. Results of Accuracy Manual Test A manual process was applied to obtain the values of the coordinates of the workpiece zero point at A, B, C, and D control points. These were the reference values. The values of the coordinates for Manual Test A (square workpiece) obtained from samples on each control point are shown in Table 1. Table 1. Coordinates for manual test A Point X coordinate Y coordinate A
−337,272
88,218
B
−186,589
86,939
C
−187,091
−63,461
D
−337,731
−62,562
In the same way, the values of the coordinates for Manual Test B (rectangular black color workpiece) obtained from samples are shown in Table 2. Automatic Test The artificial vision system developed was used to obtain the average values of the coordinates of the workpiece zero point at A, B, C, and D control points. The average values of the coordinates for the Automatic Test A (square workpiece) obtained from 7 samples are shown in Table 3. The average values of the coordinates for Automatic Test B (rectangular black color workpiece) obtained from 5 samples are shown in Table 4.
Optimization of the Setup of Workpiece Zero Point
131
Table 2. Coordinates for manual test B Point X coordinate Y coordinate A
−314,620
B
−209,690
88,220
C
−211,040
−64,680
D
−315,570
−64,720
89,310
Table 3. Average coordinates for automatic test A Control point X [mm]
Y [mm]
A
−337,469
88,087
B
−186,308
86,606
C
−187,165 −62,640
D
−337,953 −62,307
Table 4. Average coordinates for automatic test B Control point X [mm]
Y [mm]
A
−314,459
89,536
B
−209,890
88,336
C
−211,060 −64,722
D
−315,341 −64,569
Table 5. Manual and automatic test (A) coordinates on the Z-axis Number
Manual test A Z [mm]
Automatic test A Z [mm]
1
−323,472
−323.782
The values of the coordinates in the Z-axis obtained manual and automatic processes are given in Table 5. This measurement was carried out in test A. Results of Setup Time Manual Test The setup times for workpiece zero point in the manual tests are shown in Table 6. Automatic Test The setup times workpiece zero point for automatic tests are shown in Table 7.
132
A. Robalino Pinango and B. Culqui Culqui Table 6. Setup time for manual test Number Time [min] 1
9,450
2
8,450
3
7,300
4
7,500
Average 8,180
Table 7. Setup time for automatic test. Number Time [min] 1
2,100
2
1,580
3
2,120
4
2,110
5
2,010
6
1,570
7
1,570
8
1,530
9
2,010
10
2,030
11
2,150
3.1 Analysis of Results Analysis of Accuracy The analysis of accuracy computes absolute errors between the coordinates measured in automatic tests (A and B) respect to coordinates measured in manual tests (A and B) which are taken as reference values. Therefore, this analysis was done in X, Y, and Z axes, independently, and comparisons are shown as follows. The absolute errors in X-axis computed for tests (A) and tests (B) are shown in Table 8. The following Fig. 8 shows absolute errors obtained from the averages of the control points both in Test A and Test B. For X-axis, the average absolute errors of +0.194 mm for Test A and +0.153 mm for Test B were obtained. This represents a reduction of 21.13% in Test B vs Test A. The absolute errors in Y-axis computed for tests (A) and tests (B) are shown in Table 9.
Optimization of the Setup of Workpiece Zero Point
133
Table 8. Absolute error in X-axis for test A and test B Control point Eabs X (A) Eabs X (B) A
0,197
0,161
B
0,281
0,200
C
0,074
0,020
D
0,222
0,229
Average
0,194
0,153
Fig. 8. Absolute error in the X-axis.
Table 9. Absolute error in Y-axis for tests A and test B Number Eabs Y (A) Eabs Y (B) A
0,131
0,226
B
0,273
0,116
C
0,421
0,242
D
0,255
0,151
Average 0,270
0,184
The following Fig. 9 shows absolute errors obtained from the averages of the control points both in Test A and Test B. For Y-axis, the average absolute errors of +0.270 mm for Test A and +0.184 mm for Test B were obtained. This represents a reduction of 31.85% in Test B vs Test A. For X and Y axes, an overall absolute error of 0.200 mm was obtained. Finally, in this analysis the absolute errors in Z-axis computed for tests (A) are shown in Table 10.
134
A. Robalino Pinango and B. Culqui Culqui
Fig. 9. Absolute error in the Y-axis.
Table 10. Absolute error in Z-axis of test A Number
Manual test A Z [mm]
Automatic test A Z [mm]
Eabs Z (A)
1
−323,472
−323.782
0,310
Analysis of Setup Time Respect to the setup time taken from the workpiece zero point, Table 11 and Fig. 10 were obtained. Table 11. Setup time of workpiece zero point Number
Method
Time [min]
1
Manual
8,180
2
Automatic
1,890
The results show that the setup time of workpiece zero point was reduced from 8.18 min to 1.89 min; this represents a reduction percentage of 76.89%.
Optimization of the Setup of Workpiece Zero Point
135
Fig. 10. Setup time of workpiece zero point.
4 Conclusions • For the machine FADAL VMC3016 (2006), the processing time for a small part (machining operation time of 2 min) gave a total value of 15.45 min, of which the workpiece zero point setup time had a value of 8.18 min, which means that 52.9% of the total processing time the machine was on downtime by this task. • Therefore, the FADAL VMC 3016 machining center was a good case study for this project. Here, an artificial vision system was implemented to automate the micro movements required for the workpiece zero point setup with respect to the absolute reference of the machine (Home). • The accuracy in positioning with the artificial vision system was evaluated doing automatic tests from two types: Test A (square workpiece) and Test B (rectangular black color workpiece). Test A gave an average error of + 0.194 mm in the X-axis and 0.270 mm in the Y-axis. Test B gave an average error of 0.153 mm in the X-axis and 0.184 mm in the Y-axis. For X and Y axes, an overall error of 0.200 mm was obtained. In both Tests A and B, an average error of 0.310 mm in the Z-axis was found. • For optimization of setup time, manual tests gave a average time of 8.18 min, while automatic tests gave an average time of 1.89 min. Great reduction of 76.89% was achieved for this task.
References 1. Al-Kindi, G., Zughaer, H.: An approach to improved CNC machining using vision based system. Mater. Manuf. Processes, 765–774 (2012) 2. Datron, N.: Taking a workpiece measurement using with camera. https://www.datron-neo. com/lab-transforming-ideas/academy/tutorials/tutorial/taking-a-workpiece-measurementusing-the-camera/. Accessed 20 Nov 2016 3. Klancnick, S., Senveter, J.: Computer-based workpiece detection on CNC milling machine tools using optical camera and neural networks, pp. 59–68. Advances in Production Engineering & Management, Slovenia (2010)
136
A. Robalino Pinango and B. Culqui Culqui
4. Wang, Z., Wang, X.: On-machine measurement of large-scale workpiece based on machine vision 43(1), 1–7 (2012) 5. Wang, Z., Wu, C.: On-machine Measurement of Metal Parts Based on Machine vision, vol. 66–68, pp. 235–239 (2011). https://doi.org/10.4028/www.scientific.net/AMM.66-68.235 6. Forero, J., Bohórquez, C., Ruiz, V.: Automated measurement of turned parts using artificial vision. Universidad INNCA, Colombia (2013) 7. Inventcom: FANUC open CNC. http://www.inventcom.net/fanuc-focas-library/general/err code. Accessed 01 Aug 2013 8. Sanjeevi, R., Nagaraja, R., Radha Krishnan, B.: Vision-based surface roughness accuracy prediction in the CNC milling process (Al6061) using ANN, pp. 1–3 (2020) 9. Abdul-Ameer, H.K., Al-Kindi, G.A., Zughaer, H.: Hacia la retroalimentación de visión por computadora para un mecanizado CNC mejorado. En: 2011 IEEE 3rd International Conference on Communication Software and Networks, ICCSN 2011, pp. 754–760 (2011). https:// doi.org/10.1109/ICCSN.2011.6015000 10. Okarma, K., Grudzinski, M.: El sistema de escaneo 3D para el posicionamiento basado en visión artificial de piezas de trabajo en las máquinas herramienta CNC. En: 2012 17th International Conference on Methods and Models in Automation and Robotics (MMAR), pp. 85–90 (2012). https://doi.org/10.1109/MMAR.2012.6347906. http://ieeexplore.ieee.org/ lpdocs/epic03/wrapper.htm?arnumber=6347906 11. Xu, L., Fan, F., Zhang, Z., Chen, Y., Hu, D., Shi, L.: Methodology and implementation of a vision-oriented open CNC system for profile grinding (2019)
Tribological Characterization of Erosive Wear Resistance as a Criteria of Material Selection for Fabrication of Construction Equipment and Machinery Juan Angel Barella(B) , Juan Manuel Victorio Vallaro, Mercedes Lozano Rus, Eldo José Lucioni, and Huber Gabriel Fernández Departamento de Ingeniería Mecánica – Laboratorio de Metalurgia y Materiales, Facultad Regional Villa María, Universidad Tecnológica Nacional, Avda. Universidad 450, 5900 Villa María, Córdoba, Argentina
Abstract. Equipment built for mining and construction is held under extreme wear conditions during its entire operation time. Most of the times abrasive wear is the only type of wear considered. No reports or studies on erosive wear were found on the materials selected for the construction of these equipment & machinery. While choosing materials, surface hardness and mechanical properties are the main factors for the average selection criteria. Tribological properties are scarcely taken into account during material selection, making a huge impact on the overall performance of the equipment. The results of this erosion wear study were compared along surface hardness and profilometry of the eroded footprint. All the tests were done under the same environmental conditions and the same erosion parameters (counter body velocity, angle of incidence and quantity of erosive medium). Keywords: Erosion · Wear · Steel
1 Introduction Different steel grades are used in the manufacturing of the mentioned machinery. Among them: SAE1010, MA420, AR200, F26, HARDOX450 and SAE1045. Since there are no studies of the erosive wear resistance of these steels, comparison between them can’t be done. Leading to a high grade of uncertainty on which one to be used in different applications. Erosive wear results in huge economic loss as it reduces the performance of the equipment. This phenomenon happens in almost all the fields of application were mechanical devices are required. Characterization on tribological properties provides with additional data that can lead to a better selection for more suitable materials for any given application. Yield stress can be very useful while calculating extension or compression efforts but it does not give any clue on erosive wear resistance where fragile and ductile materials © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 137–150, 2021. https://doi.org/10.1007/978-3-030-72212-8_11
138
J. A. Barella et al.
have very different behaviors. With this statement as a hypothesis. Six different steel grades were used in this study. Testing was done under the standard ASTM G76-95 (standard test method for conducting erosion test by solid particle impingement using gas jets) [1].
Fig. 1. Erosion phenomenon [2]
Fig. 2. ASTM G76 test apparatus schematic [2]
1.1 Characteristics of Erosive Wear Erosive wear can be defined as a loss of material due to repetitive impacts of particles that can vary in shape and size (Fig. 1). Wear is a phenomenon that can be described as an open system where every enviromental factor has the potential to make large variations on the results. Nonetheless it can be tested under laboratory and test standards, where the obtained results are valid in real life scenarios [1]. The basic mechanism to produce erosive wear is shown on Fig. 2. Counter bodies with a higher hardness than the surface of the study specimens will cause a larger amount of degradation than those counter bodies with lower hardness [3, 4]. Particles with sharp edges will cause higher damage than those with a more rounded shape. [5] Finnie [6] demonstrated that the erosive wear decreases when the particle size is smaller than 100 µm. This mechanism of material degradation due to impact of counter body (particles of the erosive medium) can be divided into two categories: ductile and fragile materials Fig. 3.
Tribological Characterization of Erosive Wear Resistance
139
In ductile materials, the angle of incidence makes a huge impact on the obtained results. Figure 4 shows how wear gets to a maximum at impact angles between 20° and 30°. For fragile materials the maximum wear is reached when the angle of incidence gets closer to 90° [3–5, 7].
Fig. 3. Erosion cases [1] a) Ductile materials. b) Fragile materials.
Fig. 4. a) Ductile. b) Fragile [1].
Fig. 5. Fragile material footprint. [own source]
2 Materials and Methodology 2.1 General Aspects Selected materials were chosen due to their availability for local and regional manufacturing companies to acquire. All of them are mostly found in sheet metal and profiles.
140
J. A. Barella et al.
Fig. 6. Ductile material footprint. [own source]
The tests were conducted under the standard ASTM G76-95 (standard test method for conducting erosion test by solid state particle impingement using gas jets) [1] with specific adaptations described in the following sections. 2.2 Erosive Wear According to standard ASTM G76, Ottawa sand should be chosen. Due to import taxes and logistics, it was replaced with Paraná sand (erosive medium). This replacement fits the geometry and hardness recommended by the standard. All the sand samples were sieved with a #60 sieve. The testing apparatus (tribometer) was tested and calibrated in order to achieve repetitive & consistent results (Fig. 7). Before test
Aer test
Fig. 7. Erosive media. Paraná sand #60. [UTN-FRVM archive]
Each sample was blasted with the amount of 40 g of erosive media. The amounts were measured previous to the testing using a digital balance RADWAG model PS 360.R1. For this quantity of erosive medium, each test lasts approximately 40 s ±1 s. Nozzle opening diameter = 3 mm ±0.02 mm. Same as established by the standard. The normal distanced measured between the opening of the nozzle and the surface of the untested sample is of 10 mm ±0.5 mm. Erosive particles must impact at a specific range of speed. Pressurized air (20 psi ±1 psa) is used to drive the erosive medium.
Tribological Characterization of Erosive Wear Resistance
141
Impact speed determines the energy carried by the particles. Speed is measured with a two-disc spinning device. The disks are parallel and separated by a 6 mm gap. Both discs share the same rotation axis. A small hole is located at the top disk 40 mm offset from the center (Fig. 8).
Fig. 8. Schematic for speed measuring [2].
Fig. 9. Impact mismatch. [own source].
The speed measuring device is held inside the testing apparatus (tribometer). A DC electric motor spins the device at 5400 rpm. After it starts spinning it is blasted with the erosive medium as it would happen in a regular test. A small portion of the particles passes through the hole and impacts the lower disk. The mismatch (distance between centers) (Fig. 9) of the hole on the top disk and the footprint on the lower disk can be used to calculate an average speed of the particles. The average speed used in these tests is of 70 m/s. Since most of the particles impact the disk and only a small portion passes through the hole, the speed measurements are done with a larger amount of erosive medium, lasting for about 240 s ± 1 s. Surface hardness was measured in each test sample. Three measurements were taken per sample in order to get an average value. Both Rockwell C and Rockwell B scales
142
J. A. Barella et al.
were used. Same balance (RADWAG Model PS 360 R.1) was used to determine the mass of each test sample before and after the test. Profilometry measurements were done with a Mitutoyo SV-C3200 profilometer. 2.3 Test Samples All the test samples come from store available sheet metal. Dimensions are 25 mm width 60 mm height and a surface ground to 6.35 mm of thickness. Chemical compositions are detailed in the following Table 1. Table 1. Chemical compositions. Sample
Chemical composition C [%]
Si [%]
Mn [%]
P [%]
S [%]
SAE1010
0,11
0,41
0,02
0,03
SAE1045
0,46
0,71
0,02
0,01
AR200
0,35
1,20
0,01
0,01
0,25
F26
0,25
0,35
0,03
0,04
HARDOX450
0,24
0,40
1,50
0,02
0,01
MA420
0,12
0,50
1,60
0,03
0,02
Cr [%]
Ni [%]
Mo [%]
B [%]
1,32
1,40
0,48
0,00
3 Results The results are shown in Table 2. Both hardness (HB) and removed mass (MP) are direct measurements. Equivalent carbon content (CE & PCM) is obtained after chemical composition. Volume reduction (VP) and eroded area (AE) after profilometry measurements. Eroded area as can be considered as an approximate. Accurate measurements of this parameter can lead to biased results due to the nature of the erosive wear test. As expected, the portion of the test sample right below the nozzle of the test apparatus gets the most damage due to focused stream of erosive medium. Nonetheless a small amount of counter bodies damages the surroundings of the “main” footprint. This additional damage should not be accounted. It is only considered eroded area when the depth of the footprint exceeds 0,05MM. Profilometry of test samples is shown in Fig. 10. More often than not, erosive wear resistance is quantified only by the amount of lost mass. The obtained results of lost mass are shown in Fig. 12a. After testing, measurement and comparison of the results, we got to the conclusion that the geometry of the footprint produced by the erosive wear gives a more complete picture of the erosive wear itself. It is necessary to consider both the lost mass and the geometric characteristics of the damaged area.
Tribological Characterization of Erosive Wear Resistance
143
Fig. 10. Profilometry of different test samples.
As the results of the tests and measurements show, volume loss and eroded area are not direct function of the amount of lost mass. As an example, F26 test samples have a high mass loss but lower volume reduction when compared with the other steel samples. SAE 1045 is placed in the midfield hardness of the tested samples. Yet it has the lowest amount of lost mass, low volume reduction but a higher eroded area when compared with the rest of the samples. HARDOX 450 has a slightly higher amount of lost mass when compared with SAE 1045 but a considerably higher volume reduction when compared with the later. 6.00
Hr
CEr
PCMr
4.00 2.00 MA420
HARDOX450
F26
AR200
SAE1045
SAE1010
0.00
Fig. 11. Hardness (r), CE(r) and PCM (r). (IMPORTANT: Lines between values are for ease of visualization)
If we differentiate between two groups: High surface hardness (SAE 1045, AR200, HARDOX 450) and low surface hardness (MA420, F26, SAE1010): as expected, increase in hardness translates into a decrease in wear (removed material). To our surprise the increase in surface hardness is related to a higher volume reduction and eroded area. Results shown in Fig. 13b, c. This effect can be explained on the mechanics of wear in fragile materials, where the counter bodies cause micro fractures and further material removal (Figs. 3, 4, 5 and 6). As mentioned before, erosive wear can be explained as a process that involves: impact of counter body-cracking-fatigue-material removal. Being a mechanical process, it can be related to a thermal process, Welding. As both share the process of cracking, equivalent carbon content, a normal parameter used in welding, can be borrowed to be used as an indicator of the performance of any given steel against wear. Tylczak [8] was able to
J. A. Barella et al.
MA420
HARDOX450
F26
AR200
SAE1045
0.012 0.010 0.008 0.006 0.004 SAE1010
MP [g]
144
(b) Volume reduction
MA420
HARDO X450
F26
AR200
1.4 1.2 1.0 0.8 0.6 0.4 SAE101 0 SAE104 5
VP [mm3]
(a) Mass loss
(c) Eroded area
Fig. 12. Characterization of erosive wear.
relate equivalent carbon content to abrasive wear with novelty mathematical models. There’s no record of any studies relating equivalent carbon content to erosive wear. Another reason to use and relate equivalent carbon content (CE) (1) as an indicator on erosive is the fact that a portion of the tested steels have other alloying elements aside from carbon [9]. Each of these elements tends to influence the hardness and weldability of the steel to different magnitudes [10]. Equivalent carbon content allows us to quantify and compare the combined effect of all those elements. We choose to quantify these effects with two different methods. Both give close results and complement each other. The first is equivalent carbon content, calculated under the formulas developed by the American Welding Society (AWS) and the second one being critical metal parameter (PCM) (2) developed by the Japanese Welding Engineering Society (JWES) [11–13]. %Mn %(Cr + Mo + V ) %(Ni + Cu) %Si + + + 6 5 15 6
(1)
%Si %(Mn + Cu + Cr) %Ni %Mo %V + + + + +5·B 30 20 60 15 10
(2)
CE = %C + Pcm = %C +
For ease of comparative purposes and the use of a single scale. All three quantities, Hardness, CE and PCM are shown in relative values. The Adjusting value used for normalization is the lowest measured or calculated value of each quantity (Fig. 11). Summarized in a brief and concise manner, CE and PCM give an idea on how carbon and other alloy elements tend to influence the material behavior. Changes in both
Tribological Characterization of Erosive Wear Resistance
145
MP [g]
0.015 SAE1010 F26 MA420
0.01 0.005
AR200
HARDOX450
SAE1045
0 0.80
1.30
1.80
2.30
Hr
2.80
(a)
VP [mm3]
1.5 SAE1010
1.0
AR200
F26 MA420
0.5
HARDOX450
SAE1045
0.0 0.80
1.30
1.80
2.30
Hr
2.80
(b)
AE [mm2]
25 AR200
20 F26 SAE1010
15 10
HARDOX450
SAE1045
MA420
5 0 0.80
1.30
1.80
Hr
2.30
2.80
(C) Fig. 13. Effects of Surface hardness.
quantities, CE and PCM have a huge effect on surface hardness, which leads to changes in erosive wear resistance. The direct correlation between CE and PCM can be verified while comparing and contrasting the different damage indicators (MP, VP, AE) of each test sample. Increments in CE and PCM lead to a reduction of MP and changes the overall shape of the footprint, increasing AE and slightly decreasing VP. In most cases, such as SAE 1045 and F26 steel, VP, MP and AE values hold a direct function between CE and PCM. On the other hand, for AR200 steel and HARDOX, the values obtained of VP and AE are not directly proportional to the value of CE and PCM. Special attention at the moment of
146
J. A. Barella et al. 0.012
SAE1010
MP [g]
0.01 F26
0.008
AR200 HARDOX450
MA420
0.006
SAE1045
0.004 0.002 0 0.80
1.80
2.80
3.80
CEr
4.80
5.80
6.80
(a)
VP [mm3]
1.5 SAE1010
1.0
AR200 F26 MA420
0.5
HARDOX450
SAE1045
0.0 0.80
1.80
2.80
3.80
CEr
4.80
5.80
6.80
(b)
25
AE [mm2]
20
AR200
F26
15 10
SAE1045 SAE1010
HARDOX450
MA420
5 0 0.80
1.80
2.80
3.80
CEr
(c)
Fig. 14. Effects of CE
4.80
5.80
6.80
MP [g]
Tribological Characterization of Erosive Wear Resistance 0.012 0.01 0.008 0.006 0.004 0.002 0
SAE1010
F26
147
AR200
MA420 HARDOX450
0.80
1.30
1.80
2.30
2.80
PCMr
SAE1045 3.30
3.80
4.30
(d)
VP [mm3]
1.5 AR200
SAE1010
1.0
HARDOX450
0.5
F26
MA420
SAE1045
0.0 0.80
1.30
1.80
2.30
PCMr
2.80
3.30
3.80
4.30
(e)
25 AR200
AE [mm2]
20 15
F26
SAE1010
SAE1045
HARDOX450 MA420
10 5 0 0.80
1.30
1.80
2.30
PCMr
2.80
3.30
3.80
4.30
(f)
Fig. 14. (continued)
analysis should be taken while studying these cases, where this abnormal performance is related to a poor resistance against erosive wear. Being both CE and Surface Hardness independent and correlated variables (Fig. 11), The influence on erosive wear of these to variables is shown in Fig. 13 and 14. It is interesting to notice that the dependent morphologic characteristics of erosive damage (VP, AE) can be modelized as geometric parameters of an ellipsoid where the semi major axis (a) of the ellipse becomes the radius of the cross section that represents the eroded area. The lower semi volume represents the volume reduction. As in our case a = b, the cross section will be of a circle with a perceived radius (Rap) related to the eroded area.
148
J. A. Barella et al.
Fig. 15. Ellipsoid model used to describe erosive damage.
Volume reduction is defined as the lower semi volume of the ellipsoid. With a cross section of radius Rap (3) and depth Pap (4). AE (3) Rap = π Pap =
3 VP · 2 · π Rap2
(4)
These two perceived values can be used to explain with certain grade of accuracy the characteristics (both VP and AE) of the erosive wear and performances on the resistance of it. The cross section of the eroded area is considered as in the surface of the test sample and the depth Pap as from the surface of the test sample to the lowest point of the footprint [Table 2]. Table 2. Perceived Radius and depth Material
Erosive wear Volume reduction
Eroded area
Perceived radius
Perceived depth
(VP) [mm3 ]
(AE) [mm2 ]
(Rap) [mm]
(Pap) [mm]
SAE1010
1,1444
13,8719
2,99
0,06
SAE1045
0,5239
16,4245
3,25
0,02
AR200
1,0805
21,0625
3,68
0,04
F26
0,5528
15,8819
3,20
0,03
HARDOX450
0,8693
16,1539
3,22
0,04
MA420
0,5169
12,3382
2,82
0,03
As shown in Table 2 and Fig. 15. SAE 1045 and F26 have a lower perceived depth in exchange of a more extended eroded area. MA420, AR200 and HARDOX450 have a tendency to develop a footprint of greater depth, making them not suitable for applications where erosive wear resistance is critical. Finally, due to its combination of high
Tribological Characterization of Erosive Wear Resistance
Rap
0.20
Pap
0.15
2.40
0.10 2.10
0.05
1.80
Pap [mm]
Rap [mm]
2.70
149
MA420
HARDOX450
F26
AR200
SAE1045
SAE1010
0.00
Fig. 16. Perceived Radius and depth
mass loss and high-volume reduction, SAE 1010 steel should not even be considered for similar applications.
4 Conclusions This article started as a private study meant to determine which steel alloy was more suitable for fabrication of concrete truck drums. As mentioned in the introduction, abrasion is a well know phenomenon in this application. Erosion on the other hand is much more ignored. It is a common practice to directly relate surface hardness to erosive wear resistance. This might be true under certain conditions and applications; it is not the case of our study results. We found that SAE 1045 steel has a higher erosive wear resistance in comparison to HARDOX450 steel. Being the later of higher surface hardness compared to SAE 1045. As Hardness is related to Equivalent carbon content, it is important to notice the maximum values found in SAE 1045 and HARDOX 450 and the minimum found in F26 Steel. Relative values are shown in Fig. 11. Tribological characterization of materials should not be conceived as the value of a single physical quantity but a collection of them. The use of hardness measurements and relative equivalent carbon content should be combined with erosive wear tests in order to make up for a simple yet consistent method to solve the problem of material selection where erosive wear resistance becomes a critical factor. In addition, volume loss and eroded area could be used to easily compare results and correlate other physical quantities in test samples with similar erosion footprints. Being this later statement a matter of further studies.
References 1. Norma ASTM G76–95 - Standard Tests Method for Conducting Erosion Test by Solid Particle Impingement Using Gas Jets 2. Morello, N., Alejandro, R., Vallaro, V., Manuel, J.: Reingeniería del dispositivo para determinar la resistencia al desgaste en seco de materiales por erosión, CyTAL 2018 – VIII
150
J. A. Barella et al.
3. Treviño Acevedo, A.M.: Desarrollo de un metodología para pruebas de abrasión. Tesis de Maestría. Universidad Autónoma de Nuevo León. México (2004) 4. Neilson, J.H., Gilchrist, A.: Erosion by a stream of solid particles (1967) 5. Bitter, J.G.A.: A study of erosion phenomena: Part I. Wear (1963) 6. Finnie, I.: Some Reflections on the Past and Future of Erosion. Wear (1995) 7. Ruff, A.W., Wiederhorn, S.M.: Erosion by solid particle impact (1979) 8. Tylczak, J.H.: Correlating alloy composition to wear in low-alloy steels, pp. 73–78 (1987) 9. Larsen Badse, J.: The abrasion resistance of some hardened and tempered carbon steels. Trans of The Metal Soc of AIME, vol. 236, pp. 1461–1466, Act. 1966 10. Garcia, A., et al.: Influencia del carbon y el cromo en el comportamiento tribológico de aleaciones férreas. Rev. Metal. Madrid. Vol. Extr., 493–497 (2005) 11. AWS D1.1-2010 Anexo I. Directriz sobre métodos alternativos para determinar el precalentamiento. American Welding Society Inc. (2010) 12. Yurioka, N., Kasuya, A.: chart method to determine necessary preheat in steel welding. Welding World 35, 327–334 (1995) 13. Yurioka, N.: Comparison of preheat predictive methods. Weld. World 48, 21–27 (2004)
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic ´ Johanna Tobar1,2(B) , Alan Pr´ ocel1 , Andrea L´ opez1 , Bladimir Bacca2 , 2 and Eduardo Caicedo 1
Universidad de las Fuerzas Armadas ESPE, Sangolqu´ı 170501, Ecuador [email protected] 2 Universidad del Valle, Cali 76001, Valle del Cauca, Colombia
Abstract. The pandemic caused by COVID 19 has led to the transition from traditional education to other modalities that use technological resources as their primary tool. This work proposes a robotic theater as a reinforcement mechanism in education and transmission of knowledge in various areas for regular education in children. The robotic platform is made up of an automatic curtain and 3 NAO social robots. The elements are controlled from a user interface (GUI) that easily and intuitively allows the creation of theatrical scripts. As part of the investigation, the adaptation of the play “Planeta” by the author Alan Rej´ on was staged. This was presented to three groups of people. The first scenario places professionals in the area of education: teachers of regular primary education, students of initial education, and an educational psychologist. A second scenario is a group of students between the ages of 8 and 9. The results obtained show a 92% acceptance of research as an innovative tool, and 90% of those surveyed consider that robotic plays can awaken children’s awareness in education. Possible areas in education for applying the robotic platform could be history or language and communication are also exposed. Keywords: Robotic theater · Teaching processes · Educational technological tools · Virtual education · COVID 19
1
Introduction
The spread of COVID 19 in the world has led different countries to take measures to mitigate the spread of this disease. The closure of educational institutions and the implementation of virtual learning have been the first consideration when complying with social distancing policies and avoiding transmitting the virus. These new modalities present several challenges. According to the analysis [1] of the policy “Suspension of classes without stopping learning” of the Chinese government, the epicenter of the COVID19 pandemic, the possible difficulties that these new modalities face include: the weakness of the infrastructure of online teaching, the inexperience of teachers, use of information technologies and new teaching tools applied to this environment, among others. c The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 151–166, 2021. https://doi.org/10.1007/978-3-030-72212-8_12
152
J. Tobar et al.
Also, educational standards are analyzed to achieve an optimization of the school and preschool education process; and improving its quality, which requires the implementation of innovative pedagogical technologies by teachers. The challenges that teachers must take on are fully linked to the adaptability of modern pedagogical technologies and the lack of proven models to manage the introduction of these technologies in practice [2]. The transition of the teaching process modality caused by COVID 19 has increased the use of technological tools to carry out classes in the middle of virtual environments, emphasizing their correct use and selection [4]. The implementation of digital tools has become popular and promising, including digital storytelling in virtual classrooms, the use of Google applications in advanced learning through discussion groups, screencasting of assignments, motivational videos to support student engagement, recordings of meetings online as support material [5]. Thus technology is used in pedagogical methodologies. Interest in the educational use of robotics has increased in this field [23]. The success of the use of different robots in education is proven in several studies. The factors that are observed to validate the use of social robots in education are the effectiveness of cognitive and affective outcomes; the impact of using a physically incorporated robot (in comparison with alternative technologies) and the interaction role that the robot can take in an educational process [6,7]. Commonly, the implementation of social robotics in education is divided into the areas of science, technology, and language; and in the roles that are assigned to the robots are roles of tutor, companion, or tool [8]. In contrast, the relationship between art and education is considered from the perspective of the innovative world, and the impact it creates on education contributes to the development of these skills and characteristics within formal and non-formal teaching [11]. This is how the potential of social robotics is explored and implemented in integrated art platforms [12]. Robotic theaters have been designed and developed, which have been used in non-formal education in extracurricular programs [13]. The link with the arts: acting, dancing, singing, and drawing has allowed the sharing of content related to science, robotics, and computing. These platforms are also used in the context of promoting STEAM (Science Technology Engineering Art Mathematics) education in programs such as “robot opera” or “musical robot theater.” [14]. These studies involve children ages 5–7 and use a variety of social robots. The objectives of these programs focus on the interaction of children with robots and the development of motor skills, writing, acting, and participating in theatrical production with the robots. As a result, the significant potential of this technique of using robots as actors in theater productions within the STEM fields was demonstrated [15]. Social robotics is exposed as a hybrid knowledge space that encourages interaction and collaboration between many different disciplines. The development of platforms and systems with the use of social robots raises several challenges.
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
153
Among these is the sought to achieve a transdisciplinary understanding of areas taking as a complement the experience of using social robots [16]. With this context, the use of a robotic theater is proposed as a technological tool to support education for children of school and preschool age. The results of presentation sessions of theatrical production are presented as an initiative to support virtual education during the health emergency caused by COVID 19. Two scenarios are presented. First, primary education teachers, initial education students, and educational psychologists participate. The second presents the play before a group of students of 4th year of primary education, aged between 8–9 years. Through a survey presented to the validation groups, substantial results were obtained to implement robotic theater in the initial stages of education and primary education. Also, the versatility of the system based on collaborative robotics is explored to be used in various areas of knowledge of an innovative educational process.
2
System Development Methodology
The robotic theater-based system is used as a teaching tool and is developed from the guidelines and needs that arise during the learning process of children of school and preschool ages.
Fig. 1. System development methodology.
Figure 1 shows the steps used to develop the robotic system. The process starts with required parameters by teachers or tutors, this allows the next step of scriptwriting. Later the script is made using the platform’s software. Once this is achieved all the elements are configured according to their role. Finally, the script is interpreted and the play is executed. 2.1
Parameters for Script Writing
The robotic theater is presented as a flexible support tool that is not explicitly tied to a single work topic. The robotic platform is used as a resource used by the teacher or tutor to approach to the student. This is how the first considerations are given by the people responsible for the teaching process. They choose the objectives that must be met with the development of the play: the selection of the theme of the play, the characters that must participate, and the estimated duration. Considering the ease of working with languages of the NAO social robots, which are elements of the system, it is possible to select the language in
154
J. Tobar et al.
which the play is performed. Another of its advantages lies in the communication between them and their integration into the robotic theater, allowing the work of several robots without limiting the number of characters that a play can have. There are several possible uses in the different teaching modalities implemented during the pandemic. The technological tool is exposed as an adaptable model to be used in virtual education, in collaboration with online education or homeschooling with the help of parents or tutor. Modalities that each present their requirements, establishing both parameters in the creation of the script and the presentation formats. The robotic theater adapts to the teacher’s needs or tutors, respecting the rules established by the educational institution or teaching program. 2.2
Script Development
The scriptwriting starts from the requirements and parameters indicated by the teachers or tutors. The creation of the play serves the requested focus. This work develops a topic of general interest, such as caring for the environment and the planet. The script is based on the play for three characters called Planet by the author Alan Rej´ on. According to teachers’ indications, necessary modifications and adaptations are made by the programmers. The script presents a short fictional story. It details life outside of planet earth, which has become uninhabitable due to poor care and pollution. Two figures are arranged as main characters. A female character named Luna, who represents a girl who establishes a conversation with Saturn, a male character, and explains the reasons why they can no longer live on planet earth. To communicate in a more detailed and dynamic way, the script assigns expressions with gestures to each of the characters according to their dialogues. Specific actions, such as dancing, are also added. Making use of these advantages, musical scenes are incorporated with songs that capture the attention of children. Additionally, the narrator’s figure is also created; this participates in the description of the scenes and explains the context before the performances of actors. The narrations and dialogues of the script are carried out in sequence, except for the last scene. The play ends with synchronized farewell participation in which the two actors and the narrator interact. These figures will later be assigned to the robots. Knowing preschool and school-age children’s ability to pay sustained attention to a specific topic, the script works with an adaptation of a short play of approximately 7 min [24]. The creation and execution of the script take place on the robotic theater platform. The next section describes the configuration of the platform elements that allow the performance of the play.
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
2.3
155
Robotic Theater Platform Elements
The robotic theater is formed not only by social robots but also includes a user interface that allows the creation of the script and programming of the robots; and an automated stage-curtain. Script Creation and Execution Interface. The creation of the script takes place in the user interface of the robotics platform. The following subsections describe the user interface and script creation in it.
A. User Interface Description The user interface of the robotic platform performs the generation and execution of the play, which are carried out in several stages. This user-friendly interface (Fig. 2) allows the user to manage this technological model quickly.
Fig. 2. Script structure
Each act is made up of scenes, and these tasks. The programmer user assigns the tasks to the robots, which are necessary actions preconfigured in the software, such as expressions of gestures, walking, dancing, among others. Besides, the texts of the dialogues are assigned to be communicated by the robot. In this stage, the voice configuration is carried out, and the communication of the robots with the system and with each other is established to carry out collaborative work. Tasks are part of a scene. Scenes can be made up of several tasks. The tasks can be executed at the same time when the robots are synchronized with each other or executed in sequence. Therefore more than one robot actor can
156
J. Tobar et al.
Fig. 3. Execution of the example script using virtual NAO robots. Scene 1-left. Scene 4-right
participate in each scene. This work was carried out with the incorporation of physical robots. However, the tool also offers the possibility of connection with free access Choregraphe software. Choregraphe has a virtual NAO robot, which allows the simulation of actions and gestures, as shown in Fig. 3. This option facilitates the development of works if not having physical robots. This increases the possibility, that the child or the teacher, to learn without the presence of the programmer or the robot. The work scenario with the combination of physical and virtual robots can also be considered.
B. Script Development in User Interface The play developed by programmers is divided into 4 acts. It is shown in Fig. 4a. The first integrates scenes in which the narrating robot gives an introduction to the story and performs tasks such as greeting, pointing, exclamation and calm gestures. The second act has the participation of the two robot actors. Dialogues are assigned to each of them in order to establish a conversation. They complement the scenes with tasks such as dancing, greeting, tranquility, and pointing gestures. In the third act, new dialogues are assigned to the robot actors; both of them interact with each other, and the tasks for each of the scenes include movements that represent exclamation, affirmation, denial, tranquility, and pointing and crying gestures. The final act reunites the three robots. Dialogues are assigned to the narrator robot for interaction with children. The movements that this robot performs represent gestures of tranquility, exclamation, and denial. Finally, the three robots act in a synchronized way executing a goodbye greeting while the narrator expresses the final dialogue and goodbye. The script is then
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
157
Fig. 4. a) Creating the script in the technological tool interface. b) Interpretation of the script in the technological tool interface
saved for interpretation and execution. This is done through a second interface that allows the execution act by act, and also shows the progress of the work performed in the bars located at the bottom of it. As shown in Fig. 4b. Stage - Curtain. The stage - curtain of the robotic theater is an automated semi-mechanical structure. This structure consists of fixed and mobile parts in analogy to a real theater stage with participation of human actors. With the use and activation of sensors and actuators, the curtain opens and closes automatically. This allows the perception of changing scenes and creating a more theatrical/artistic environment for the performance. According to the play presented, an animated projection of the environment in which the script is developed is also added as theater background. Social Robots NAO. The execution of the play also involves the participation of three NAO humanoid robots. The use of these robots in the field of education focuses on the results of interaction within the classroom and learning. [7] In a teaching process social robots can used as a tutor, partner, or tool [8]. The robotic theater places NAO robots as a tool for teaching support. Taking the advantages of social robots, characters are assigned to them, with their corresponding dialogues, actions, and tasks according to the dialogue with gestural expressions. Two robots took the role of main actors representing a female and a male character, and a third robot served as the storyteller and scenarios narrator. Thanks to the ease of configuration of the robots, to provide characteristics and increase the distinctiveness between the characters, the tone of voice settings are made. This allows the robot to take on more personality and fulfill the role
158
J. Tobar et al.
of playing a boy or girl; or an adult man or woman. Additionally, accessories were developed, which without interrupting functions or impairing the robot’s operation were incorporated into the visual presentation, giving the robot a feminine or masculine style.
3
Experimental Procedure
The experimental procedure begins with the identification of the target population, contemplating ethical considerations. Then the presentation format of the play is selected and it is presented in sessions to the target population. The steps are shown in Fig. 5.
Fig. 5. Experimental procedure
3.1
Target Population
For the validation of the system as a technological aid tool in the teachinglearning process, three target populations were considered. The first group places regular primary education teachers. Teachers profile contemplate the teaching-learning management component oriented to the planning of educational processes with the use of technological and didactic resources to meet the objectives of professional competence development. [17] This first validation counted a total of 7 teachers and an educational psychologist who work with children aged 6–12 years. The second group, that participates in this work, are students of the Bachelor’s degree in initial education of the Universidad de las Fuerzas Armadas ESPE. Their work roles are based on the evolutionary and integral development of children from 0 to 6 through pedagogical and social innovation [18]. A total of 24 5th-level students participated in the presentation and subsequent survey. Within the third group are 12 children in primary school, aged between 8 and 9 years from the educational Institution EMDI school [19]. As extracurricular activities, the children receive workshops and/or tutorials. The work theme selected by the Institution is caring for the environment, on which the script of the play is based. During the presentation, there was also the participation of the teacher guides of the group of children.
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
3.2
159
Ethical Considerations
Because the validation studies with robotic theater involve human participants, the research is carried out under the considerations issued in the Universidad del Valle, Colombia, with registration 003–020. The primary purpose of ethics committees is to protect participants in human research [20]. This being the most critical factor for the selection of the target population. 3.3
Presentation Format
With the outbreak and spread of COVID-19, educational institutions have implemented various study modalities intending to mitigate the risk of contagion of this disease; consequently, students are not allowed to attend institutions, and education has had to change its traditional form to an online one [21]. Regarding the new virtual education modality, the effective selection and use of learning tools play an important role [22]. Due health emergency this work was adapted to a presentation in a prerecorded format. The option of a real-time presentation was also considered; however, this leads to other challenges related to digital and communication resources. The execution of the work was carried out without an audience only with the participation of the researchers. Audio and video were recorded in separate parts to preserve the quality of both formats. Using video editing software, these two parts were synchronized. It is important to emphasize that no special effects were added during the video editing, or any modification was made to the video or audio files. 3.4
Presentation Sessions
The play was presented in three different sessions to the previously described target population groups. The first session includes the online presentation of the recording of the play to primary education teachers and educational psychologists. An exposition about the project objectives is carried out. Later the files to be presented are housed in an online storage space allowing teachers to have access to them. The play’s video was watched individually, and then they proceeded to complete a questionnaire about the play presented. Second session aimed at students of the initial education career, was carried out with the use of the MEET video conference platform. The project manager gave brief introduction to the robotic theater. Developers of the project made known the operation of the system. Then the video of the play is presented. Subsequently, questions and concerns are received, which are resolved by the researchers. Finally, the students proceed to complete a form with questions about the presentation observed. The results of the same are exposed in the following section. Finally, the third session held for the children of the EMDI school educational institution was held during the assigned tutoring space using the same methodology of videoconferencing platforms. Extracurricular space where the
160
J. Tobar et al.
Fig. 6. Left-Presentation session with teachers. Right- Presentation session with children
teacher approaches her students to solve doubts about the educational process and offer a guide. The session begins with the welcome of the tutor-teacher, who presents the topic to be discussed and gives an introduction to the robotic theater developers, who briefly explain how it works. Then the play is presented. Once the presentation is finished, a space for dialogue is established where children share reflections and interpret the information received. Figure 6 shows the presentation sessions with teachers and children.
4
Results
For the validation of the project, both the first group made up of primary education teachers and the second group of initial education students responded to a questionnaire with specific questions about robotic theater and the theater play seen. The questionnaire was focused on objective questions related to using of the robotic theater as an innovative tool in education and its possible uses in other subjects of preschool and primary education. Technical aspects of the construction and design of the tool were not contemplated in the questions. For a quantitative assessment, the following weighting is established for the answers to the questions. Being 4 the maximum weight corresponding to the highest level of user satisfaction and 1 to the lowest level. 1. 2. 3. 4.
Strongly disagree Somewhat disagree Somewhat agree Strongly agree.
The questions that were considered in the survey are shown in Table 1. With the graphic representation of the answers obtained, the high degree of acceptance of the technological tool implemented based on the robotic theater
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
161
Table 1. Questions presented in the survey Reference
Question
Target population
Question 1 Do you think that the technological tool system will generate interest in children?
Group 1 and 2
Question 2 Do you consider that technological tool is innovative?
Group 1 and 2
Question 3 Do you consider that robotic theater can be Group 1 and 2 a useful tool in the learning process of children? Question 4 The interpretation of the work is done in a fluid way
Group 1 and 2
Question 5 Consider that these types of robotic plays can arouse the attention of children in education
Group 1 and 2
Question 6 Consider that the interpretation of robots can clearly convey a message or knowledge
Group 1 and 2
can be observed in the results. The interest that robotic theater awakens in students and future educators is high. The survey responses are displayed in bar graphs in Fig. 7, 8. These use the blue color to indicate the number of people who chose the answer “Strongly agree” in each of the six questions. The orange color corresponds to the answer option, “somewhat agree.”. The answers “somewhat disagree” and “strongly disagree” are represented by the colors gray and yellow, respectively. Figure 7 shows the responses obtained from the questionnaires applied to the first target group. The answer option most used by the participants in group 1 corresponds to “Strongly Agree.”. The variant “somewhat in agreement” is also used by two people to answer most of the questions, except for number 4.
Fig. 7. Analysis of the reason “strongly agree” with regular basic education teachers
162
J. Tobar et al.
The latter refers to the fluidity of the work’s development and divides the group to choose the weightings above in the same ways. The options “somewhat disagree” and “strongly disagree” were not used as responses by the first group. Figure 8 presents the results obtained from the validation with the second target group. The majority use of the option “strongly agree” is evident. The second most used option corresponds to the option “somewhat agree.” Participants in this group also chose the answers “somewhat disagree” and “strongly disagree.” Question 2 on the consideration of robotic theater as innovative shows the acceptance of 20 people in the group with the answer “strongly agree” is the question with the most frequent use of this option. The option “strongly disagree” is not presented in this question. Question number 4, which refers to the fluidity with which the work is developed, is positioned as the least used with the option “strongly agree.” Half of the participants in group 2 chose this option to define the ability to interpret the script. However, only four people disagreed, opting for the options “strongly disagree” and “somewhat disagree” equally. The remaining 8 participants used the option “somewhat agree” in this question.
Fig. 8. Analysis of the reason “strongly agree” with initial education students
Statistical tables are also presented on the tabulated results obtained. Table 2 corresponds to the responses obtained from group 1(G1) and group 2(G2). In group 1 according to the weight assigned to questions, the mean of the majority of questions corresponds to 3.75 points, except for question 4, with 3.5 points out of a total of 4. The frequency of weighing four corresponds to “strongly agree” is visible in the answers to all questions. 93.75% of those surveyed in this group consider, according to questions 1,2,3,5 and 6, that the system will generate interest in children, that is innovative, that can be useful as an educational tool, that awakens the attention, and that can be used to convey a message or knowledge clearly. It can also be observed that 87.5% consider that the work is done fluidly.
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
163
Table 2. Statistics of group 1 responses Average Trend % G1 G2 G1 G2 G1
G2
Std. Dev Var. Coeff. G1 G2 G1 G2
Question 1 3.75 3.58 4
4
93.75 89.5
Question 2 3.75 3.71 4
4
93.75 92.75 0.46 0.69 0.12 0.19
0.46 0.72 0.12 0.2
Question 3 3.75 3.5
4
4
93.75 87.5
3.33 4
4
87.5
Question 5 3.75 3.63 4
4
93.75 90.75 0.46 0.71 0.12 0.2
Question 6 3.75 3.46 4
4
93.75 86.5
Question 4 3.5
0.46 0.72 0.12 0.21
83.25 0.53 0.76 0.15 0.23 0.46 0.72 0.12 0.21
For second group of professionals in the area of initial education, question 2 receives the highest score of 3.71. The question with the lowest score corresponds to number 4, which refers to the fluidity of the work with 3.5 points. The frequency of use of weight four corresponding to “strongly agree” is notable in all questions. Question number 2 indicates that 92.75% of the participants consider robotic theater to be innovative. Followed by question 5 with 90.75%, which considers that plays can capture children’s attention and question 1 with 89.5% on the interest that robotic theater arouses in children. Question number 4 is the one with the lowest percentage of 83.25% considers that the play develops fluently. Additionally, the survey had an unweighted question considered as a basis and feedback for future work. This space allowed respondents to know the areas of knowledge in which the tool can be used to support teaching.
Fig. 9. Possible areas of application of technological tool
164
J. Tobar et al.
Figure 9 shows the results obtained from the first group. In this question, teachers indicate the subjects of history, natural sciences, and arts as the areas with the most significant potential in which robotic plays can be implemented during the teaching process. The option with the least reception, and that would present the most significant difficulty for teachers when implementing a play, corresponds to the subject of Mathematics. Additionally, he found the option of adding an area or subject that, under his expertise, can handle robotic plays in which the subject of ethical values was suggested. According to the consideration of specialists in initial education, the possible areas of application, corresponding to the second group of the target population. The option with the most significant potential corresponds to the subject of communication and language, followed by the subjects of natural science, arts, and history. In this group, the area that would represent the most significant difficulty for the development of plays would be the subject of coexistence. In the same way, the subject of Mathematics at the time of indicating the possibilities of working with robotic theater, only 10 participants would work out of a total of 25 people.
5
Conclusions
The pandemic caused by COVID 19 has caused a significant impact on the world. Several areas have been affected, including education. The acceleration process in approaching a more technological society in which innovative tools play a critical role is evident. Without a doubt, within the educational process to promote the quality of education, it is contemplated integrating methodologies with the use of technology. Likewise, the health emergency and the transition to the virtual education modality have prompted teachers to seek practical tools adapted to a virtual environment. In this context, a technological tool based on a robotic theater was presented as a reinforcement tool in the teaching process in virtual education, through a play it was possible to motivate children to care for the planet. Besides, the use of this system based on collaborative robotics demonstrated in both primary and initial education teachers the high interest that robotic theater has as a tool. This could be evidenced through the results obtained from the surveys carried out. Where 92.75% of the participants consider robotic theater to be innovative and 89.5% agreed on the interest that robotic theater arouses in children. The ease of work that the robotic theater allows for the implementation of scripts shows it as a versatile tool adapted to the educational process’s needs. According to the new modalities of education implemented: virtual modality, online modality, and homeschooling, the potential of work in presentation formats of the works is explored. Also, the robotic platform allows work with free access software for work with a virtual simulation of robots. The potential reflected by the work for its implementation in various knowledge areas could be tested through the survey. With the selection of possible
Robotic Tool as Support in Teaching Processes During COVID 19 Pandemic
165
areas in which technological tools can work, nine subjects are indicated to realize plays. Thus, the adaptability of robotic theater to work with various themes is demonstrated, achieving interdisciplinarity. On the other hand, the surveys also reflected the points that should considered for future work. Being the results of the fluidity of the script’s execution, one of the elements with 83.25% acceptance. The work allows knowing the possible areas of work of the plays in initial education and regular primary education. Another important point the exploration of proposals to the difficulties that the integration of theater would have in subjects that received low acceptance, such as mathematics.
References 1. Zhang, W., Wang, Y., Yang, L., Wang, C.: Suspending classes without stopping learning: China’s education emergency management policy in the COVID-19 outbreak. J. Risk Financial Manag. 13, 55 (2020) 2. Demina, E., Atemaskina, Y., Kazyuk, N., Mikova, M.: International Scientific and Practical Conference on Education, Health and Human Wellbeing (2019) 3. Otterborn, A., Sch¨ onborn, K., Hult´en, M.: Surveying preschool teachers’ use of digital tablets: general and technology education related findings. Int. J. Technol. Des. Educ. 29, 717–737 (2019) 4. Basilaia, G., Kvavadze, D.: Transition to online education in schools during a SARS-CoV-2 coronavirus(COVID-19) pandemic in Georgia. Pedagog. Res. 5(4) (2020) 5. Ferdig, R.E., Baumgartner, E., Hartshorne, R., Kaplan-Rakowski, R., Mouza, C.: Teaching, technology, and teacher education during the COVID-19 pandemic: stories from the field. Association for the Advancement of Computing in Education (AACE). https://www.learntechlib.org/p/216903/ 6. Belpaeme, T., Kennedy, J., Ramachandran, A., Scassellati, B., Tanaka, F.: social robots for education: a review. Sci. Robot. 3 (2018) 7. Choudhury, A., Li, H., Greene, C.M., Perumalla, S.: Humanoid robot-application and influence. Arch. Clin. Biomed. Res. 2, 198–227 (2018) 8. Mubin, O., Stevens, C.J., Shahid, S., Mahmud, A.Al., Dong, J.-J.: A review of the applicability of the robots in education. Technol. Educ. Learn. 1, 13 (2013) 9. Van den Berghe, R., Verhagen, J., Oudgenoeg-Paz, O., van der Ven, S., Leseman, P.: Social robots for language learning: a review. Rev. Educ. Res. 89(2), 259–295 (2019) 10. Paul, V., de Haas, M., de Jong, C., Peta, B., Emiel, K.: Child-robot interactions for second language tutoring to preschool children. Front. Hum. Neurosci. 11, 73 (2017) 11. Kwiatkowska-Tybulewicz, B.: Arts education and theatre pedagogy as a tool for education in the 21st century. Polish case study 12. Potentials of Robot-Theater as a Platform for Integrated Art, STEAM Education,and HRI Research Insert Subtitle Here Myounghoon Jeon Industrial and Systems Engineering Computer Science Virginia Tech Blacksburg, VA, USA [email protected] Myounghoon Jeon Industrial and Systems Engineering Computer Science Virginia Tech Blacksburg, VA, USA [email protected]
166
J. Tobar et al.
13. Barnes, J., FakhrHosseini, S.M., Vasey, E., Park, C.H., Jeon, M.: Child-robot theater: engaging elementary students in informal STEAM education using robots. IEEE Perv. Comput. 19(1), 22–31 (2020). https://doi.org/10.1109/MPRV.2019. 2940181 14. Jeon, M., et al.: Robot Opera: a modularized afterschool program for STEAM education at local elementary school. In: 2017 14th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Jeju, 2017, pp. 935–936. https://doi.org/10.1109/URAI.2017.7992869. 15. Barnes, J., FakhrHosseini, M.S., Vasey, E., Duford, Z., Jeon, M.: Robot theater with children for STEAM education. In: Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 61, no. 1, pp. 875–879 (2017). https:// doi.org/10.1177/1541931213601511 16. Making Friends: Building Social Robots Through Interdisciplinary Collaboration Selma Sabanovi cRensselaer Polytechnic Institute [email protected] Marek P. Michalowski Carnegie Mellon University [email protected] Linnda R. Caporael Rensselaer Polytechnic Institute [email protected] 17. Direcci´ on nacional de formaci´ on iniciale inducci´ on profesionaldocumento en revisi´ onconstrucci´ on y validaci´ on de perfiles profesionales educativos. https://www.ces.gob.ec/doc/cuarto-seminario/perfiles%20ministerio%20de %20educaci’on.pdf 18. https://www.emdischool.edu.ec/index.html 19. https://www.espe.edu.ec/educacion-inicial/ 20. https://www.fhi360.org/sites/default/files/webpages/sp/RETC-CR/sp/RH/ Training/trainmat/ethicscurr/RETCCRSp/pr/Contents/SectionIII/a3sl25.htm 21. Basilaia, G., vavadze, D.: Transitionto On line Education in Schools during a SARSCoV-2 Coronavirus(COVID-19) Pandemic in Georgia. Pedagog. Res. 5(4), em0060 (2020). https://doi.org/10.29333/pr/7937 22. Huang, R.H., Liu, D.J., Tlili, A., Yang, J.F., Wang, H.H., et al.: Handbook on Facilitating Flexible Learning During Educational Disruption: The Chinese Experience in Maintaining Undisrupted Learning in COVID-19 Outbreak. Smart Learning Institute of Beijing Normal University, Beijing (2020) 23. Alimisis, D., Kynigos, C.: Chapter 1 Constructionism and robotics in education. http://roboesl.eu/wp-content/uploads/2017/08/chapter1.pdf 24. El desarrollo de la atenci´ on, la memoria y la imaginaci´ on. http://www.waece.org/ biblioteca/
Trends in Technological Advances in Food Dehydration, Identifying the Potential Extrapolated to Cocoa Drying: A Bibliometric Study A. D. Rincón-Quintero1,2(B) , L. A. Del Portillo-Valdés1 , A. Meneses-Jácome2,3 , C. L. Sandoval-Rodríguez1,2 , W. L. Rondón-Romero2 and J. G. Ascanio-Villabona1,2
,
1 University of the Basque Country UPV/EHU, 48013 Leioa, Bizkaia, Spain
{arincon,csandoval,jascanio}@correo.uts.edu.co, [email protected] 2 Unidades Tecnológicas de Santander UTS, Bucaramanga, Santander 680005, Colombia [email protected] 3 Universidad Autónoma de Bucaramanga UNAB, Bucaramanga, Santander 680003, Colombia
Abstract. This work presents a systematic review of the literature based on bibliometric networks structured with the VOSviewer application, which was run on different scientific databases, seeking for improvements of drying processes of food and other biomasses in rural contexts. Analyzed data reveals a growing interest on topics dealing with storage of solar thermal energy and mathematical modelling to predict properties of air, as a drying fluid in solar and hybrid dryers. Solar-biomass hybrid drying technologies that use residual heat are currently being applied. At the level of mathematical models, the literature analysis cover heat transfer mechanisms related to changes in absolute and relative humidity of the air and food biomasses. Therefore, a promissory horizon to develop a continuous dryer for cocoa beans, enabling an improvement of the final quality of dried biomass and reducing operation periods for this process in low-economy countries has been elucidated through this review. Keywords: Bibliometric analysis · Scientific research · VOSviewer tool · Dryer · Biomass · Mathematical models
1 Introduction Bibliometric mapping is an important research topic in exploratory research; it includes two aspects clearly distinguished, firstly the construction of bibliometric maps and then the graphic representation of such maps. VOSviewer software can be used for this task, for instance, to build author or journal maps based on citation data or to build keyword maps based on matching data. This program offers a viewer that allows you to examine bibliometric maps in detail [1]. By means the visualization of bibliometric networks © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 167–180, 2021. https://doi.org/10.1007/978-3-030-72212-8_13
168
A. D. Rincón-Quintero et al.
through VOSviewer tool enables a complete vision of a research area and the research orientation that should be taken in order to generate a significant contribution to the new knowledge generation. In this document, VOSviewer has been used to perform a systematic search and analysis of high-impact databases on sustainable food drying systems for developing countries; particularly post-harvest processes, such as fermentation and drying of some biomasses, as critical steps to address the quality of plant materials derived from coffee and cocoa, providing them inherent and unique characteristics of flavor and aroma [2, 3]. During, drying process, moisture is extracted from the harvested products to reduce deterioration possibilities during their storage [4]. Solar drying is a traditional method widely used in developing countries around the world. Drying at home Elba or hoppers which expose the products directly to solar radiation undergoing changes. Climatic conditions, pollution such as dust, smoke, animals among others, in fact a bad execution of the drying can lead to the proliferation of toxic fungi [5, 6]. In order to overcome these unfavorable circumstances for high quality standards for food drying, different authors have developed different types of sustainable dryers, which continue to take advantage of the solar energy resource, specifically by implementing collectors for heating air to be circulated through isolated drying chambers that protect food and allow enhanced process control [7]. In this perspective, the present review focuses on a methodology that involves the implementation of a tool for the visualization of bibliometric networks, and aims to identify the most recent advances in environmentally and economically sustainable food dryers. Results of this review would endorse a second research stage to be structured, where a prototype that integrates some of these technologies can be proposed, modeled and implemented to drive a continue and short-time drying process of cocoa beans (e.g. drying in 36 h), being focused on improving the productivity and efficiency of small and medium producers in Colombia. In addition, this study could be used as a basis for the future implementation of thermal energy storage systems given their high technological potential in terms of geographic location in a country with abundant energy resources.
2 Materials and Methods This work is based on the application of VOSviewer, a tool for the graphic visualization of bibliometric networks, addressing the identification of technologies and advances in methodologies for food biomass drying and their specific application in the processing of cocoa biomass. A map of keywords for data mining has been elaborated, starting from the summaries of relevant documents in scientific databases. Taking into account that small producers traditionally make the drying of the cocoa with solar energy, by spreading the cocoa beans in hoppers or Elba houses [8, 9], it has been inferred that this abundant renewable energy in rural areas for cocoa production would be implicit in new technologies to be implemented for this purpose.
Trends in Technological Advances in Food Dehydration
169
Thus, the first research criteria applied has been the Boolean equation “dry* AND renewable energy”. The records processed and networked with the visualization tool VOSviewer are shown in the Fig. 1. Indeed, the network is built with keywords matching full coincidence at least 3 documents of a total of 3472 present in a set of 2000 relevant records identified by the search engine. - In fact the network shown in Fig. 1 is composed of 539 nodes (keywords), and its visualization was improved by configuring some parameters in the software, such as the limit of components within a group to identify thematic axis. Biomass processing (blue), energy systems (green) and performance (yellow) were the central concepts identified by this basic research. The group of red bubbles is most disperse and seems to be more related with environmental issues. Then, the research for the context of our work would be centered in the findings of blue and green sets of bubbles, which are linking aspects of biomass processing and technological issues; namely, the kind of perspective need to address the development of biomass drying process. Specifically, the green group, enables visualizing topics of interest, such as solar drying, dehydration of the bio-dough or food, being linked to notions of performance, energy, simulation, water heating, among others. So it can be hypothesized that the drying of food is coupled with other heat requirements, for optimal use of energy.
Fig. 1. Keyword network (First search criteria: 539 nodes, 4 groups, 9833 links)
170
A. D. Rincón-Quintero et al.
Starting from the analysis of keywords and the graphical visualization of the basic network (Fig. 1), a new equation was elaborated, by deepening the search engine as follows: “dryer * AND (model * OR system * OR design) AND renewable energy AND solar”. This equation allowed concentrating the research to nearly 100 documents, dealing with dryer’s investigations, it is important to highlight the inclusion of the “Solar” word, considering that it is the main mechanism implemented for drying food or biomass. The keywords analysis by using the VOSviewer tool is summarized by the Fig. 2. In this network, it is possible to observe that “dryers” word has greater centrality, as well systems, energy and performance concepts. It makes possible to point out that the researchers remain focused on improving the energy use of drying processes. The network identifies ancillary topics such as air heating, efficiency, simulation, design, solar dryers, collectors, performance analysis, among others that are topics of special interest for this bibliographic research and the subsequent research.
Fig. 2. Keywords (final document record: 69 nodes, 3 groups, 478 links)
Over the same network, VOSviewer enables another type of appraisal, named “temporary analysis”; this resource allows identifying new trends in research and topics engaging academic and scientific interests. Figure 3 displays the “temporary analysis” derived from the network initially presented in Fig. 2. More recent interests are closer to topics related with exergetic analysis, thermal storage, simulations, drying kinetics, among others yellow-highlighted in Fig. 3. These subjects seek for drying efficiency
Trends in Technological Advances in Food Dehydration
171
improvement pathways, for instance, the implementation of thermal storage devices and heat exchange by phase changes.
Fig. 3. Temporary keyword network
Other analysis tool available in VOSviewer is the called “bibliographic coupling”. By means this option, it is possible to find out documents having, both the highest number of citations, as well the stronger links; this later characteristic makes reference to documents being published recently, obviously with a lower number of citations than other being published for longer, but that are receiving citations by other publications within the same group or similar period. Figure 4 presents the results of the “bibliographic coupling” application. Taking the advantage of this display, it is possible to select each of the nodes to make detailed reading of the documents of interest, to give it an appropriate order by starting with those documents with the greater link strength. In addition, Table 1 has been prepared to summarize the results of this analysis. Review papers about solar dryers for agricultural applications [10, 11] are the documents with the highest strength of links. Classifications and components of dryers, as well parameters that affect the quality of the products, are some of the topics developed therein. Advantages, disadvantages and limitations of such technologies are also included in these key references and consequently, they represent a stock of useful and valuable information for developing research on drying in the sector of cocoa production, which is the final objective of this exploratory research.
172
A. D. Rincón-Quintero et al.
Fig. 4. Bibliographic coupling network (73 nodes, 3 groups, 539 links)
Table 1. List of documents ordered by link strength. Author
Citations
Strength of ties
Documents
El hage (2018)
17
171
[10]
Lamidi (2019)
19
145
[11]
Tiwari (2018)
13
144
[12]
Kumar (2016)
88
129
[7]
Pirasteh (2014) 71
110
[13]
Prakash (2013) 68
102
[14]
18
82
[15]
Ferreira (2014)
8
71
[16]
Atalay (2019)
13
61
[17]
Perea-moreno (2016)
15
60
[18]
Sahota (2017)
3 Results The exploratory review of technological advances in food drying technology, with major options to be scaled up or implemented in drying of cocoa beans, was carried out on the basis of previous reviews [7, 10, 19–21] and [11] and the critical analysis of the selected references presented in Table 1. Applications of solar drying of agricultural biomasses in developing countries, is the common subject of these reviews; consequently, specific issues dealing with the proper implementation of these devices to guarantee the product quality and reach the expected economic and environmental benefits in a context of
Trends in Technological Advances in Food Dehydration
173
limited industrialization for food production, are conveniently and repeatedly treated through these documents. The identified perspective for the classification of solar dryers in these documents is applied here and it is mostly based on the implemented technique for drying food or agricultural biomasses, which can vary on the heat source and heat transfer mechanisms (e.g. natural or forced convection) as more relevant aspects. 3.1 Classification of Dryers In the work carried out by [7] the dryers are classified into three groups, which are direct, indirect and hybrid solar dryers, the latter being the core for these authors. Direct Solar Dryers. In direct solar drying, solar radiation is responsible of direct heating to evaporate water contained in the biomass by natural or forced convection. It can be described as an isolated drying chamber that is commonly covered by glass or transparent plastic, in order to allow the passage of solar radiation and protect the product, and usually it also the chamber is provided with holes to facilitate air circulation within it [22]. To favor the drying conditions, it is common to find out the implementation of fans to generate a forced-air flow within the chamber. The incident radiation crosses the enclosure, to heat the biomass particles inside it and evaporates a share of its water content. Part of this radiation is also reflected back to the atmosphere. The advantages of the direct solar dryer are its economy and simple construction; these allow protecting the product from pollution, animals, dust, smoke among others external factor that can affect its flavor, aroma and other quality parameters. In contrast, these systems tend to generate excessive temperatures, that negatively can affect the quality of the products. Additionally, the capacity for drying is limited [23] (Fig. 5).
Fig. 5. Direct solar drying
Within the bibliography the term passive and active drying is also commonly used, as equal terms to natural and forced convection respectively. Greenhouses are a clear example of this type of drying, as outlined elsewhere [18]. Here, the authors analyze the differences between drying by making piles of biomass chips inside the greenhouse against direct solar drying of a single pile of these chips. Temperatures of up to 25 °C higher than the temperature of direct drying devices and 20% more humidity reduction are reached for greenhouses. Indirect Active Solar Dryers. The main characteristic of these systems is to have a solar collector separated from the drying chamber, which is why it is necessary to include
174
A. D. Rincón-Quintero et al.
in their designs, ducts for the passage of a forced air flows by a fan, these being 4 the components of these dryers. With the separate collector unit, it is possible to achieve higher drying temperatures and control of the air flow through ventilation. Then it is possible to achieve higher energy efficiencies and better moisture removal as long as the design is optimal [24]. In general, the collectors are built with wood, due to its thermal insulation and accessibility, and by adding a metallic sheet as an heat absorbent coating the expected temperatures would be reached [22, 25, 26]. Hybrid Solar Dryer. Specifically, it refers to the incorporation of two or more sources of energy for the drying process, having solar energy as the main source. This concept was introduced as an opportunity to generate continuous drying in the time-span where solar energy is not available (e.g. cloudy days or nights). Among the different alternative sources that are implemented for continuous drying is the use of waste biomasses, natural gas, fuels, geothermal energy, among others. Biomass represents an important alternative considering its availability in rural areas and its low cost of implementation [27–29]. Different researches point out that the dryers under forced convection and implementing different sources of energy, to generate a constant drying over time, are able to greatly exceed the speed and quality of drying compared to direct drying. Due to the characteristics previously described, hybrid dryers are profiled to be implemented in areas with low solar radiation and high humidity. Taking into account that the hot air collector or solar collector is a unit of vital importance in indirect drying, the incorporation of improvements leads to an increase in drying performance. In addition, the implementation of alternative energy sources such as biomass usually requires modelling to predict a better control of variables, such as air flow and temperature, if they are implemented. Other type of hybrid systems integrates a photovoltaic thermal greenhouse system to heat sludge in a plant located at IIT Delhi, India which produces biogas [30]. Mixed Solar Dryers. From other approach the classification of dryers is extended by adding the “mixed solar” dryers type, which can basically be summarized as a combination between a direct and an indirect dryer [10]. In these the coating in the drying chamber is transparent. The solar collectors preheat the air to be entered into the drying chamber under forced convection, while the solar radiation increases the temperature of the feedstock. This type of system uses to have a highest drying performance compared to single indirect and direct dryers, amount other characteristics, because they can exceed ambient temperature in a short time, which contributes to reach the required humidity levels. Due to its complexity, it stands out for its high implementation costs. An example of this type is the work was carried out previously [31], by developing a high-performance solar dryer by incorporating a collector into a greenhouse. In other work, researchers run drying tests to analyze the behavior of the integrated collector. This identifies that the collector has a good performance; by having an efficiency close to 0.6 ± 0.05. This same system is implemented for the analysis of grape drying [32] achieving efficient drying in 128 h. 3.2 Technological Advances in Drying Since the analysis of the network of keywords has been possible to identify a growing interest in the development of more efficient dryers, particularly applying exergy analysis
Trends in Technological Advances in Food Dehydration
175
[33, 34], and other researches in the field of thermal storage, simulation, heat exchange, and drying kinetics. Specifically a previous work [11] established the main methods for developing of solar drying processes complemented by natural or forced air convection) to their application in rural areas [35, 36]. By addressing recent research on hybrid drying systems, two works that integrate geothermal energy into conventional drying systems stand out, the first of which is carried out by [37], which implements a heat supply technique, by implementing an exchanger double-level tubular heat source, using geothermal water as supply, which is built in the LENREZA laboratory. The authors deliberately place the heat exchanger on the absorbent plate in front of the holes to allow air penetration. For the experimental tests, water is adjusted to 70 °C to mimic geothermal water present in the southern regions of Algeria. Authors greatly improve the performance of solar drying by coupling the heat exchanger using geothermal energy; drying were between 46 and 58 °C. Once the sun goes down, geothermal energy keeps stable the drying temperature overnight in the dry chamber at 46 °C. Authors highlighted the continuity of drying over time, which can be maintained at night and in cloudy weather. The second relevant work [4] presents a design and numerical optimization of a dryer that implements a solar collector, connected to a system of pipes installed on the ground, whose function is air preheating. The numerical simulation shown a significant efficiencies increase close to 20.5% with the PA-FPSC hybrid solar-geothermal dryer, compared to a flat-plate solar collector. The authors aim to contribute to the development of sustainable technologies that favor the improvement of the food industry without negatively affecting climate change caused by the implementation of fossil fuels and indicate that this technology contributes to this purpose. However, due to the dependence on the sun energy availability to weather conditions, it is recurrent to generate operational delays, therefore hybrid drying, appears as an economic alternative to provide continuous heating services. In this sense, hybrid drying especially addresses thermal uses of surplus waste agricultural biomasses as the most feasible alternative to be implemented in farms and plantations of crops and other commercial species (i.e. cocoa, coffee, etc.). 3.3 Thermal Energy Storage Another important factor in the advancement of dryers is the incorporation of thermal energy storage, during the hybrid drying process the sun and biomass, which are the main alternatives for renewable drying, are unstable in nature, with the implementation of materials that have the capacity to store energy by increasing its temperature to later transfer it to the air when the main sources are absent. Under this concept there are two types of storage that have been implemented, sensible heat and latent heat, the latter being the most interesting for researchers, due to the advantage of the phase change of the materials for the process absorption and release of energy [38–41]. Within the research on thermal storage applied in food drying, it is highlighted that paraffin has properties according to the drying needs, among which it stands out, its low cost, its safety, reliability and without generating corrosion. For this reason, it still plays an important role in the development of these technologies, because with it an efficient use of thermal energy has been achieved and it maintains stability at drying temperatures [42, 43].
176
A. D. Rincón-Quintero et al.
Among the most recent advances that have been documented on dryers with thermal storage, is the work of [17, 44] in which the drying behavior of orange slices is analyzed in detail. The results obtained by the researchers that achieve a decrease in humidity from 93% to 10% are highlighted, the energy consumption during the experiments was close to 88.5 MJ, with this they calculate a humidity decrease rate close to 0.5292 and 0.5049 kg * KWh−1 . They also calculate the exergetic efficiency of the system during hours with solar radiation, which varies from 50.18 to 66.58% and the efficiency for those cases in which thermal energy storage is used exergy of the drying process, ranges between 54.71 and 68.37%. With these results, the authors conclude that the incorporation of drying systems with thermal energy storage significantly increases the efficiency of solar dryers. In fact, recent advances not only allow to take advantage of drying, with renewable energy for agricultural products in rural areas, this concept is also applied at an industrial level, as is the case of the research carried out by [45]. Whose system is used for the drying of wood and within its object, is the improvement of efficiency through the incorporation of thermal energy storage and air recirculation, as a methodology they study the thermal parameters, drying times and consumption of Energy. Due to the seasons and climatic changes, the results vary depending on the month, which is why the researchers achieve shorter drying times in June than in December. Drying times improve as the area of the solar collector increases and the thickness of the wood decreases. With the incorporation of the air heater, efficiencies close to 0.88 are achieved, having great value for the development of the industry. The storage of thermal energy manages to reduce drying times by up to 50% which varies throughout the year. The simultaneous combination of recirculation and thermal storage manages to greatly reduce energy consumption by up to 50% depending on the time of year. These types of implementations are a clear example of the improvement achieved by incorporating sustainable energy systems such as solar collectors to improve heat processes. Paraffin stands out thanks to its melting temperature close to 60 °C, so it is within the optimal temperature ranges for drying, this is something that remains constant in research, as is the example of [46]. In this work the methodology for the design of a system that implements thermal energy storage in a Hohenheim type tunnel dryer is exposed. Among its results, they highlight the importance of different variables such as the heat transfer area of the storage material with respect to the air to be heated, the diameter of the tube, the total mass of the paraffin, among others, giving recognition to the improvement capacity of the implementation of this type of methodologies.
4 Conclusions and Recommendations Novel displays of bibliometric networks facilitate data analysis, showing the main links and strengths that keywords have and whole available literature about a specific topic and for instance it enables to identify technologies in early stages of development. These emergent methodologies for the visualization of bibliometric networks is openly recommended in order to facilitate researchers’ understanding of the progress of a certain topic (e.g. Drying technologies applied to the processing of cocoa beans). That enables to refine literature review procedures in order to discover new routes for technological development or the improvement of their efficiency, among others.
Trends in Technological Advances in Food Dehydration
177
Currently, conventional open dryers with direct solar radiation from is the broader used technology in the field of drying process but it does not represent any interest for the generation of quality products, due to its low precision and efficiency, in spite of its low cost. Therefore, the implementation of sustainable drying systems such as indirect drying with solar collectors, hybrid and mixed dryers with renewable energies, such as biomass and solar, are the most promising technologies for agricultural applications in developing countries. Likewise, thermal storage systems represent an important opportunity for the development of drying systems, particularly to extend drying times and stabilize the temperature inside the chambers, which would not be depending on environmental changes, such as cloud cover. Modeling and simulation are the specific research areas more explored in order to develop self-sustainable drying systems, particularly those involving alternative energies such as solar, biomass and geothermal energy sources. The identification of critical points generating the greatest loss of energy and guarantee an efficient use of renewable resources are the main goals in this approach. In the specific modeling of solar-biomass hybrid technologies, that seem to be the most adaptable for agricultural contexts, it would be considered the integration of the thermal storage system, the instantaneous dehumidification of the air and a simple thermodynamic variable control system, which can be implemented in the drying of cocoa beans continuously, reducing the drying time by up to 70%, bringing with it competitiveness, profitability and efficiency in the use of energy resources for small and medium producers of this fruit in emergent countries Due to the complexity in controlling the temperature by biomass combustion, research and testing of drying systems incorporating solid plant material or biogas is recommended. The use of these resources represents one of the greatest opportunities within a circular economy in which waste is used and generates extra value in the generation of quality products.
References 1. van Eck, N.J., Waltman, L.: Software survey: VOSviewer, a computer program for bibliometric mapping. Scientometrics 84(2), 523–538 (2010). https://doi.org/10.1007/s11192-0090146-3 2. Gil, M., et al.: Effect of temperature on the formation of acrylamide in cocoa beans during drying treatment: an experimental and computational study. Heliyon 6(2), e03312 (2020). https://doi.org/10.1016/j.heliyon.2020.e03312 3. Rincón-Quintero, A.D., et al.: Manufacture of hybrid pieces using recycled R-PET, polypropylene PP and cocoa pod husks ash CPHA, by pneumatic injection controlled with LabVIEW Software and Arduino Hardware. In: IOP Conference Series: Materials Science and Engineering, vol. 844, no. 1 (2020). https://doi.org/10.1088/1757-899X/844/1/012054 4. Ananno, A.A., Masud, M.H., Dabnichki, P., Ahmed, A.: Design and numerical analysis of a hybrid geothermal PCM flat plate solar collector dryer for developing countries. Sol. Energy 196, 270–286 (2020). https://doi.org/10.1016/j.solener.2019.11.069 5. Sharma, A., Chen, C.R., Vu Lan, N.: Solar-energy drying systems: a review. Renew. Sustain. Energy Rev. 13(6–7), 1185–1210 (2009). https://doi.org/10.1016/j.rser.2008.08.015 6. Ascanio-Villabona, J.G., et al.: Incidence of corrosion in low voltage electrical conductor. In: IOP Conference Series: Materials Science and Engineering, vol. 844, no. 1 (2020). https:// doi.org/10.1088/1757-899X/844/1/012055
178
A. D. Rincón-Quintero et al.
7. Kumar, M., Sansaniwal, S.K., Khatak, P.: Progress in solar dryers for drying various commodities. Renew. Sustain. Energy Rev. 55, 346–360 (2016). https://doi.org/10.1016/j.rser. 2015.10.158 8. FEDECACAO: Guía técnica del cultivo de cacao (2012) 9. Cárdenas-Arias, C.G., et al.: Elasticity modulus variation of the AISI SAE 1045 steel subjected to corrosion process by chloride using tension test destructive. In: IOP Conference Series: Materials Science and Engineering, vol. 844, no. 1 (2020). https://doi.org/10.1088/1757899X/844/1/012059 10. El Hage, H., Herez, A., Ramadan, M., Bazzi, H., Khaled, M.: An investigation on solar drying: a review with economic and environmental assessment. Energy 157, 815–829 (2018). https:// doi.org/10.1016/j.energy.2018.05.197 11. Lamidi, R.O., Jiang, L., Pathare, P.B., Wang, Y.D., Roskilly, A.P.: Recent advances in sustainable drying of agricultural produce: a review. Appl. Energy 233–234, 367–385 (2019). https://doi.org/10.1016/j.apenergy.2018.10.044 12. Tiwari, S., Agrawal, S., Tiwari, G.N.: PVT air collector integrated greenhouse dryers. Renew. Sustain. Energy Rev. 90, 142–159 (2018). https://doi.org/10.1016/j.rser.2018.03.043 13. Pirasteh, G., Saidur, R., Rahman, S.M.A., Rahim, N.A.: A review on development of solar drying applications. Renew. Sustain. Energy Rev. 31, 133–148 (2014). https://doi.org/10. 1016/j.rser.2013.11.052 14. Prakash, O., Kumar, A.: Historical review and recent trends in solar drying systems. Int. J. Green Energy 10(7), 690–738 (2013). https://doi.org/10.1080/15435075.2012.727113 15. Sandá, A., Moya, S.L., Valenzuela, L.: Modelling and simulation tools for direct steam generation in parabolic-trough solar collectors: a review. Renew. Sustain. Energy Rev. 113, 109226 (2019). https://doi.org/10.1016/j.rser.2019.06.033 16. Ferreira, A.G., Gonçalves, L.M., Maia, C.B.: Solar drying of a solid waste from steel wire industry. Appl. Therm. Eng. 73(1), 104–110 (2014). https://doi.org/10.1016/j.applthermaleng. 2014.07.047 17. Atalay, H.: Performance analysis of a solar dryer integrated with the packed bed thermal energy storage (TES) system. Energy 172, 1037–1052 (2019). https://doi.org/10.1016/j.ene rgy.2019.02.023 18. Perea-Moreno, A.J., Juaidi, A., Manzano-Agugliaro, F.: Solar greenhouse dryer system for wood chips improvement as biofuel. J. Clean. Prod. 135, 1233–1241 (2016). https://doi.org/ 10.1016/j.jclepro.2016.07.036 19. Chel, A., Kaushik, G.: Renewable energy for sustainable agriculture. Agron. Sustain. Dev. 31(1), 91–118 (2011). https://doi.org/10.1051/agro/2010029 20. Panwar, N.L., Kaushik, S.C., Kothari, S.: A review on energy and exergy analysis of solar dying systems. Renew. Sustain. Energy Rev. 16(5), 2812–2819 (2012). https://doi.org/10. 1016/j.rser.2012.02.053 21. Singh, P., Shrivastava, V., Kumar, A.: Recent developments in greenhouse solar drying: a review. Renew. Sustain. Energy Rev. 82, 3250–3262 (2018). https://doi.org/10.1016/j.rser. 2017.10.020 22. Romuli, S., Schock, S., Nagle, M., Chege, C.G.K., Müller, J.: Technical performance of an inflatable solar dryer for drying amaranth leaves in Kenya. Appl. Sci. 9(16), 3431 (2019). https://doi.org/10.3390/app9163431 23. Vergara, S., Hadzich, M., Tipula, R., Perez, J., Lopez Vasquez, E., Herrera, E.: Thermal analysis and validation of a geodesic dome dryer for capsicum baccatum, p. 10 (2019). https:// doi.org/10.18086/eurosun2018.08.06 24. Camas-Nafate, M.P., Alvarez-Gutiérrez, P., Valenzuela-Mondaca, E., Castillo-Palomera, R., Perez-Luna, Y.D.C.: Improved agricultural products drying through a novel double collector solar device. Sustainability 11(10), 2920 (2019). https://doi.org/10.3390/su11102920
Trends in Technological Advances in Food Dehydration
179
25. Azouma, Y.O., Drigalski, L., Jegla, Z., Reppich, M.: Indirect convective solar drying process of pineapples as part of circular economy strategy. Energies 12, 2841 (2019). https://doi.org/ 10.3390/en12152841 26. Nukulwar, M.R., Tungikar, V.B.: Thin-layer mathematical modeling of turmeric in indirect natural conventional solar dryer. J. Sol. Energy Eng. 142(4), 041001 (2020). https://doi.org/ 10.1115/1.4045828 27. Okoroigwe, E.C., Eke, M.N., Ugwu, H.U.: Design and evaluation of combined solar and biomass dryer for small and medium enterprises for developing countries. Int. J. Phys. Sci. 8(25), 1341–1349 (2013). https://doi.org/10.5897/IJPS2013.3937 28. Dhanushkodi, S., Wilson, V.H., Sudhakar, K.: Design and thermal performance of the solar biomass hybrid dryer for cashew drying. Facta Univ. Mech. Eng. 12(3), 277–288 (2014). https://casopisi.junis.ni.ac.rs/index.php/FUMechEng/article/view/564/270 29. Dhanushkodi, S., Wilson, V.H., Sudhakar, K.: Life cycle cost of solar biomass hybrid dryer systems for cashew drying of nuts in India. Environ. Clim. Technol. 15(1), 22–33 (2015). https://doi.org/10.1515/rtuect-2015-0003 30. Tiwari, S., Tiwari, G.N.: Thermal analysis of photovoltaic thermal integrated greenhouse system (PVTIGS) for heating of slurry in potable biogas plant: an experimental study. Sol. Energy 155, 203–211 (2017). https://doi.org/10.1016/j.solener.2017.06.021 31. Azaizia, Z., Kooli, S., Elkhadraoui, A., Hamdi, I., Guizani, A.A.: Investigation of a new solar greenhouse drying system for peppers. Int. J. Hydrog. Energy 42(13), 8818–8826 (2017). https://doi.org/10.1016/j.ijhydene.2016.11.180 32. Hamdi, I., Kooli, S., Elkhadraoui, A., Azaizia, Z., Abdelhamid, F., Guizani, A.: Experimental study and numerical modeling for drying grapes under solar greenhouse. Renew. Energy 127, 936–946 (2018). https://doi.org/10.1016/j.renene.2018.05.027 33. Boulemtafes-Boukadoum, A., Benzaoui, A.: Energy and exergy analysis of solar drying process of mint. Energy Procedia 6, 583–591 (2011). https://doi.org/10.1016/j.egypro.2011. 05.067 34. Fernández, A., Dieste, J.A.: Low and medium temperature solar thermal collector based in innovative materials and improved heat exchange performance. Energy Convers. Manag. 75, 118–129 (2013). https://doi.org/10.1016/j.enconman.2013.06.007 35. Lowder, S.K., Skoet, J., Raney, T.: The number, size, and distribution of farms, smallholder farms, and family farms worldwide. World Dev. 87, 16–29 (2016). https://doi.org/10.1016/j. worlddev.2015.10.041 36. Ascanio-Villabona, J.G., et al.: Building a prototype for functional analysis of the energy potential of the water flow in pipe 1/2 “using microturbines applied to Unidades Tecnológicas de Santander. In: IOP Conference Series: Materials Science and Engineering, vol. 844, no. 1, p. 012056 (2020). https://doi.org/10.1088/1757-899X/844/1/012056 37. Sandali, M., Boubekri, A., Mennouche, D., Gherraf, N.: Improvement of a direct solar dryer performance using a geothermal water heat exchanger as supplementary energetic supply: an experimental investigation and simulation study. Renew. Energy 135, 186–196 (2019). https://doi.org/10.1016/j.renene.2018.11.086 38. Sharma, A., Tyagi, V.V., Chen, C.R., Buddhi, D.: Review on thermal energy storage with phase change materials and applications. Renew. Sustain. Energy Rev. 13(2), 318–345 (2009). https://doi.org/10.1016/j.rser.2007.10.005 39. Sreerag, T.S., Jithish, K.S.: Experimental investigations of a solar dryer with and without multiple phase change materials (PCM’s). World J. Eng. 13(3), 210–217 (2016). https://doi. org/10.1108/WJE-06-2016-028 40. Bhagwat, V.V., Salve, S.P., Debnath, S.: Experimental analysis of a solar dehydration with phase changing material. In: AIP Conference Proceedings, vol. 1998, no. 1, p. 20003 (August 2018). https://doi.org/10.1063/1.5049099
180
A. D. Rincón-Quintero et al.
41. Romero-Tarazona, B.E., et al.: Development of an artificial vision system that allows nondestructive testing on flat concrete slabs for surface crack detection by processing of digital images in MATLAB. In: IOP Conference Series: Materials Science and Engineering, vol. 844, no. 1 (2020). https://doi.org/10.1088/1757-899X/844/1/012058 42. Shalaby, S.M., Bek, M.A.: Experimental investigation of a novel indirect solar dryer implementing PCM as energy storage medium. Energy Convers. Manag. 83, 1–8 (2014). https:// doi.org/10.1016/j.enconman.2014.03.043 43. Reyes, A., Mahn, A., Vásquez, F.: Mushrooms dehydration in a hybrid-solar dryer, using a phase change material. Energy Convers. Manag. 83, 241–248 (2014). https://doi.org/10.1016/ j.enconman.2014.03.077 44. Atalay, H., Turhan, M., Olcay, K.: Modeling of the drying process of apple slices: application with a solar dryer and the thermal energy storage system. Energy 134, 382–391 (2017). https:// doi.org/10.1016/j.energy.2017.06.030 45. Khouya, A.: Effect of regeneration heat and energy storage on thermal drying performance in a hardwood solar kiln. Renew. Energy 155, 783–799 (2020). https://doi.org/10.1016/j.ren ene.2020.03.178 46. Escobar, M., Cuervo-Andrade, S., Rincon-Prat, S.: Methodology for the design of a thermal energy storage module for a solar tunnel dryer using phase change materials (PCM). Rev. UIS Ing. 13(1), 9–20 (2018). https://doi.org/10.18273/revuin.v17n1-2018001
Analysis of the Energy Potential of a Tangential Microturbine for Application in a Passivhaus Environment J. G. Ascanio-Villabona1,2(B) , L. A. Del Portillo-Valdés2 , O. Lengerke-Pérez1 B. E. Tarazona Romero1,2 , A. D. Rincón-Quintero1,2 , and M. A. Durán-Sarmiento1,2
,
1 Unidades Tecnológicas de Santander UTS, Bucaramanga, Santander 680005, Colombia
{jascanio,Olenguerke,btarazona,arincon}@correo.uts.edu.co 2 University of the Basque Country UPV/EHU, 48013 Leioa, Bizkaia, Spain [email protected], [email protected]
Abstract. In this document, an analysis was made of the energy potential of the flow of water in pipes, of an innovative prototype for a building certified with the standard Passivhaus (PH). Passive interventions are based on insulation through the housing envelope and implementing technologies according to criteria and principles of the standard, thus increasing their energy efficiency [1], however, in most of the projects carried out, the concept of “zero energy building”, with energy savings of approximately 80% [2], therefore, the need arises to improve energy performance, through the use of tangential micro turbines with low pressure drop, constituting a model on a small scale for the study of electric generation thereof, establishing operating parameters similar to a hydraulic system for the distribution of residential drinking water. In order to take advantage of electrical energy, which is obtained by transferring flow through a micro turbine, increasing the percentage of performance in passive construction. Keywords: Passivhaus standard · Efficiency · Generation · Hydraulic system · Microturbine
1 Introduction According to the International Energy Agency IEA, 40% of primary energy consumption corresponds to the residential sector [3], where in many climatic situations current technology may not be appropriate for the conditioning of these spaces [4]. Therefore, increasing energy efficiency, reducing electricity consumption and maintaining interior thermal comfort in buildings [5], have become a high potential, to comply with environmental agreements [6]. Therefore, there are standards that promote energy efficiency and the use of renewable energy in buildings [7]. One of the standards that fits with the concept of “zero energy building” is the Passivhaus standard [8], since it consists of a construction with almost zero energy consumption and high internal comfort, based mainly on minimizing this energy operational © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 181–195, 2021. https://doi.org/10.1007/978-3-030-72212-8_14
182
J. G. Ascanio-Villabona et al.
in buildings, through the use of more advanced insulation materials and equipment [9]. However, certified homes already built according to Passivhaus requirements, can reduce energy demand between 70–90% [10]. In order to achieve a zero-energy upgrade, that is, to achieve 100% savings in energy consumption, studies are carried out in order to find an alternative that can generate electricity in the midst of daily activities. The installation of microturbines in a water system in spaces where water flows naturally [11] without the intervention of a pumping system [12], could produce electricity [13], from the moment there is consumption [14]. There are different studies on power generation from microturbines. In the context of the technical feasibility of a hybrid solar microturbine, Chahartaghi and Baghaee [15] investigated a new cycle of solar gas microturbine combined cooling, heating and power (CCHP) systems for a residential building in Iran. They estimated the performance of the system and the cost using the method the levelized cost of electricity (LCoE). One of the characteristics of the gas microturbine is its ability to combine with solar energy. By adding a solar receiver to available micro gas turbines, solar energy can provide some of the required cycle heat. The components of this cycle included: micro gas turbine, solar dishes (parabolic dishes), solar receiver, absorption chiller, auxiliary boiler and heat recovery steam generator. The authors investigated 49 different methods by changing the number of microturbines and the surface area of the parabolic dishes in 10 economic scenarios. One of the benefits of this cycle is the generation of less pollutants, greater reliability and less dependence on fossil fuels, obtaining an efficiency of approximately 55%. Wang et al. [16] presented a new integrated design with a 25 kWe hybrid solar microturbine. In their proposed prototype, the receiver, combustion chamber and turbine have been integrated to improve efficiency and minimize the size of the system. Arroyo et al. [17], presented a linear rotor dynamic analysis and a comparison of three mechanical arrangements of a 6 kW micro gas turbine (MGT) intended to use concentrated solar energy (CSP) using a parabolic dish concentrator. Giostri and Macchi [18], studied the effect of increasing the turbine inlet temperature on cycle efficiency for an integrated solar panel and microturbine system. To withstand high temperatures, they proposed a ceramic turbine, obtaining an annual solar to electrical efficiency of 26.4%. On the other hand, Vargas, Velásquez and Torres [19], designed the prototype of an electric hydrogenerator. The design, construction and commissioning of the prototype was carried out on the Guatiquía River, which allowed transforming the kinetic energy of this water source into mechanical energy and generated electrical energy with satisfactory results of 12 V to 1 kW of electrical power. Among the studies that have been carried out so far on the use of microturbines in terms of electricity generation, there is no application for residential buildings in the water system. In this way, this article presents and evaluates the feasibility of implementing an innovative sanitary water system, in a house based on the Passivhaus standard, by analyzing the behavior of the microturbine and its variables involved in the study.
Analysis of the Energy Potential of a Tangential Microturbine
183
2 Method The domestic energy demand for a typical residential unit consists mainly of electrical energy (for lighting, electrical appliances), thermal energy, (for hot water and in the winter season for space heating) [20]. Passivhaus houses are based on making constructions that have excellent thermal insulation, which controls air infiltration through the envelope and optimal indoor air quality, among other aspects. These measures, in addition to the use of solar energy, can reduce energy consumption by 70% compared to traditional buildings [21]. Regarding the above, the study addresses a descriptive research methodology [22], because it intends to perform the analysis of the energy potential of the water flow in a ½ pipe in the construction of an innovative prototype, by using a micro turbine, where later a quantitative investigation will be carried out, collecting variables involved in the study such as current, voltage, power, ON-LINE, OFF-LINE [23], in order to study the feasibility of its implementation in a house based on the Passivhaus standard.
3 Prototype Description and Materials The study house has a water design built on a pipe with a diameter of D = ½ PVC, under the ICONTEC 1500 [24] technical standards for plumbing. In the selection process of the materials and in the development of the prototype, the main parameters such as the flow, pressure presented by a hydraulic system of the study house, were also analyzed: • • • •
Inlet and outlet pressures of the microturbine. The output flow of the motor pump. The recirculation of the water to avoid its waste. The construction of the base or support for the installation of the tank and the motor pump, as well as • PVC pipe, ball valves, universal couplings, elbows, reductions, unions, tees. The design of the prototype was carried out by means of the computer-aided CAD program, as shown in Fig. 1. CAD prototype, successively in Table 1. Main elements of the prototype, the main components used in the system are described. Motor Pump The system consists of a Ranger QB60 motorized pump with a 1 outlet diameter, shown in Fig. 2, which generates a maximum pressure of 50 psi and a maximum flow of 35 L/min, complying with the NTC-1500 standard, where they indicate the range of pressures in conventional water systems ranging from 25 to 35 psi and Table 2 shows the technical characteristics.
184
J. G. Ascanio-Villabona et al.
Fig. 1. CAD prototype
Microturbine Figure 3 shows a micro turbine with ½ inlet and outlet diameter, brand AIYIMA, model hydroelectric generator microturbine, where the main features are the unique double magnetic coupling clutch, inlet and outlet pipe are available to match with a straight line or L-type right angle mount, ultra-low flow loss, low water pressure start, an isolation diode to supply power directly to the rechargeable battery or supply power to the 5F farad capacitor. On the other hand, the main variables were characterized in Table 3. The mentioned microturbine was selected because it was the most successful model for the following measurements: • • • •
Inlet and outlet diameter of ½ Maximum working pressure greater than or equal to 20 psi Lower flow to start electricity generation Does not require a minimum operating time.
It should be noted that the chosen microturbine is of European origin, since, in the country, there is no available market with the necessary characteristics.
Analysis of the Energy Potential of a Tangential Microturbine
185
Table 1. Main elements of the prototype Element
Description
Function
Motor pump
Ranger 1/2 hp; qMax 35 L/min QB60 Converts electrical energy into velocity and flow pressure, through the impeller or turbine rotating in the housing. Its main objective is to move a fluid on two different levels and take advantage of the pressure it supplies
Micro turbine
Hydroelectric generator microturbine High power generator that could - AIYIMA - 3.5W 12 V guarantee 60 light emitting diodes (LEDs)
Pressure gauge Glycerin ø63-0-60 PSI
Also called pressure gauges. They are instruments that measure the pressure of fluids (liquids, gases) in closed circuits. Measuring the difference between real or absolute pressure and atmospheric pressure
Valve
They are used to prevent the return of a fluid within a line, since this valve closes instantly, allowing only the flow that flows in the right direction to pass, so they are also called non-return valves
CHECK vertical de 1 Pulg
Fig. 2. Motor pump
Pressure Gauge To obtain variation in the values and subsequently compare the electrical variables, manometers with a range between 0–60 PSI were installed, as shown in Fig. 4, measuring pressure at the outlet of the motor pump, inlet pressure in the turbine and the loss of fluid pressure when passing through it.
186
J. G. Ascanio-Villabona et al. Table 2. Technical specifications of the pump Inlet and outlet diameter
½
Voltage range
9.8 a 18.5 V DC
Current range
120 mA a 260 mA
Minimum operating pressure 0.08 MPA Loss of charge
3.6% a (0.25 MPA)
Minimum operating flow
1 LPM
Maximum output power
5W
Minimum operating time
It does not require
Fig. 3. Micro turbine
Table 3. Technical specifications of the micro turbine Inlet and outlet diameter
½
Output voltage
9.8 V a 18.5 V DC
Current range
128 mA a 260 mA
Pressure range
0.08 MPA a 0.55 MPA
Maximum pressure
0.55 MPA
Traffic loss
3.6% (a 0.25 MPA)
initial flow
1 L/min
Working temperature range 5 °C a 85 °C Dimensions
81.4 mm × 43.8 mm × 80 mm
Maximum output power
5W
Minimum run time
It does not require
External diameter
20 mm
Analysis of the Energy Potential of a Tangential Microturbine
187
Fig. 4. Pressure gauge
Check Valve In order to avoid possible discharge in the motor pump chosen for the study, a check valve shown in Fig. 5 will be used.
Fig. 5. Check valve
In addition to the above, a Bypass is integrated, increasing or decreasing the pressure, depending on the action required at the inlet of the micro turbine, since, if the ball valve opens or closes, there would be a variation in pressure allowing different measures to be tabulated of them and obtain different voltages in the electrical generation. At the end of the tests, a purge is performed to drain the liquid retained in the upper pipe.
4 Experimental Process In the experimental process, two tests were carried out demonstrating the operation and behavior of the hydraulic microturbine, obtaining the best design alternative in the study for the simulation of the water system. 4.1 Characterization of the Prototype – Test 1 The first test was carried out with a manometer attached to the entrance of the micro turbine, as shown in Fig. 6, starting with minimum pressure measurements up to the maximum that the water system of the house can reach.
188
J. G. Ascanio-Villabona et al.
Fig. 6. Assembly Test 1, pressure gauge attached to the microturbine inlet.
According to Fig. 7, to determine the measurements of the voltage generated by the microturbine at a pressure of 10 PSI, a multimeter was used. It was placed at the terminals of the microturbine, obtaining measurements of 13.80 V, which means that the system is optimal for charging a 12 V battery.
Fig. 7. Measurements of generated voltage at a pressure of 10 PSI, Test 1
The trend values are the estimates of the variables with respect to time [25]. There are several types of fitted-trend equations, shown in Table 4. Figure 8 presents the graphic representation of the selected fitted trend equation, shown in the expression Eq. (2). y = −0.0195x2 + 1.125665x + 2.0699 Where: Y: Pressure in Psi X: Voltage in Volts.
(1)
Analysis of the Energy Potential of a Tangential Microturbine
189
Table 4. Types of fitted-trend equations Tipo de modelo
Ecuación
Lineal
yt = b0 + (b1 * t)
Cuadrático
yt = b0 + b1 * t + (b2* t2)
Crecimiento exponencial
yt = b0 + (b1t)
Curva S (logística de Pearl-Reed) yt = (10a)/(b0 + b1 * b2t) Note: The form of the fitted trend equation depends on the type of model selected, where: yt: the variable; b0: The constant; b1 and b2: The coefficients and t: Value of the unit of time [25].
Voltaje vs Presión 25 20 15 10 5 0 0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34
Fig. 8. Graphical representation of voltage vs pressure for behavior tests.
4.2 Characterization of the Prototype - Test 2 In the second test, two manometers are assembled, one at the inlet and the other at the outlet of the microturbine as given in Fig. 9. In this study, the losses generated by flow friction are determined, by therefore, the losses described by the technical sheet provided by the manufacturer are taken as a basis, which are in a range of 3.6% at 0.25 MPA. The test is run under a maximum pressure of 22 psi, performed with air pressure, to obtain greater accuracy. Therefore, in this case the manufacturer’s data would vary. It is worth noting that the inlet of the microturbine is brought to a pressure of 20 psi as seen in Fig. 10. Subsequently, the result of the outlet pressure is shown in Fig. 11, being approximately 20 psi, which means that it is relatively low and that it does not affect the pressures of the sanitary system.
190
J. G. Ascanio-Villabona et al.
Fig. 9. Assembly Test 2 - connect two manometers to the inlet and outlet of the microturbine.
Fig. 10. Inlet pressure in the microturbine
Fig. 11. Taking measurements at the manometer at the outlet of the microturbine.
5 Results To determine the generation of the microturbine, the value of the flow that passes through the pipe and the daily water consumption of the home must be known.
Analysis of the Energy Potential of a Tangential Microturbine
191
Therefore, Eq. (1) is proposed, calculating the estimate of total daily consumption, assuming that 4 people live in the residence and that each person consumes 210 Lt/inhab/day, value taken as a reference from the NTC-1500, Colombian Code Of Hydraulic And Sanitary Installations, for the studio apartment. TCD = Nhab ∗ ECD
Lt hab
(2)
Where: TCD: Total daily consumption Nhab: Number of inhabitants in the home ECD: Estimate of daily consumption per inhabitant in liters. Applying the previous equation, we obtain: TCD = 4 hab ∗ 210
Lt hab
TCD = 840 Lt Likewise, in Table 5 an inventory of the elements and units of expenditure is made, in order to calculate the flow, knowing that the pressure available in the external network is 20 PSI. Table 5. Inventory of components and units of expenditure Component
Quantity
Shower
2
Toilet
2
Handwash
2
Laundry
1
Washing machine 1 Dishwasher
1
Total
9
Based on the data of the characterization of the turbine and taking into account the water consumption per daily house, we proceed to perform Eq. 3, the calculation of electric power generation of the microturbines as follows: Q = V /t Where: Q: Flow
(3)
192
J. G. Ascanio-Villabona et al.
V: volume t: time. Q=
840 Lt 10 h ∗ 60 min
Q = 1.4 lpm Taking into account the flow obtained by housing and contrasting it with the minimum generation flow in the characterization of the microturbine, it can be seen that if energy is generated with the calculated flow. Based on the above, Eq. 4 is used, calculating an approximate generation per turbine if it were implemented in the Passivhaus: P =I ∗V
(4)
Where: P: power I: current V: voltage. Therefore, the current and voltage values are replaced, assuming the average value between their maximum and minimum values. P = 194 mA ∗ 14.15 V P = 2.74 W Subsequently, by means of Eq. 5 the power generated in hour / month per microturbine is calculated. G = P ∗ dm ∗ hQ Where: G: generation P: power dm: days of the month hQ: Flow hours in the day. G = 2.74 W ∗ 30 dias ∗ 10 horas G = 822 Wh al mes
(5)
Analysis of the Energy Potential of a Tangential Microturbine
193
Finally, the pressure losses of each one of the microturbines between the inlet and outlet of them have to be considered, which is negligible as evidenced in the field tests, therefore, a series of microturbines can be implemented in series and It will not affect the pressure of the water flow in the system. Taking into account the above, the calculation is made through Eq. 6 of the total generation to be implemented in the Passivhaus if two microturbines were installed per output component of water flow, which are mentioned in Table 5. GT = G ∗ NC ∗ NM
(6)
Where: GT: total power generated in the home G: power generated in the month NC: Number of components NM: Number of microturbines in series. GT = 822 Wh per month ∗ 9 units ∗ 2 microturbines in series GT = 14.800 Wh per month The electrical power generated by the prototype was equivalent to 822 Wh, however, if the number of microturbines were increased in each output unit, the generated power would be 14.8 kWh per month. Different studies classify generation systems from 0.5 to 5 kW as Pico hydroelectric plants, for which the prototype manufactured applies within this classification. Comparing the energy supplied by the developed system against a little more known systems, such as photovoltaic solar systems, the prototype presents similar characteristics in relation to electricity generation, voltage levels and electrical power, which allows its application in electrical systems low demand. The results presented previously showed the viability of the water system of the microturbines.
6 Conclusions This project presents an innovative prototype for the analysis of the energy potential of the water flow in pipes, the development offers positive results in terms of the generation of electrical energy through the microturbine, providing energy for domestic consumption with standard Passivhaus (PH). Passivhaus homes, despite their trends being 0% external energy consumption, need additional external energy from the PH standard to supply 19.25 KWh per month, that is, 100% of the average single-family energy a house needs 20% of energy, preferably renewable, which would represent 3,850 kWh per month, based on this, the implementation of micro hydro turbines in the home’s clean water system projects a generation and injection of usable electrical energy of 0.822 kWh per month equivalent to 21.31% with respect to the renewable energy required in the PH architecture.
194
J. G. Ascanio-Villabona et al.
References 1. Sierra-Pérez, J., Rodríguez-Soria, B., Boschmonart-Rives, J., Gabarrell, X.: Integrated life cycle assessment and thermodynamic simulation of a public building’s envelope renovation: conventional vs. passivhaus proposal. Appl. Energy 212, 1510–1521 (2018). https://doi.org/ 10.1016/j.apenergy.2017.12.101 2. TOBIAS HATT: EL ESTÁNDAR ‘PASSIVHAUS’ EN EL CENTRO-SUR DE CHILE. UN ESTUDIO PARÁMETRICO MULTIFACTORIAL. TESIS DOCTORAL CONCEPCIÓN (2012) 3. International Energy Agency IEA: Hacia un modelo de edificio energía cero interconectado a la red. En: Encuentro internacional Ekotectura. Towar. net zero energy Sol. Build., pp. 27–29 (2012) 4. O’Kelly, M., Walter, M.E., Rowland, J.R.: Simulated hygrothermal performance of a Passivhaus in a mixed humid climate under dynamic load. Energy Build. 81, 211–218 (2014). https://doi.org/10.1016/j.enbuild.2014.06.015 5. Zhao, D., McCoy, A.P., Du, J., Agee, P., Lu, Y.: Interaction effects of building technology and resident behavior on energy consumption in residential buildings. Energy Build. 134, 223–233 (2017). https://doi.org/10.1016/j.enbuild.2016.10.049 6. CORPORACIÓN AUTÓNOMA REGIONAL DEL TOLIMA: Principales Convenios Internacionales En Materia Ambiental (2018). https://www.cortolima.gov.co/principales-conven ios-internacionales-materia-ambiental 7. European Commission: Energy efficiency directive 2012/27/EU (2012) 8. Directive 2010/31/EU of the European parliament and of the council: Directive 2010/31/EU of the European parliament and of the council of 19 May 2010 on the energy performance of buildings (2010) 9. Finkbeiner, M., Schau, E.M., Lehmann, A., Traverso, M.: Towards life cycle sustainability assessment. Sustainability 2(10), 3309–3322 (2010). https://doi.org/10.3390/su2103309 10. Feist, W., Peper, S., Görg, M.: CEPHEUS-Project information No. 36 (2001). www.passiv.de. Accessed 11 Oct 2020 11. Creutzfeldt, B., Güntner, A., Thoss, H., Merz, B., Wziontek, H.: Measuring the effect of local water storage changes on in situ gravity observations: case study of the geodetic observatory wettzell, Germany. Water Resour. Res. 46(8) (2010). https://doi.org/10.1029/2009WR008359 12. Yahyaoui, I., Tina, G., Chaabene, M., Tadeo, F.: Design and evaluation of a renewable water pumping system. IFAC-PapersOnLine 48(30), 462–467 (2015). https://doi.org/10.1016/j.ifa col.2015.12.422 13. Rashid, M.H., Hussien, Z.F., Rahim, A.A., Abdullah, N.: Electric power transmission. In: Power Electronics Handbook, pp. 829–846. Elsevier, Amsterdam (2018) 14. Vieux, F., Maillot, M., Constant, F., Drewnowski, A.: Water and beverage consumption patterns among 4 to 13-year-old children in the United Kingdom. BMC Public Health 17(1), 1–12 (2017). https://doi.org/10.1186/s12889-017-4400-y 15. Chahartaghi, M., Baghaee, A.: Technical and economic analyses of a combined cooling, heating and power system based on a hybrid microturbine (solar-gas) for a residential building. Energy Build. 217, 110005 (2020). https://doi.org/10.1016/j.enbuild.2020.110005 16. Wang, W., Ragnolo, G., Aichmayer, L., Strand, T., Laumert, B.: Integrated design of a hybrid gas turbine-receiver unit for a solar dish system. Energy Procedia 69, 583–592 (2015). https:// doi.org/10.1016/j.egypro.2015.03.067 17. Arroyo, A., McLorn, M., Fabian, M., White, M., Sayma, A.I.: Rotor-dynamics of different shaft configurations for a 6 KW micro gas turbine for concentrated solar power. In: Proceedings of the ASME Turbo Expo, vol. 8 (September 2016). https://doi.org/10.1115/GT201656479
Analysis of the Energy Potential of a Tangential Microturbine
195
18. Giostri, A., Macchi, E.: An advanced solution to boost sun-to-electricity efficiency of parabolic dish. Sol. Energy 139, 337–354 (2016). https://doi.org/10.1016/j.solener.2016.10.001 19. Vargas Guativa, J.A., Velásquez Clavijo, F., Torres Gómez, C.: Desarrollo del prototipo de un hidrogenerador eléctrico como alternativa de generación de energía limpia en zonas rurales. Universidad Libre Seccional Barranquilla (2016). https://dialnet.unirioja.es/servlet/articulo? codigo=5980557&info=resumen&idioma=SPA. Accessed 11 Oct 2020 20. Ancona, M.A., et al.: Combined heat and power generation systems design for residential houses. Energy Procedia 158(2018), 2768–2773 (2019). https://doi.org/10.1016/j.egypro. 2019.02.036 21. certificados energeticos.com: Obtener certificado energético. Certificado de eficiencia energética (2019). https://www.certificadosenergeticos.com/. Accessed 30 Sep 2020 22. de Cabo, J.V., de la Fuente Díez, E., Verdejo, M.Z.: Modelos de estudios en investigación aplicada: conceptos y criterios para el diseño. Med. Segur. Trab. (Madr) 54(210), 81– 88 (2008). https://scielo.isciii.es/scielo.php?script=sci_arttext&pid=S0465-546X20080001 00011. Accessed 11 Oct 2020 23. McDonald, R.P.: The informative analysis of individual trend curves. Multivar. Behav. Res. 39(3), 517–563 (2004). https://doi.org/10.1207/s15327906mbr3903_5 24. ICONTEC: NTC-1500 CODIGO COLOMBIANO DE INSTALACIONES. Bogota (2017). https://www.aprocof.co/descargas/icontec/PRESENTACION. ICONTEC NTC-1500 2.pdf 25. Interpretar todos los estadísticos y gráficas para Análisis de tendencia - Minitab. https://sup port.minitab.com/es-mx/minitab/18/help-and-how-to/modeling-statistics/time-series/howto/trend-analysis/interpret-the-results/all-statistics-and-graphs/. Accessed 11 Oct 2020
Development of a Fresnel Artisanal System for the Production of Hot Water or Steam B. E. Tarazona-Romero1,2(B) , A. Campos-Celador2 , Y. A. Muñoz-Maldonado3 , J. G. Ascanio-Villabona1,2 , M. A. Duran-Sarmiento1,2 , and A. D. Rincón-Quintero1,2 1 Unidades Tecnológicas de Santander UTS, Bucaramanga, Santander 680005, Colombia
{btarazona,jascanio,mduran,arincon}@correo.uts.edu.co 2 University of the Basque Country UPV/EHU, 48013 Bizkaia, Spain [email protected] 3 Universidad Autónoma de Bucaramanga UNAB, Bucaramanga, Santander 680003, Colombia [email protected]
Abstract. The development of the Fresnel artisanal system or Fresnel linear collector aims to take advantage of solar radiation and local materials, to generate a low-cost alternative for the production of hot water or steam, with scope for integration to desalination systems that require feed water in a range of 60 °C to 95 °C. For the prototype development, the solar radiation in the town Giron, Santander, Colombia was used, later mathematical models were identified to carry out the dimensioning, followed by the selection of local materials based on availability and technical specifications, in order to carry out the assembly and field tests. The system was designed to perform manual solar monitoring counting with three auxiliary systems: solar preheater, pumping system and data acquisition system for temperature, level and flow variables. Keywords: Fresnel linear collector · Solar radiation · Pumping system · Primary and Secondary reflection system
1 Introduction Concentrating solar energy (CSP) converts solar energy into thermal or electrical energy, through the use of mirrors with large areas of solar concentration in a receiver [1, 2]. A fluid circulates inside the receiver that heats up to high temperatures. In general, there are four types of CSP technologies [3, 4]: parabolic trough, Fresnel, tower and disc. Within the 4 types of technologies, the linear Fresnel collector (LFC) is the technology with the least maturity [5], but with the highest development projection [6]; This is due to its simplicity in design, land use, maintenance, and cost. Based on the above, LFC is an attractive system for water heating, steam generation and electricity, depending on the application environment [7]. LFC is composed of a series of rectangular mirrors that make up the reflection area or primary reflection system [8], the sun’s rays hit the mirrors and direct the radiation © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 196–209, 2021. https://doi.org/10.1007/978-3-030-72212-8_15
Development of a Fresnel Artisanal System
197
towards a specific focal point [9]. The latter has two types of technology [10]: multitube receivers, made up of a trapezoidal cavity with a series of parallel tubes connected in series through which the heat transfer fluid of transported [11, 12] or a secondary reflector system, which basically consists of in a tube covered by a reflector that redirects the solar rays that come from the primary reflection system [13]. LFC reaches temperatures up to 500 °C [14] and historically, most LFC systems have been used to generate low or medium temperatures, applied to cooling [15, 16] and heating processes in buildings, heat supply in industrial processes [17], water treatment, desalination [18, 19], etc. Currently, innovative state-of-the-art systems are being built in order to withstand high temperatures, for steam generation or large-scale electricity generation [20]. Since the cost represents an approximate saving of 30% compared to the other types solar concentration and presents less optical losses [7]. The objective of the technological development presented in this article is to develop an LFC system that can produce hot water in a range of 60 to 95 °C to be subsequently integrated into a desalination system. It is important to highlight, that the system will be developed with low-cost and locally accessible materials. On the other hand, the methodological development of the project meets with a strict chronological order of development from the searching, dimensioning, selection of components, assembly and experimentation. Finally, the development of this document begins with the description in Sect. 3 of the methodology where the implemented flow diagrams, the sizing of the system and the development of CAD modeling are presented to carry out the project development as well as the materials used. This section focused on the auxiliary equipment required for the operation of the collector. Apart from that, Sect. 3 presents the experimentation process carried out with the prototype, the initial field tests and the background tests of the system. In Sect. 4 the results obtained in the experimentation are analyzed and finally, Sect. 5 corresponding to the most relevant conclusions of the project is developed.
2 Method and Materials 2.1 Method The present technological development with an investigative nature that is presented in this article, presents a descriptive methodology [21, 22] with a quantitative approach [23]. Figure 1 presents the flow chart with the sequence of activities methodologically carried out to achieve the optimal development of the work. The first step was to identify the location or place, in order to know latitude, longitude and meteorological parameters through open access databases. Subsequently, a search for academic information was carried out to identify the mathematical models that, together with the location data, allowed the sizing of the Fresnel system. Subsequently, the dimensioned Fresnel system was modeled using computer aided design (CAD) software together with the three existing auxiliary systems, which allowed to generate real-scale plans of the parts that make up the system for its construction, through the use of low-cost materials in the local market. Finally, field tests are carried out to determine the functionality of the prototype through the auxiliary data acquisition system and thus determine its efficiency.
198
B. E. Tarazona-Romero et al.
Fig. 1. Workflow diagram
Fig. 2. Data acquisition
It is important to highlight the data acquisition system (See Fig. 2), which used the free access Arduino programming tool or software where the data was processed as follows: Starting with the reception of analog signal input from sensors which takes measure of temperature, pressure and flow. Then it continues with the conversion of analogue signal to digital, taking into account that the instruments used to measure variables provide analogue signal. The next step is to treat and process the digital data for, finally, the
Development of a Fresnel Artisanal System
199
acquisition of data in external memory and indication of values of variables on LDC screen. Sizing To sizing the Fresnel System, it was necessary to identify the place and obtain geographic data. The prototype test site is the town of Girón Santander, Colombia with the following geographic coordinates (Table 1): Table 1. Geographic coordinates Place
Latitude Longitude Elevation
Giron-Santander 7.1
−73.12
930 m
The sizing was started from the following parameters: • Focal Distance 0.75 m (Distance between primary and secondary reflector system) • Area of each (Unit) 0.10 m * 1 m • Trapezoidal secondary reflection system in aluminum, transparent glass and U in 1/4 tube in 2 m copper • Working time range 8 am to 2 pm. For the development of the mathematical models, the ones used by Tarazona-Romero et al. in 2019 [24], and the following parameters were determined (Table 2): Table 2. Sixing parameters Description
Variable
Date to evaluate system Nd
210
Solar tilt
δ
−16.8295
Hour angle
w
−45
Latitude
ϕ
7.1
Solar incidence angle
Cos θ
0.6358
θ
50.51724
Solar height angle
U
89.36415
Test start time
HP
9
Once the values of the variables were calculated, and taking Tarazona-Romero et al. as a reference, who implemented 10 equal mirrors, 10 cm wide and at a distance (Li) of 15 cm each from center to center, the angles of initial inclination of the reflector (βi) and the angle formed between the vertical axis of the receiver and the reflected solar radiation for each reflector (i) as presented in the following Table 3:
200
B. E. Tarazona-Romero et al. Table 3. Primary reflector area parameters Ei E1
Li
βi°
7.5 28.11
i° 5.71
E2
22.5 33.60 16.69
E3
37.7 38.60 26.68
E4
52.5 42.75 34.99
E5
67.5 46.25 41.98
E6
7.5 22.40
5.71
E7
22.5 16.91 16.69
E8
37.7 11.92 26.68
E9
52.5
7.76 34.99
E10 67.5
4.27 41.98
Finally, Fig. 3 presents the geometric distribution of the mirrors or Fresnel prototype primary reflection area.
Fig. 3. Distance between primary reflectors system
3D Modeling The CAD modeling of the Fresnel linear solar collector prototype was developed in Solid-Work software and is presented in Fig. 4. The mirror area is 1 m2 and receives the solar radiation and reflects it to a focal point of concentration that has a trapezoidal cover in aluminum with a glass cover to reduce losses due to conduction, convection and radiation. The mirrors have an individual area of 10 cm wide by 1 m long. The base of the mirrors was made of 1 cm wide aluminum profile and has the same dimensions as
Development of a Fresnel Artisanal System
201
the mirrors; this material allows to reduce the weight of the structure to manually track the sun. The receptacles tube presents a curvature or geometric U-shape, the material is copper with a diameter of 1/4 in., length of 2.1 m, with a theoretical absorption coefficient of 0,98 and a theoretical emissivity coefficient of 0,98. The base of the primary and secondary reflection system was built with a 1 1/2-in. galvanized steel tube or profile.
Fig. 4. Fresnel system CAD modeling
The base of the primary and secondary reflection system was built with a structural tube of 1 1/2 in., by 1.5 mm thick, the structure was fixed through screws and plates of 8 cm * 8 cm * 3 mm, facilitating the clearance. Additionally, two aluminum rails were used to mount the reflectors, because they provide easy sliding of the reflector supports. Subsequently, the mirror was selected and has the following characteristics: 3 mm thick flat mirrors, with Ag coating with a reflection coefficient of 0,95. Finally, Fig. 4 also presents the auxiliary systems that will be described in the materials section. 2.2 Materials This section describes the auxiliary materials or equipment that were used to operate the Fresnel system, which was described in the previous section on CAD modeling. Pumping System. The portable pumping system has a 1/4 HP three-phase pump, with a maximum flow rate of 25 L/min, which is regulated through a three-phase LSIS Brand
202
B. E. Tarazona-Romero et al.
Variable Frequency Drive. Additionally, an electromechanical protection circuit and a tank with a capacity of 8 Gallons (See Fig. 5).
Fig. 5. Pump control
Solar Pre-heating System. The solar pre-heating system consists on an aluminum coil, which is responsible for delivering water at a temperature in the range between 45 and 60 °C at the inlet of the Fresnel system (See Fig. 6).
Fig. 6. Solar pre-heating system
The measurement and data acquisition system of pressure, temperature and flow variables; has two pressure transducers with a range of 0–500 PSI, two PT100 thermo resistors connected to two fixed temperature transmitters with a rate of 0–200 °C. The
Development of a Fresnel Artisanal System
203
flowmeter is a tangential turbine type with a measurement range of 0 to 33 L/min. The data sent by the sensors are stored in an electronic circuit made through the Arduino tool, which stores the physical values of the field instruments at an interval of every 15 s (See Fig. 7).
Fig. 7. Measurement and data acquisition system
3 Experimentation The testing or experimentation process began with the implementation of the Fresnel system integrated in the auxiliary systems, with a control and variation of flow through the frequency variator during a day from 8:00 to 2:00 h in Girón Santander with an average direct normal radiation of 4.10 kWh/m2 per day using the NREL NSRDB data viewer, contrasted and verified by a meteorological data file obtained through the Meteonorm software. The tests were carried out in time intervals of 1.30 h per frequency, varying from 10 Hz from 20 Hz to 60 Hz. This process was carried out for two days as an initial test to initially determine which frequency values would allow an adequate subsequent analysis. In total, each sample captured 360 data by daily frequency, which were surpassed and averaged for the variables of average inlet and outlet temperature, pressure, and inlet and outlet flow (Table 4). Table 5 shows the results of the initial test on the second day. It is highlighted that this process was carried out during two days and the solar radiation was apparently constant, passing by moments of cloudiness that lower the temperature in the system, showing that the tests were carried out in a month with mixed weather (Sun-rain). With the initial tests, it was determined to carry out three tests on different days between 9 am and 3 pm, with controlled flow rates at frequencies of 30, 45 and 60 Hz,
204
B. E. Tarazona-Romero et al. Table 4. Initial tests 01 Frequency HZ
P inlet (Psi)
P outlet (Psi)
T inlet °C
T outlet °C
Flow L/min
20
2
8
40
56
0.5
30
2.55
9
55
95
1.5
40
2.40
9
52
80
4.1
50
2
8
50
70
8.3
60
1.75
7
45
60
12.5
Table 5. Initial tests 02 Frequency HZ
P inlet (Psi)
P outlet (Psi)
T inlet °C
T outlet °C
Flow L/min
20
2
8
41
57
0.5
30
2.55
9
56
97
1.52
40
2.40
9
51
79
4.09
50
2
8
49
69
8.28
60
1.75
7
45
60
12.5
to identify the variation in heating of water compared to the flow handled by the system. The flow variation was carried out every 2 h, which means that for each frequency, approximately 480 data were taken through the data acquisition system per day. It should be noted that with a frequency lower than 30 Hz, the flow is not enough to transfer the fluid throughout the system, due to the diameter of the pipe and the distance that the fluid must travel, so an analysis below this value.
4 Results and Analysis Figure 8 presents the artisanal Fresnel collector model and the auxiliary systems used in the present study, in the test site, where the following stand out: Fresnel collector, the data acquisition system, the pumping system and the preheater. The results of the tests carried out during three days, with frequency variation, are described below. Table 6 details the inlet and outlet temperatures, inlet and outlet pressure, and flow rate of the system put into service during the first day of testing. The 30 Hz frequency being the one with the highest temperature. Furthermore, Table 7 details the inlet and outlet temperatures, inlet and outlet pressure, and flow of the system put into service during the second day of testing. Since the
Development of a Fresnel Artisanal System
Fig. 8. Central system and auxiliary for tests Table 6. Test 01 Frequency HZ
P inlet (Psi)
P outlet (Psi)
T inlet °C
T outlet °C
Flow L/min
30
2.51
8.8
55
90
1.52
45
2.35
8.5
50
75
4.9
60
1.78
7.1
46
61
12.4
Table 7. Test 02 Frequency HZ
P inlet (Psi)
P outlet (Psi)
T inlet °C
T outlet °C
Flow L/min
30
2.55
9
58
98
1.53
45
2.39
8.7
50
77
4.99
60
1.78
7.1
47
64
12.42
205
206
B. E. Tarazona-Romero et al.
30 Hz frequency is the one with the highest temperature on the second day, keeping the traceability with day one. Additionally, Table 8 details the inlet and outlet temperature, inlet and outlet pressure, and flow rates of the system put into service during the third and last day of testing. Being, the frequency of 30 Hz the one with the highest temperature like the previous days, but with the characteristic that it was a day passed by high cloud cover, therefore the temperature levels are considerably lower than the previous ones. Table 8. Test 03 Frequency HZ
P inlet (Psi)
P outlet (Psi)
T inlet °C
T outlet °C
Flow L/min
30
2.4
8.5
53
80
1.53
45
2.2
8
51
67
4.92
60
1.78
6.9
45
52
12.5
Once the tests have been performed, the average of the inlet and outlet temperatures, inlet and outlet pressure, and flow of the tests carried out, not including the last day, is carried out due to the high cloudiness presented that day (See Table 9). It is highlighted that the frequency of 30 Hz is the one with the highest temperature and the lowest flow, this relationship reflects that the lower the flow, the higher the temperature can be obtained in the system. Table 9. Average tests Frequency HZ
P inlet (Psi)
P outlet (Psi)
T inlet °C
T outlet °C
Flow L/min
30
2.53
8.9
56.5
94
1.525
45
2.37
8.6
50
76
4.945
60
1.78
7.1
46.5
62.5
12.41
Finally, the efficiency of the system is calculated, for which the direct solar radiation value of 4.10 kWh/m2 per day using NREL NSRDB data viewer, compared with the Meteonorm software. Furthermore, it is necessary to estimate the coefficients involved in the materials, such as the primary reflectors that contain a thin layer of silver with a reflectivity coefficient of 0.95, with an interception factor and an absorptivity of the receiver tube of 0.9. These data allow to determine that the direct solar radiation reflected to the focal point is 77% of the solar rays. Subsequently, some additional collector parameters are required and are presented in Table 10.
Development of a Fresnel Artisanal System
207
Table 10. Collector parameters Frequency HZ
P inlet (Psi)
Receiver tube external diameter m 0.00635 Stefan-Boltzmann constant W /m2 ∗ K 4
5.67 ∗ 10−8
Concentration factor
18.8
Average direct radiation Wh/m2
900
Receiver tube emissivity
0.9
Receiver tube surface area m2
0.041893
Receiver tube length m
2.10
Finally, it was necessary to take data from the Thermodynamics in SI Units by Yunus Cengel [25], on the properties of saturated water at 90 °C, which on average was the maximum point obtained in the process, where which the radiation losses of 12.3 W/m2 K were determined. Next, again the properties of the air at 25 °C are taken to calculate the losses by convection and radiation of the system of the same book mentioned above; obtaining a coefficient of losses 49.88 W/m2 K. The theoretical efficiency of the system is 60.48% and finally, the real efficiency of the system is 15% for the climatic conditions of the dates of experimentation.
5 Conclusions The Fresnel collector was built with soft and low-cost materials, available at all times in the test area, Giron Santander, Colombia. Being a development of soft and alternative technology with potential for study and development in the local field for the application of alternative thermo solar systems, it is important to highlight that the solar monitoring was performed manually and a series of auxiliary systems were integrated for its implementation and operation, which is not sustainable at the moment. The results of temperature increase in the heat transfer fluid were affected by the cloudiness of the test days and the mixed climate of the region. Therefore, it is necessary to carry out tests in long periods and the dynamic simulation of the system, in the current region and in the regional area of Colombia with the highest solar radiation constantly present to evaluate its applicability in highly potential areas with more meteorological conditions favorable. The pre-heater system reduces initial heat losses and the reduced diameter of the pipe induces flow losses of approximately 50% of the pump capacity. An inversely proportional relationship between temperature and flow was evidenced in the system, where the higher the temperature at the outlet of the system, the lower the flow rate and the lower the temperature at the outlet of the system, the greater the flow, indicating that this relationship should be reduced to improve the overall efficiency of the collector.
208
B. E. Tarazona-Romero et al.
References 1. Steinmann, W.-D.: 21 - thermal energy storage systems for concentrating solar power (CSP) technology. En: Cabeza, L.F. (ed.) Advances in Thermal Energy Storage Systems, pp. 511– 531. Woodhead Publishing (2015) 2. Häberle, A.: 19 - concentrating solar technologies for industrial process heat and cooling. En: Lovegrove, K., Stein, W. (eds.) Concentrating Solar Power Technology, pp. 602–619. Woodhead Publishing (2012) 3. Mills, D.: Advances in solar thermal electricity technology. Sol. Energy 76(1), 19–31 (2004). https://doi.org/10.1016/S0038-092X(03)00102-6 4. Lovegrove, K., Pye, J.: 2 - fundamental principles of concentrating solar power (CSP) systems. En: Lovegrove, K., Stein, W. (eds.) Concentrating Solar Power Technology, pp. 16–67. Woodhead Publishing (2012) 5. Tsekouras, P., Tzivanidis, C., Antonopoulos, K.: Optical and thermal investigation of a linear Fresnel collector with trapezoidal cavity receiver. Appl. Therm. Eng. 135, 379–388 (2018). https://doi.org/10.1016/j.applthermaleng.2018.02.082 6. Baharoon, D.A., Rahman, H.A., Omar, W.Z.W., Fadhl, S.O.: Historical development of concentrating solar power technologies to generate clean electricity efficiently – a review. Renew. Sustain. Energy Rev. 41, 996–1027 (2015). https://doi.org/10.1016/j.rser.2014.09.008 7. Bellos, E.: Progress in the design and the applications of linear Fresnel reflectors – a critical review. Therm. Sci. Eng. Prog. 10, 112–137 (2019). https://doi.org/10.1016/j.tsep.2019. 01.014 8. Montes, M.J., Rubbia, C., Abbas, R., Martínez-Val, J.M.: A comparative analysis of configurations of linear Fresnel collectors for concentrating solar power. Energy 73, 192–203 (2014). https://doi.org/10.1016/j.energy.2014.06.010 9. Mills, D.R.: 6 - linear Fresnel reflector (LFR) technology. En: Lovegrove, K., Stein, W. (eds.) Concentrating Solar Power Technology, pp. 153–196. Woodhead Publishing (2012) 10. Qiu, Y., Li, M.-J., Wang, K., Liu, Z.-B., Xue, X.-D.: Aiming strategy optimization for uniform flux distribution in the receiver of a linear Fresnel solar reflector using a multi-objective genetic algorithm. Appl. Energy 205, 1394–1407 (2017). https://doi.org/10.1016/j.apenergy. 2017.09.092 11. Abbas, R., Muñoz, J., Martínez-Val, J.M.: Steady-state thermal analysis of an innovative receiver for linear Fresnel reflectors. Appl. Energy 92, 503–515 (2012). https://doi.org/10. 1016/j.apenergy.2011.11.070 12. Abbas, R., Muñoz-Antón, J., Valdés, M., Martínez-Val, J.M.: High concentration linear Fresnel reflectors. Energy Convers. Manag. 72, 60–68 (2013). https://doi.org/10.1016/j.enconman. 2013.01.039 13. Cagnoli, M., Mazzei, D., Procopio, M., Russo, V., Savoldi, L., Zanino, R.: Analysis of the performance of linear Fresnel collectors: encapsulated vs. evacuated tubes. Sol. Energy 164, 119–138 (2018). https://doi.org/10.1016/j.solener.2018.02.037 14. Bellos, E., Said, Z., Tzivanidis, C.: The use of nanofluids in solar concentrating technologies: a comprehensive review. J. Clean. Prod. 196, 84–99 (2018). https://doi.org/10.1016/j.jclepro. 2018.06.048 15. Bermejo, P., Pino, F.J., Rosa, F.: Solar absorption cooling plant in Seville. Sol. Energy 84(8), 1503–1512 (2010). https://doi.org/10.1016/j.solener.2010.05.012 16. Gallego, A.J., Merello, G.M., Berenguel, M., Camacho, E.F.: Gain-scheduling model predictive control of a Fresnel collector field. Control Eng. Pract. 82, 1–3 (2019). https://doi.org/ 10.1016/j.conengprac.2018.09.022 17. Sebastián, A., Abbas, R., Valdés, M., Casanova, J.: Innovative thermal storage strategies for Fresnel-based concentrating solar plants with East-West orientation. Appl. Energy 230, 983–995 (2018). https://doi.org/10.1016/j.apenergy.2018.09.034
Development of a Fresnel Artisanal System
209
18. Askari, I.B., Ameri, M.: Techno economic feasibility analysis of Linear Fresnel solar field as thermal source of the MED/TVC desalination system. Desalination 394, 1–7 (2016). https:// doi.org/10.1016/j.desal.2016.04.022 19. Sharan, P., Bandyopadhyay, S.: Solar assisted multiple-effect evaporator. J. Clean. Prod. 142, 2340–2351 (2017). https://doi.org/10.1016/j.jclepro.2016.11.043 20. Zhu, G., Wendelin, T., Wagner, M.J., Kutscher, C.: History, current state, and future of linear Fresnel concentrating solar collectors. Sol. Energy 103, 639–652 (2014). https://doi.org/10. 1016/j.solener.2013.05.021 21. Atehortúa, F.H.R., Zwerg-Villegas, A.M.: Metodología de la investigación: más que una receta, p. 21 (2012) 22. Martínez, C., de la Luz, M.: INTRODUCCIÓN A LA METODOLOGÍA DE LA INVESTIGACIÓN EN BIOÉTICA: SUGERENCIAS PARA EL DESARROLLO DE UN PROTOCOLO DE INVESTIGACIÓN CUALITATIVA INTERDISCIPLINARIA. Acta Bioethica 14(1), 97–105 (2008). https://doi.org/10.4067/S1726-569X2008000100013 23. Riera, J.R.M., del-Pino-Casado, R.: Manual práctico de enfermería comunitaria. Elsevier, Amsterdam (2014) 24. Tarazona-Romero, B.E., Campos-Celador, Á., Muñoz-Maldonado, Y.A., SandovalRodríguez, C.L., Ascanio-Villabona, J.G.: Prototype of lineal solar collector Fresnel. Visión Electrónica 14(1), 4 (2020). https://doi.org/10.14483/22484728.16013 25. Yunus, A.C., Michael, B.: Thermodynamics: An Engineering Approach, 5th edn. McGrawHill Higher Education, New York (2007)
Portable Robotic Modular Kit for Teaching Gestures in Children with ASD Johanna Tobar1,2(B) , Joffre Delgado1 , Brandon Muñoz1 , Bladimir Bacca2 , and Eduardo Caicedo2 1 Universidad de las Fuerzas Armadas ESPE, Sangolquí 170501, Ecuador
[email protected] 2 Universidad del Valle, Cali 76001, Valle del Cauca, Colombia
Abstract. Technological advancement in recent years has allowed effective development in various fields, the link between technology and education is undoubtedly one of them. The health emergency that has arisen in recent months worldwide is a call for innovation to continue with learning processes that adapt to new study modalities. That is why this work proposes the development of a portable robotic modular kit for teaching gestures in children with autism spectrum disorder. The aim is to contribute to the development of non-verbal communication. The portable robotic modular kit consists of an Android application, a robot with 20 interchangeable parts and one container box to transport the robotic kit. The Android application has 3 working modes. Two of them are for working without the robot and the main working mode is the one that controls the robot. The validation of the tool was carried out by experts in the field of teaching to children with ASD. 65.57% of the people evaluated considered that the system can contribute to the teaching of gestures to autistic children and 70.49% of people considered that the tool can be used in children without disorder. Finally, the tool was presented in 5 therapy sessions in children with ASD within the Neuropsychological Rehabilitation Center CERENI. The results obtained are quite encouraging. We obtained a satisfaction of 100% in children with autism grade 1, while in children with autism grade 2 and 3 we obtained a satisfaction of 50% on the scale proposed with an increasing trend. Keywords: Teaching · Autism spectrum disorder · Modular robots · Gestures · Non-verbal communication · Human - robot interaction · Laterality
1 Introduction Autism spectrum disorder (ASD) is considered a neurological condition, of which very little is known. Several investigations focus on the help of social, communication and cognitive skills, all these related to ASD [1]. According to WHO conclusions, there are 62 people with ASD in every 10,000 inhabitants, that is, 1% of the world’s population is affected by the disorder. [2]. Delfos M. and Groot N. said that in Ecuador there are between 85,841 and 165,960 people with autism spectrum disorder [3]. For this study, it is essential to consider the most important problems presented by people affected by this © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 210–225, 2021. https://doi.org/10.1007/978-3-030-72212-8_16
Portable Robotic Modular Kit for Teaching Gestures
211
disorder. To do this, we refer to the explanation provided by Lorna Wing which says that autism spectrum disorder falls on the well-known Wing triad [4]. According to Lorna Wing, social relationships and social communication are part of the problems that most affect people with ASD. For this reason, it is essential to teach forms of daily communication, such as non-verbal. Section 1.1 presents the use of technology to contribute to the teaching of this type of communication and socialization. Highlighting social relationships, social communication, and rigidity of through behavior and play. The final details of each part of the portable robotic modular kit are presented in the subsequent subsections, detailing its most important characteristics. One of the principal problems was detected by a group of scientists who developed an experiment called “Communication Play Protocol”. In this experiment, it was concluded that in the communication between children, the use of gestures is essential, and children with ASD present a significant deficit in the performance of gestures, especially deictic gestures [5]. Also, it is detailed in other studies the great problem presented by children in cooperative games because they don’t understand basic gestures, for which they continue with repetitive actions typical of autism [6]. 1.1 Autism therapy by robotics When we talk about therapy, one of the main questions that arises is: Are social robots helpful for children? There are several studies that reflect the great attention that people involved have and the ways of teaching with social robots. In “Robotic assistants in therapy and education of children with autism: can a small humanoid robot help encourage social interaction skills?” [7], it is concluded that after several sessions, social robots become mediators between the child and the therapist, or also once the child gets used to the humanoid, he starts therapies on his own initiative. The child-robot interaction, once trust is established by the patient, can become a form of teaching with quite positive results, or only as the direct link between the teaching field and the child. In a study by Yaoxin Zhang and others, it is concluded that humanoid robots can adapt to the needs of learning, and training of social rules [8]. In view of the great acceptance of people with ASD towards the technology, there are already several robots that are being used in various therapies. However, most robots are designed with more general purposes, so several experts adapt them to the needs of the patient. Such is the case of the robotic kit Mindstorms developed by Lego [9], which seeks a development in the social part of the patient. The Milo robot is another similar case, this robot presents a teaching method aimed at children of early ages and it is used in various investigations [10]. One of the main characteristics that allow the interaction between patient and robot is anthropomorphism, which is defined as the tendency to attribute human characteristics to a robot. However, we should keep in mind not to fall into the valley of the inexplicable. A study carried out by Masahiro Mori called Unexplained Valley, analyzes a level of approval of robots in these patients. As a result, robots that resemble human appearance in a large percentage cause rejection by observers [11]. Therefore, it is essential to be clear about the final appearance of the used robots. Robots with shapes not so similar to humans are ideal for acceptance and communication with children with ASD. AISOY1 robot is another one widely used within this field, but with an approach directed at emotions [12]. NAO humanoid robots are the best known within gestural therapies, imitations,
212
J. Tobar et al.
socialization among others. The ease of use and multiple applications of these robots makes them quite adaptable to various needs. [13]. In an experiment with the ZENOS robot, to measure the interaction between the child and the therapist, a spontaneous participation of those affected was evidenced only when they worked with the robot and very little attention when they worked with humans. The learning results were quite satisfactory and with a relevant imitation of basic gestures for communication. [14]. This makes essential the feedback in teaching or an increase in the number of interactions between the child and the robot. In other therapies with adapted robots we find the article “Autonomous Robot-mediated Imitation Learning for Children with Autism” [15], which describes an imitation system for autistic children using a NAO Robot and data collection with Microsoft Kinect. The authors detail in their results that the population tolerated the robotic system and children with ASD worked better with this system than with a human therapist, obtaining relatively better performances [15]. NAO robots are also used in a study by Ryazantsev and Baranova L. In the investigation, one of the most encouraging results was measured by an electroencephalogram. It was evidenced that in a child with autism brain activity increased sharply compared to therapies without robots and this reaction remained throughout the session with the NAO robot [16]. Finally, Zheng in his article detected that a real-time imitation of patients with ASD can occur and that feedback is of great importance for adaptive intervention [17]. Other research affirms the use of interchangeable parts, both in robots and other methodologies to attract the attention of users. However, in the robotic area, the only one found that resembles a modular design is Mindstorms, which is a therapy adapted and oriented to children of advanced ages or young people, so it has different results from this research. However, the similarity between these therapies and this work lies in the level of attention obtained by the patients during therapy, confirming and guaranteeing a high level of attention while the child interacts directly with the interchangeable pieces. 1.2 Study proposal Currently, various robots adapted to therapies are used with quite favorable results. There are also robots developed based on the specific needs of patients with autism spectrum disorder. However, one of the recommendations of specialists is constant feedback on the topics discussed to reinforce the knowledge acquired [18]. This is a very conflictive problem when children with ASD are in therapies with robots considered high cost or fragile. This due to the high possibility of patients damaging a expensive robot. That is why an economical and robust robot is developed. This robot is capable of being portable for users and with intuitive operation for use of therapists and parents of the child with ASD. This robot is also capable of being considered a reinforcement tool at home. This research seeks to contribute to the development of Social Understanding, seeking a human-robot interaction, and then reinforcing other areas such as imitation and gesture recognition, laterality, and fine motor skills through the use of the robotic kit. All this will be done in this work with a portable robotic modular kit of interchangeable parts controlled from a mobile application capable of being used by the parents of each patient at home at any time.
Portable Robotic Modular Kit for Teaching Gestures
213
2 Development of the Modular Robotic Kit This section presents a brief description of the methodology used in the development of the portable robotic modular kit, as well as the methods used in the design and construction of the kit. It also details in more depth the operation of the robot and the app. 2.1 Design methodology As described in Sect. 1, the most important characteristic of this design is the direct approach to autistic children between the ages of 3 and 8 years, based on their needs. To do this, a mechatronic design based on the VDI 2206 methodology is used through the “V” model. This model describes the product requirement until reaching the exit. Based on the VDI methodology, it was carried out a conceptual design and sizing, component selection, mechanical design, virtual simulation, construction, and assembly. It was verified the operation with surveys directed to experts in the field of design. The results are presented in the next chapter. Development of the Portable Robotic Modular Kit Hardware: To give a preamble to the complete robotic kit, Fig. 1 shows all the components of the work developed. The robotic kit has 3 heads, 3 pairs of feet, 5 pairs of arms, 1 fixed torso, 1 container box, the mobile application for its control, 1 system charger and its corresponding user manual. Brief features of the design, manufacture and use are specified in the following sections. The method used for manufacturing and testing, as well as the technologies invested to obtain the complete kit, are also detailed.
Fig. 1. Parts of the modular robotic kit
2.2 Conceptual design of the robotic kit Table 1 shows the need for a special approach to the aesthetics of the kit, the gestures it performs, portability, robustness, and its intuitiveness, which are the parameters that
214
J. Tobar et al.
the experts consider most important. As it is a technology focused directly on children with autism spectrum disorder, the first step before the design was a needs assessment based on the bibliography which recommends robotic aspects with appearances don’t so similar to the human [11]. The use of soft colors and without many details. These concepts are validated with experts in education and therapists who work with children affected with ASD. The needs identified for the project are presented in the Table 1. Table 1. Needs matrix Need
Importance
1
The robot must be visually attractive
5 of 5
2
The robot must be modular (didactic)
4 of 5
3
The robot must teach gestures
5 of 5
4
Two distinctive colors to assemble and as support in the laterality
4 of 5
5
The robot must be small
4 of 5
6
The robot must be easy to transport
5 of 5
7
The robot must be robust
5 of 5
8
The application must be visually attractive
4 of 5
9
The application must be easy to use
5 of 5
10
The application must be able to be used without the robot
3 of 5
In order to meet the needs found and relate them to the technical part, Table 2 shows the technical metrics that the modular robot will contain to cover each need, as well as the evaluation range for subsequent validation of operation. Table 2. Metric need matrix Need
Metrics
Importance
3
Robot movement 5
Degrees of freedom
3
Replicate gestures
5
Subjective
2,4
Interchangeable parts
4
Fits and tolerances
9,10
Control software 5
Subjective
5,6,7
Portable Design
Subjective
5,6
Weight
5
Kg
5,6
Dimensions
3
mm
1,2,4,5
Aesthetic design
5
Subjective
8,9,10
Mobile app
4
Subjective
3
Measure
Portable Robotic Modular Kit for Teaching Gestures
215
Finally, Table 3 was developed based on the metrics and needs. It describes the main technical characteristics that the system needs to cover to function at the users’ home and thus be considered as a reinforcement therapy at home for patients with autism at established ages. A selection of concepts is made for each parameter and specification, with it the best components and methods that were used for the development of the robot are chosen. With all of this we can give the direct focus to children with autism spectrum disorder. Table 3. System specifications Parameter
Specification
Software
App in android Studio, through communication with Arduino
Gestures
Yes, no, ok, wrong, greet, hunger, cry
Architecture
Modular Type (detachable)
Mechanism
Minimum 2 degrees of freedom
App language
Spanish
System dimensions
Max 40 cm high and 25 cm wide
Weight
Max 1.5 kg
Power supply
USB charger 5 [V]
2.3 CAD design of modular robots (Simulation) Using CAD software, the design of the robot is proposed for its external part and its internal components. See Fig. 2.
Fig. 2. Robotic kit in exploded mode
In Fig. 3, the simulation of the gestures that can be assembled is shown. 2.4 Final product development The final result of the portable robotic modular kit is achieved through additive manufacturing. PLA is used as the base material, for subsequent processing with putty and
216
J. Tobar et al.
Fig. 3. Simulated gestures
automotive paint. Then, final modular robot and its parts are presented in a robust and smooth way capable of working with children. In the subsequent subsections, the final details of each part of the portable robotic modular kit are presented, detailing its most important characteristics. Assemble. This technological tool is a system that has a total of 20 interchangeable parts between upper and lower extremities, trunk, and heads. The robotic kit allows 165 possible ways of assembling it by varying its pieces. Figure 4 shows some possible combinations. With these motor games it is hoped to contribute to the development of the users’ spatial awareness, as has been done in other studies [19].
Fig. 4. Possible assembly models
For the particular case of teaching gestures, there is a special design for each movement. This design is presented in the App before executing the movement. These 7 special assemblies are presented in Fig. 5, with a brief description of their movement. To ensure correct assembly, the modular robots have a light warning system located near each upper extremity and head, as shown in the Fig. 6A. The robotic kit was designed with a direct approach to people with ASD. Therefore, it ensures the positioning of the parts before movement with grooves that allow the parts to be inserted in only one direction and position, as shown in Fig. 6B. Which means that the piece will always be placed in the ideal way to start the imitation of the selected gesture. Another feature that the portable robotic modular kit presents is the identification of its right and left part. With this, the user cans identify the correct positioning of each part. It was described in the CAD design section, then the final result is presented in the Fig. 4. This presentation of the robot also seeks to work on laterality in children.
Portable Robotic Modular Kit for Teaching Gestures
217
Fig. 5. Gestures
Fig. 6. Assembly details
Container Box: A box is designed; it can transport the modular robotic kit and it is made with laser technology in MDF to make it light. The box is finished inside with sponges to take care of the integrity of the pieces. Mobile App: The operation of the robotic kit is controlled and monitored from the mobile app. It was developed for any device that has an Android operating system. This App is part of the robotic kit. The application has the screens that you can see in Fig. 7. The mobile app has three modes of operation, they can be shown in the Fig. 7A and which are: A: Button that opens the screen “Sin Robot”; B: Button that opens the screen “Con Robot”; C: Button that opens the screen “Cuentos”; D: Button that closes the application. The screen “Sin Robot” is a setting for the system, in which the user can select the desired gesture and the application will show the robot as a gif for recognition. The details of this screen can be seen in Fig. 7B. This screen has the following functions: A1: Gesture selection buttons; A2: Animated gif showing selected gesture; A3: Return button to main screen; A4: Button that closes the application.
218
J. Tobar et al.
Fig. 7. App screens
The screen “Con Robot” is the interface that has the characteristics to control the robot. After connecting via Bluetooth, the App will show the tab seen in Fig. 7C. This screen is intended for the child’s tutor during the therapy and it has the following functions: B1: Gesture selection buttons; B2: Animated gif showing selected gesture; B3: Robot colors picker; B4: Button that commands the robot to move; B5: Button that disconnects the App from the robot; B6: Buttons that control the chronometer for recording part recognition data; B7: Return button to main screen; B8: Button that closes the application. Within the option “Con robot”, when we select the gesture, a complementary tab will open. It will show the robot that the child should assemble before executing the movement of the robot, the interface of this tab can be seen in the Fig. 7D. This screen contains the following functions: D1: Image showing the robot that the child should assemble; D2: Buttons that control the chronometer to record arming time data; D3: Button that alerts the application that the child has finished the arming. The screen “Cuentos” is a waiting method for children in which the user can choose between several stories selected by experts as a distraction or wait. These stories are recommendations of education experts, and it has pictograms in their description. This screen has the following functions: Story selection buttons; Return button to main screen; Button that closes the application. Through the features presented in Subsects. 2.2 and 2.3, it is expected that the therapist or tutor who carried out the therapy, can collect information in color distinction, laterality, and location of upper and lower extremities. This result can be delivered to the professional in charge of the patient to observe the evolution of the child affected with autism spectrum disorder. 2.5 Operation mode Once the communication between the robot and the application is established, the parent or person with whom the reinforcement therapy is carried out will follow the following steps:
Portable Robotic Modular Kit for Teaching Gestures
219
– Choosing the characteristic color of the robotic kit (2 colors available) – Selecting the desired gesture. The application will show a specific way of arming the robot. – Showing the image to the child and take the time that the child takes to assemble the robot (The application has tools for data collection) – Once the assembly process is completed, the application will send the robot the movement that it has to realize. The patient will be asked if he recognizes the gesture or if he can imitate it. – The person in charge provides the information of the assembly times and the number of gesture acknowledgments to the charge therapist. Figure 8 shows an operation diagram of the robotic kit.
Fig. 8. Operation diagram of the robotic kit
3 Results and Discussion The two test scenarios are presented in this section. The first one is a validation with experts in education and therapy. The second scenario is the field tests. These tests are divided into 2 parts. The first tests are realized with a sample of 11 children without ASD to validate the functioning of the system. The second tests are realized with 4 children affected with the disorder who regularly attend therapy at the “CERENI” center. 3.1 Preliminary tests Before having direct contact with children, the tool is presented to education professionals who currently work with healthy children and in some cases with children affected with autism spectrum disorder. The portable robotic modular kit is presented virtually and with the support of multimedia tools. To validate the use of the tool in the target population, the opinion provided was from 61 experts in the area of education, among which we have: Psychologist, Teachers of Basic Education and Students in the last levels of initial Education. The opinion was obtained through questions given to the participants on a scale with ranges from 1 to 4, where 1 represents total dissatisfaction and
220
J. Tobar et al.
4 total satisfaction in the question asked. According to the data, except for question 4 which refers to the movement of the “WRONG” gesture, in general the observations and considerations of the education experts are positive. It was obtained a 65.92% of responses totally agree, which means a value of 4, and 23.65% of responses with value 3. Regarding more detailed questions such as 8 in which it refers to the final appearance of the kit, an average evaluation of 3.79 is obtained in a satisfaction scale of maximum 4. Which is why it is considered that modular robots are attractive for the children. With respect to the portability of the system an average evaluation equal to 3.58 of 4 is obtained. This result affirmed that the kit is considered portable for the users. With the results obtained, it can be concluded that the portable robotic modular kit is suitable for working and contributing to the learning of gestures, laterality, and fine motor skills in children. 3.2 Field tests The second evaluation was carried out in two parts. The first one directed to children without ASD in the province of Pichincha and Ambato. The second one directed to children with autism spectrum disorder at the Comprehensive Neuropsychological and Neurological Rehabilitation Center “CERENI”. This subsection presents data and evidence from the sessions. To carry out the field tests, the research followed the rules and recommendations of the ethics committee belonging to the Universidad del Valle and approved with the document number 003–020. Children without ASD. Once the signing of the informed consent for participation in research and publication of results for academic purposes has been obtained, the study was carried out with a sample of 11 children between 4 and 7 years of age, among them 5 male and 6 females. In these tests the portable robotic modular kit was delivered to the parents with a user manual and after its use they were asked to fill out an evaluation survey, which contains the questions shown in Table 4. Some of the tests carried out can be seen in Fig. 9.
Fig. 9. Function test, children without ASD
Once the information was collected, we obtained the data shown in Table 5. It should be noted that they are statistical data of opinions of the parents or people with whom the children did the therapy.
Portable Robotic Modular Kit for Teaching Gestures
221
Table 4. Questions for parents of chindren without ASD No.
Question
Q1
Consider that that modular robots (interchangeable parts) arouse interest in children?
Q2
To what extent do you consider these robots are innovative?
Q3
To what extent do you consider that the process with modular robots is intuitive?
Q4
Consider that that the application interface allows easily handling of the robot?
Q5
To what extent do you consider that a modular robot is a useful tool?
Q6
Consider that that the repetition of gestures is attractive and useful for children?
Q7
Do you consider that the colors of the robot help the recognition of the left and right?
Q8
To what extent do you consider that the size and shape of the robots are adequate to be considered a reinforcement tool at home? (Easy transportation to home)
Q9
Consider that that the process of assembling the robot is easy for children?
Q10
To what extent do you consider that the entire kit is easy to use?
With a total of 70.91% responses with a value of 4, “Strongly agree”, the tool is validated as useful in children without autism spectrum disorder, and it doesn’t present any inconvenience during its use. These results allow the passage to the next stage of tests in children with autism spectrum disorder. Table 5. Statistics of children without ASD Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
Q10
Media
3,91
3,64
3,55
3,91
3,36
3,45
3,91
3,82
3,82
3,64
Std. dev
0,30
0,50
0,52
0,30
0,50
0,69
0,30
0,40
0,40
0,50
Var. coeff
0,08
0,14
0,15
0,08
0,15
0,20
0,08
0,11
0,11
0,14
Var
0,09
0,25
0,27
0,09
0,25
0,47
0,09
0,16
0,16
0,25
In questions with more interest, the result of question Q1 is presented, which gives an average of 3.91 out of 4. This tells us that the portable robotic modular kit caused a high interest from users. In question Q8, which refers to the portability of the robotic modular kit, we get a 3.82 out of 4. This shows that the parents of the children don´t had problem transporting the robot and its components. Children with ASD. In children with ASD the tests performed were different from those performed with children without ASD. This is because, in children without ASD only the function was tested, while in children with ASD, the kit was used as a therapy tool. Children with autism spectrum disorder who participated in the testing sessions are between the ages of 4 and 8, among them affected by autism in grade 1, 2 and 3. They regularly attend the Neuropsychological Rehabilitation Center and integral neurological “CERENI”, which is currently working with all biosecurity standards, and with
222
J. Tobar et al.
admission protocols such as: admission with some symptom is prohibited, disinfected footwear and body, mandatory use of mask, hand disinfection and keep a distance of 2 m. The participants were selected by the specialist therapists of the center according to the time they have been treating them. Before the process the parents of each one was informed, recording their consent in writing to participate in the study and publication of the data with academic purposes. The field test was carried out on 4 children, some details of them are presented below: Male of 8 Age, Grade 1, Total session 5; Male of 7 Age, Grade 1, Total session 3; Male of 4 Age, Grade 2, Total session 5; Male of 4 Age, Grade 3, Total session 5. Within the first intervention, it was observed a behavior in children with autism of grade 1 of great interest in the tool. In grade 2 there is a very slight interest, so the therapist intervenes very frequently to attract the child’s attention. While in grade 3 the interest is totally null. A stronger attention is shown in the following sessions compared to the first in all cases. One of the children of grade 1 perfects the assembly and reduces their times to less than a minute in the second session. While the children of grade 2 and 3, despite not assembling the robot until the third session, they begin to recognize gestures. The results of the last session were quite favorable for the few sessions carried out due to the pandemic. The 2 children of grade 1 assembled the robot without the help of therapists in less than 1 min. The child of grade 2 recognized all the gestures except for the “WRONG” gesture and he assembled the robot in less than 3 min. While the child of grade 3 recognizes all the gestures except for the “HUNGER” gesture, and he manages to assemble the robot with minimal help from the therapist in less than 4 min. Figure 10 shows evidence from the last session of each child. The results of the tests performed on children with ASD are separated by the degree of autism.
Fig. 10. Children with ASD last sessions; A: Child with ASD grade 1, fifth session, 06/08/20; B: Child with ASD grade 1, third session, 07/08/20; C: Child with ASD grade 2, fifth session, 06/08/20; D: Child with ASD grade 3, fifth session, 07/08/20
In children with autism grade 1, it is evident that gesture recognition increases as the sessions increase. Figure 11A shows the media of gestures recognized in each session. The help of the therapist decreases so they can handle the system (Robotic kit and application) alone as can be seen in Fig. 11B. The arming times decrease until reaching the initial objective of minus 1 min, the results are observed in Fig. 11C.
Portable Robotic Modular Kit for Teaching Gestures
223
Fig. 11. Results of children grade 1
The results in the child with autism grade 2 are also favorable. The gestures learned increase as more sessions are performed. An increase in the media from 0.29 to 1.29 is evidenced in a range of 3, as can be seen in Fig. 12A. The therapist’s help, although still a bit high, tends to decrease as shown in Fig. 12B. And the arming times went from not assemble the robot to assemble it in less than 2 min, despite only placing the extremities, the results are observed in Fig. 12C. In the fourth session a peak is observed in Figs. 12A and 12B, this is due to the fact that in that session the child presented very high hyperactivity, according to the therapists this is due to his diet in that day.
Fig. 12. Results of children grade 2
Finally, in the child with autism grade 3, considering that his degree of autism is serious, the advances that he presents, as more sessions are carried out, are significant. He passes in the recognition of gestures of a media from 0.14 to 0.86 of a maximum value of 3, as we can be seen in Fig. 13A. The therapist’s help, however, continues to be maximum, but it came to this by not paying any attention to the robot in the first sessions, as shown in Fig. 13B. The arming times went from not assemble the robot to assemble it in a time of less than 3 min, despite only placing some parts, the results are observed in Fig. 13C.
Fig. 13. Results of children grade 3
224
J. Tobar et al.
4 Conclusions The portable robotic modular kit is a tool designed for reinforcement therapy at home (easy portability), with a direct approach to children with ASD. It contributes in the teaching of gestures, laterality and development of motor skills. It allows children with ASD to do reinforcement work at home, with a data collection for later delivery to the therapist in charge. The portable robotic modular kit is an educational tool. According to an average of 3.82 out of 4 on the satisfaction scale by the parents who used the robotic kit, they consider that it has the correct shape and weight to be carried and used as a reinforcement tool at home. In addition, the system allows a rapid collection of information by parents or guardians of patients, the same that is provided to therapists for the evaluation of progress in the therapies carried out. In all the children with autism, there was a noticeable improvement for each session. The best results are observed in children with autism grade 1, where on the scale proposed by therapists, a 100% of satisfaction is achieved. While in children with autism grade 2 and 3, a 50% of satisfaction in the proposed criterion is achieved with an increasing trend. Also, several recommendations given by parents of children without disorders show the desire to use the tool as a toy for the development of motor skills in their children. Acknowledgment. The authors thank Dr. Martha de la Torre and the CERENI center for their valuable participation and help in the study.
References 1. Robertson, C.E., Baron-Cohen, S.: Sensory perception in autism. Nat. Rev. Neurosci. 18(11), 671–684 (2017). https://doi.org/10.1038/nrn.2017.112 2. López-Chávez, C., Larrea-Castelo, M.-L., López-Chávez, C., Larrea-Castelo, M.-L.: Autismo en Ecuador: un grupo social en espera de atención. Revista Ecuatoriana de Neurología 26(3), 203214 (2017) 3. Moncayo, C., Carofilis, J.: Situación de la inclusión eduacativa de niños y jovenes con trastorno del espectro autista (TEA), en las instituciones educativas en las ciudades de Santo Domingo y Manta durante el año lectivo 2013 – 2014. Pontifica Universidad Católica del Ecuador (2014) 4. Naguy, A., Abdullah, A.: Autism: the second triad of impairment demystified. J. Nerv. Ment. Dis. 207(5), 417 (2019) 5. Dimitrova, N., Özçalı¸skan, S., ¸ Adamson, L.B.: Parents’ translations of child gesture facilitate word learning in children with autism, Down syndrome and typical development. J. Autism Dev. Disord. 46(1), 221231 (2016) 6. Boucenna, S., Narzisi, A., Tilmont, E., Muratori, F., Pioggia, G., Cohen, D., Chetouani, M.: Interactive technologies for autistic children: a review. Cogn. Comput. 6(4), 722–740 (2014). https://doi.org/10.1007/s12559-014-9276-x 7. Alhaddad, A.Y., Cabibihan, J.-J., Bonarini, A.: Head impact severity measures for small social robots thrown during meltdown in autism. Int. J. Soc. Robot. 11(2), 255270 (2019) 8. Zhang, Y., Song, W., Tan, Z., Zhu, H., Wang, Y., Lam, C.M., Weng, Y., Hoi, S.P., Lu, H., Man Chan, B.S., Chen, J., Yi, L.: Could social robots facilitate children with autism spectrum disorders in learning distrust and deception? Comput. Hum. Behav. 98, 140149 (2019)
Portable Robotic Modular Kit for Teaching Gestures
225
9. Soares, F., Costal, S., Gon, N.: Robotic-Autism project: technology for autistic children. In: 2013 IEEE 3rd Portuguese Meeting in Bioengineering (ENBENG), pp. 14 (2018) 10. Sánchez, X.E.B.: Efectividad del ‘Robot Milo’ en el desarrollo de habilidades sociales y comunicación en niños de 5 a 7 años con trastorno del espectro del autismo de grado 1. Universidad San Francisco de Quito, Quito (2018) 11. Matovelle, L., César, D.: Diseño y construcción de una cabeza mecatrónica de aspecto realista 2013. Accessed 23 Apr 2020 12. Paracuellos de los Santos, D.: Programación de Robots para mejorar la atención terapeutica de niños con Trastornos Generalizados del Desarrollo (TGD). Dec. 2017 13. Telisheva, Z., Turarova, A., Zhanatkyzy, A., Abylkasymova, G., Sandygulova, A.: Robotassisted therapy for the severe form of autism: challenges and recommendations. En: Salichs, M.A., Ge, S.S., Barakova, E.I., Cabibihan, J.-J., Wagner, A.R., Castro-González, Á., He, H. (Eds.) Social Robotics (pp. 474–483). Springer International Publishing (2019) 14. Schadenberg, B.R., Reidsma, D., Heylen, D.K.J., Evers, V.: Differences in spontaneous interactions of autistic children in an interaction with an adult and humanoid robot. Front. Robot. A I, 7 (2020). https://doi.org/10.3389/frobt.2020.00028 15. Zheng, Z., Das, S., Young, E.M., Swanson, A., Warren, Z., Sarkar, N.: Autonomous robotmediated imitation learning for children with autism. In: 2014 IEEE International Conference on Robotics and Automation (ICRA), pp. 2707–2712 (2014). https://doi.org/10.1109/ICRA. 2014.6907247 16. Ryazantsev., A., Baranova, L.A.: Android NAO robot as a correction tool of abnormalities in children with autism. TEST Eng. Manage. 83 (2020) 17. Zheng, Z., Young, E.M., Swanson, A.R., Weitlauf, A.S., Warren, Z.E., Sarkar, N.: Robotmediated imitation skill training for children with autism. IEEE Trans. Neural Syst. Rehabil. Eng. 24(6), 682–691 (2016). https://doi.org/10.1109/TNSRE.2015.2475724 18. Martos, J., Freire-Prudencio, S., Llorente-Comi, M., Ayuda-Pascual, R., Gonzalez-Navarro, A.: Autism and intelligence quotient: stability? Rev. Neurol. 66, S39–S44 (2018) 19. Sandoval, V., Irene, C.: Programa de juegos psicomotrices para el desarrollo de las nociones espaciales en niños de 5 años de la I.E.I. N° 011 “Juan Ugaz”. Chiclayo-2017 (2018)
Descriptive Study of a Rotary Machine Affected by Misalignment and Imbalance Applying the Wavelet Transform Camilo Leonardo Sandoval-Rodriguez1,2(B) , Brayan Eduardo Tarazona-Romero1,2 , Omar Lengerke-Perez1 , Carlos Gerardo Cárdenas-Arias1,2 , Diana Carolina Dulcey Diaz1 , and Oscar Arnulfo Acosta Cárdenas1 1 Unidades Tecnológicas de Santander, Bucaramanga, Colombia
{csandoval,btarazona,olengerke,ccardenas,ddulcey, oacosta}@correo.uts.edu.co 2 University of the Basque Country, UPV/EHU, 48013 Bilbao, Spain
Abstract. The study of the effects produced by both, imbalance and misalignment in rotating machines, allow identifying the effects produced by these two phenomena under normal operating conditions of the equipment. Initially, an analysis is carried out using the wavelet transform, through the application of the LabVIEW and Matlab software; the goal was to compare these results with those obtained with the application of the Fourier transform. Additionally, a subsequent comparison of the results is carried out as a consequence of an initial comparison with an antecedent, where the Cepstrum-Fourier transformer and the Wavelet transform were applied to verify the existence of faults in a rotating machine. The present work is developed in order to determine which of these two studies represent these effects of misalignment and imbalance in a clear, specific, and/or significant way. Keywords: Wavelet · Vibrations · Algorithm · Misalignment · Measurement and unbalance
1 Introduction Vibrations are propagations of waves through a medium [1, 2]; vibrations can become a common problem [3] because they are repetitive, for that reason, they could generate consequences over machines. In the case of rotary machines [4, 5] they are caused by factors such as wear (loss of material) [6] and/or loads [7], generating both imbalance and misalignment, which in turn would lead to overheating and rupture of parts, loss of equipment efficiency, excessive noise and collateral damage, among others [8]. Predictive maintenance tasks [9] and condition monitoring [10] are developed in the industry applying non-destructive techniques in machines that provide relevant information about the internal structure of the equipment [11], allowing to determine when changes of parts are required, taking precedence over the occurrence of the failure [9]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 226–242, 2021. https://doi.org/10.1007/978-3-030-72212-8_17
Descriptive Study of a Rotary Machine Affected by Misalignment
227
Due to the above-mentioned, vibration analysis arises [12]; this analysis allows versatile and effectively monitoring machines operating conditions [13], The vibration analysis technique [14] consists of both measuring and interpreting vibrations [15]. The method consists of taking the produced signals by vibration through an analyzer, equipped with a sensor in the time domain, converting the signal into the frequency domain [16] and later, perform the analysis of both the obtained data and the spectra to predict failures in parts or equipment [17]. Analysis of the stored data through the measurement equipment has been classified over time in various ways such as [18] for example [19]: measurement of general vibration level [20], spectral analysis of vibration [21], monitoring discrete frequency [22], shock pulse monitoring [23], kurtosis measurement [24], signal averaging [25], Cepstrum analysis [26], Fourier analysis [27, 28], and Wavelet analysis [29] among others. It is relevant to stick out that some of the mentioned categories are widely applied to validated processes, whereas others are still in the process of development and research. Finally, the development of the presented work is both investigative and experimental in nature; it is based on the analysis of vibrations which are present in rotary machines. Data acquisition is carried out using a vibration sensor belonging to the Laboratory of Industrial Control and Automation in the Unidades Tecnologicas de Santander, in which signals are obtained for subsequent analysis. The study of these signals is developed through the application of the Wavelet transform and the construction of an algorithm capable of processing the information, to identify comparison markers for imbalance, misalignment, and normal operation. These data are contrasted with a parallel research development, in which a similar analysis process was carried out applying the CeptrumsFourier and Fourier transform (TFF).
2 Method and Materials 2.1 Method Figure 1 represents the flow chart concerning the sequential development of the required activities of the present work. The methodology has a mixed approach: quantitative and qualitative it starts with the characterization of the unbalance and misalignment signals of a rotary machine applying the analysis through the Wavelet transform (TW), contrasting it with the Ceptrums-Fourier and Fourier transform. The first step was to carry out a state of the art that would allow both defining parameters and goals of the work. Then, the physical condition of the rotating machine was tested using a vibration analyzer. The aforementioned devices belong to the Laboratory of Industrial Control and Automation in the Unidades Tecnologicas de Santander. Figure 2 shows a diagram of the process measurement chain. The laptop has the function of performing data acquisition and information processing through LabVIEW and MATLAB software. LabVIEW carries out the ON-LINE shots; MATLAB carries out the succeeding OFF-LINE analyzes (FFT, Cepstrum-Fourier, and Wavelet). Figure 3 corresponds to the graphical interface of the block diagram developed for the capture and subsequent acquisition of information. The stored data is processed OFF-LINE using MATLAB to perform the next calculations: maximum values, average values, standard deviations, coefficient of variation, Euclidean distance, and root mean square error.
228
C. L. Sandoval-Rodriguez et al.
Fig. 1. Workflow diagram
Fig. 2. Measurement chain
Experimental Development A factorial experimental design was developed to carry out the final tests; this experimental design focuses on the characterization called 2k, where k is the number of factors at two levels; in this case, the factors to be evaluated are the states of the rotary machine:
Descriptive Study of a Rotary Machine Affected by Misalignment
229
Fig. 3. Blok diagram in LabVIEW
normal, misalignment, and imbalance. The number of registers or iterations was calculated applying the Eq. (1) where: n is the number of iterations for the groups and k is equal to 2. n = 2k
(1)
Based on the calculation of Eq. (2), the design would have more than four combinations; however, the number of iterations for the present work was 8, for each state of the rotary machine. n = 2k ; 22 ≥ 4
(2)
Experimentation On-line Process Once the data processing was finished and tested through the developed program using Lab-VIEW, it was carried out a series of continuous sample records or ON-LINE tests, in a voltage range of 10 to −10, taking 10k readings, in a waiting time of 20 ms. This information was generated according to a series of groups (See Table 1), storing eight records. For each group, the ON-LINE program automatically saves the information using an Ivm format in a folder assigned as follows: Off -line Process The OFF-LINE process is carried out through data processing using MATLAB Software; this process uses the data stored in the ON-LINE process. It was necessary to develop eight algorithms to process the data as follows: (1) loading and classification of stored data, (2) calculating of the spectrum to the registers of each group, (3) calculating of the maximum values of the spectrum, (4) calculating of Wavelet transformation, (5) calculating of averages values and standard deviation, (6) calculating of the coefficient of variation, (7) determination of the Euclidean distance and finally (8) the calculating of the root mean square error.
230
C. L. Sandoval-Rodriguez et al. Table 1. Ivm Storage logs Nomenclature Item
Description Group - Test
Registers name’s
1
Normal
NT20_01…. NT20_08
2
Misalignment 0,5 grades
MDA05T20_01 … MDA05T20_08
3
Misalignment 1 grades
MDA1T20_01 … MDA1T20_08
4
Misalignment 1,5 grades
MDA15T20_01 … MDA15T20_08
5
Disbalance 30 g; 1 mass
MDB1T20_01 … MDB1T20_08
6
Disbalance 60 g; 2 mass
MDB2T20_01 … MDB2T20_08
7
Disbalance 90 g; 3 mass
MDB3T20_01 … MDB3T20_08
2.2 Materials The materials used for the development of the research were basically a laptop with the LabVIEW and MATLAB software, as well as the bank of mechanical vibrations for rotary machines located in the Structural Health Laboratory of the Unidades Tecnologicas de Santander; Fig. 4 shows this device. The vibration analysis module allows simulating the performance of a rotary machine in real-time; this module has the next components: an electric motor whose axis has a series of scales or weights, a frequency variator, and an accelerometer.
Fig. 4. Vibration analysis module
A Dytran sensor reference 305682 and a Dytran amplification system reference 4110C were used to perform the measurements; Fig. 5 shows this module. After amplifying the signal, it is stored on an NI 6001 card, and the A/D conversion process is performed.
Descriptive Study of a Rotary Machine Affected by Misalignment
231
Fig. 5. Dytran module reference 4110
3 Results and Analysis 3.1 Algorithm in MATLAB This section presents an example of the obtained results using the processing algorithm developed in MATLAB, for each condition, normal, misalignment, and imbalance. Table 2 presents the records classified according to the reference variable group using a given nomenclature to allow its identification and grouping. Table 2. Relationship of variables with their group of belonging Reference variable Group Normal Group
NT20
Misalignment Group MDA05T20 MDA1T20 MDA15T20 Disbalance Group
MDB1T20 MDTB2T20 MDB3T20
It is relevant to mention that the information storage process generates matrices. Due to the generated matrices corresponds to column vectors, it is necessary to guarantee that the length of each vector is equal; therefore, it was needed to develop one algorithm; Table 3 presents part of the obtained results.
232
C. L. Sandoval-Rodriguez et al. Table 3. Longitude of some used vectors Grupo Normal Register
Nomenclatures Longitude Longitude
NT20_01 LongN1
2500
NT20_02 LongN2
2500
NT20_03 LongN3
2500
NT20_04 LongN4
2500
NT20_05 LongN5
2500
NT20_06 LongN6
2500
NT20_07 LongN7
2500
NT20_08 LongN8
2500
3.2 Fourier Transform Figure 6 shows the Fourier transform for each of the classified groups: normal, misaligned, and unbalanced. As it can observe, at the beginning of each graph, a peak of zero is presented in every register; that peak does not represent information for signal processing, since it is a start condition of the bench’s test. For this reason, it was necessary to cut this first data.
Fig. 6. Fourier Transform in MATLAB
3.3 Application of Maximums to the Spectrum with Filter Figure 7 presents the results of the calculations of the maximum values from each record. These values correspond to the mean of the maximum of the filtered spectrum. Table 4 shows the used values in the construction of the graph.
Descriptive Study of a Rotary Machine Affected by Misalignment
233
Fig. 7. Signal Maximum
Table 4. Maximum data sample Normal Group
Disbalance Group 1 mass
Vmax FFT
max
Vmax FFT
Max
MXN1
52,7268
MXM1301
44,8670
MXN2
55,5886
MXM1302
48,4924
MXN3
54,9005
MXM1303
38,6234
MXN4
42,7948
MXM1304
37,0925
MXN5
46,0414
MXM1305
36,4315
MXN6
53,9810
MXM1306
49,6852
MXN7
42,4441
MXM1307
41,8099
MXN 8 67,7028
MXM1308
48,0578
3.4 Wavelet Transform Approximation Coefficient Daubuchies, Symlet, and Coileft were proposed for selecting of the mother wavelet. Due to Symlet is the improved version of Daubechies, which is appropriate for the vibration signals treatment, the order 4 Symlet family (Sym4) was selected for the present work. Figure 8 presents an example of the graphs corresponding to the signals captured in the tests without any treatment, compared to the Sym4 mother Wavelet signal for each machine condition. The figure reflects the length of the approximation coefficients of the Wavelet transform in the time domain, starting from a shorter time, but with a greater degree of information about low frequencies without signal treatment.
234
C. L. Sandoval-Rodriguez et al.
Fig. 8. Wavelet transform approximation coefficient
3.5 Coefficients of Variation, Averages and Euclidean Distance Table 5 presents the coefficients of variation of normal, misalignment, and imbalance groups for the Fourier transform. Table 5. Variation coefficients of the Normal, misalignment and imbalance groups of the Fourier transform Description
Fourier Transforms Variable
Normal
CIDN
CV % 2,27
Misalignment 0,5° C1DM05G
8,09
Misalignment 1°
2,61
C1DM1G
Misalignment 1,5° C1DM15G
2,21
Disbalance 1 M
C1DM130
2,93
Disbalance 2 M
C1DM260 14,48
Disbalance 3 M
C1DM390
9,29
Table 6 shows the coefficients of variation of normal, misalignment, and imbalance groups for the maximums of the Fourier transform.
Descriptive Study of a Rotary Machine Affected by Misalignment
235
Table 6. Variation coefficients of the Normal, misalignment and imbalance groups of the maximum Fourier transform Description
Normal
Maximum Fourier Transforms Variable
CV %
CDN
15,98
Misalignment 0,5° CDM05G 19,12 Misalignment 1°
CDM1G
18,30
Misalignment 1,5° CDM15G 26,23 Disbalance 1 M
CDM130 12,46
Disbalance 2 M
CDM260 28,53
Disbalance 3 M
CDM390 12.17
Finally, Table 7 displays the coefficients of variation of normal, misalignment, and imbalance groups for the wavelet transform. Table 7. Variation coefficients of the Normal, misalignment and imbalance groups of the Wavelet transform Description
Wavelet Transforms Variable
CV %
Normal
CWIDN
0,66
Misalignment 0,5°
CW1DM05G
3,86
Misalignment 1°
CW1DM1G
2,18
Misalignment 1,5°
CW1DM15G
2,01
Disbalance 1 M
CW1DM130
1,79
Disbalance 2 M
CW1DM260
13,11
Disbalance 3 M
CW1DM390
3,20
Based on the carried-out tests, the wavelet transform has a shorter distance between the groups compared to normal group, imbalance group, and misalignment group applying both the Fourier techniques and the Fourier maxima. The results show as well that the variables with the best response are the next: misaligned at 1°, misaligned at 1.5°, and unbalanced with 1 weight, as evidenced in Table 8. Figure 9 shows the averages and the Euclidean distance calculated by the Fourier transform technique. The upper graph represents the averages values of the conditions normal, imbalance, and misalignment. The lower Figure corresponds to the Euclidean distances that are calculated from the normal point. These results do not have cohesion
236
C. L. Sandoval-Rodriguez et al.
Table 8. Euclidean distance of the related variables between normal vs. delineation and Misalignment phenomena Euclidian Distance Description Normal Vs Misalignment 0,5°
Fourier Transforms
Máximum Fourier Transforms
Wavelet Transforms
3,7195089
37,82971223
0,1955434
Normal Vs Misalignment 1°
1,619799176
6,20203238
0,03192805
Normal Vs Misalignment 1,5°
0,354759101
16,5387557
0,0985183
Normal Vs Disbalance 1M
0,598595753
8,890051608
0,01494572
Normal Vs Disbalance 2M
3,198316889
39,2232912
0,12145389
Normal Vs Disbalance 3M
4,891222749
9,99791044
0,19655908
Fig. 9. FFT average and distance
between them; for that reason, the signal processing of the Fourier transform did not carry out and the analysis of low frequencies was developed using the wavelet transform. Figure 10 presents the comparison of the average value of the approximation coefficients that have greater cohesion compared to the other applied techniques. Additionally, it is evident that some differences exist between the bars 1 and 6, which correspond to the distances between the normal condition and the misalignment condition at 0.5°, and the existing distances between the normal condition and the unbalance condition using three weights.
Descriptive Study of a Rotary Machine Affected by Misalignment
237
Fig. 10. Wavelet transform average and distance
3.6 Coefficients of Variation Comparison Table 9 shows the coefficients of variation resulting from the application of the processing of the Cepstrum transform. Table 10 presents the coefficients of variation corresponding to the Wavelet and Fourier transforms. Table 9. Variation coefficients of the Cepstrum processing application Variation coefficients Descripción
Cepstrum Transforms Variable
CV%
Normal
GCAB
12,44
Misalignment 0,5°
DA05G
6,24
Misalignment 1°
DA1G
8,21
Misalignment 1,5°
DA15G
9,45
Disbalance 1 M
DBV2R1 M
4,82
The results in Tables 9 and 10 suggest that the wavelet transform has the lowest percentages of dispersion. These results allow concluding that the applied technique is assertive for the analysis and characterization of misalignment and imbalance phenomena in rotating machines, compared to an aligned and balanced system.
238
C. L. Sandoval-Rodriguez et al. Table 10. Variation coefficients of Wavelet and Fourier processing application Coeficientes de Variación Descripción
Variable
Normal
CDN
TW
TFF
CV% 0,66 2,27
Misalignment 0,5° CDM05G
3,86 8,09
Misalignment 1°
2,18 2,61
CDM1G
Misalignment 1,5° CDM15G Disbalance 1 M
2,01 2,21
DBV2R1M 1,79 2,93
3.7 Euclidean Distance Comparison Euclidean distance is a criterion for the analysis of vibration signals for phenomena of misalignment and imbalance; Table 11 shows the calculation results of this distance. These results suggest that the Wavelet transform has values close to the Cepstrum Transform, which allows concluding that they do not permit characterizing faults as easy as the Fourier transform. Table 11. Processing techniques; Euclidean distance comparison of Cepstrum, Wavelet and Fourier Descripción
T. Cepstrum
TFF
TW
Normal vs Misalignment 0,5° 0,0039332
3,7198 0,1955
Normal vs Misalignment 1°
0,0046229
1,6198 0,0319
Normal vs Misalignment 1,5° 0,0109488
0,3548 0,0985
Normal vs Disbalance 1M
0,5986 0,0149
0,0172853
3.8 Root Mean Square Error 1 Xi n n
X =
i=0
Table 12 presents the results of the root mean square errors calculated by Eq. 3.
(3)
Descriptive Study of a Rotary Machine Affected by Misalignment
239
Table 12. Root Mean Square error Root Mean Square error EMC Cepstrum vs Wavelet 0,011464 EMC Cepstrum vs Fourier
4,247445
EMC Wavelet vs Fourier
3,922081
4 Discussion The analysis of the Wavelet transform applied in the present study has a better performance compared to the Fourier Transform when it comes to the coefficient of variation for the maximum values of each record in the transformed domain. However, the Wavelet technique reflects values with smaller distances between the evaluated groups. On the other hand, both the Wavelet transform and the Fourier transform have results with low cohesion in the data of each class in terms of Euclidean averages and distances. Based on the above mentioned, a low-frequency analysis was developed with the Wavelet transform (using the approximation coefficients), determining the maximums of each signal that, when compared with Fourier, reflect even smaller distances between classes, that indicates that there is a disadvantage in the use of the Fourier transform. Additionally, as a result of the research developed in the antecedents [18, 30, 31] and using the coefficients of variation and Euclidean distance for the Wavelet, Cepstrum, and Fourier transforms, it is possible to conclude that the lowest percentages of dispersion correspond to the Wavelet transform. That suggests that Wavelet Transform is a technique that could be viable if average data are taken from each signal instead of point data (maximum) for the analysis and characterization of the misalignment and unbalance phenomena. Finally, the Wavelet transform applied in this project generates intermediate values closer to the Cepstrum transform. It implies that the distances of the phenomena related to the normal state are very close, and it indicates that the failure is not characterized as easy as the Fourier transform, which exhibits distances fifteen times wider than the Wavelet transform.
5 Conclusions In the signal processing, the samples must be of equal length and be continuous in time because it is necessary for comparing data; for that reason, a counter is included in the LabVIEW program to comply with this analysis condition. The comparison of the three transformation techniques Cepstrum, Fourier, and Wavelet using the average values, the coefficients of variation, and the Euclidean distances allow us to verify that the Fourier transformation in the frequency domain is a better descriptor of the misalignment and balancing phenomena. The mean square error supports what is concluded by the comparison of the Euclidean distance, including the obtained results using techniques such as the Cepstrum transform,
240
C. L. Sandoval-Rodriguez et al.
which is reported in the antecedents. Fourier transform allows analyzing the phenomenon with high discrimination capacities between classes and low values of the coefficients of variation, which permits detecting the phenomena studied in the present work. However, it is interesting to include the detail coefficients (high-frequency components) of the studied wavelets in the analysis, considering that it would be possible to lose some relevant information for the characterization at high frequencies when the Wavelet transform is working at low frequencies.
References 1. Cheung, S., Das, R., Mo, J.: Demystifying Numerical Models Step-by Step Modeling of Engineering Systems. Butterworth Heinemann (2019) 2. Cárdenas, C., Sandoval, C., Gómez, J.: Implementación de una mesa vibratoria triaxial neumática para el análisis de estructuras y el movimiento sísmico, Rev. Colomb. Tecnol. Av. RCTA, vol. 2, no. 32, Art. no. 32, November 2018. https://doi.org/10.24054/16927257. v32.n32.2018.3032 3. Blanco-Ortega, A., Beltrán-Carbajal, F., Silva-Navarro, G., Méndez-Azúa, H.: Control de vibraciones en maquinaria rotatoria, Rev. Iberoam. Automática E Informática Ind. RIAI, vol. 7, no. 4, Art. no. 4, October 2010. https://doi.org/10.1016/s1697-7912(10)70058-3 4. Romero-Tarazona, B.E., Rodriguez-Sandoval, C.L., Villabonai-Ascanio, J.G., RincónQuintero, A.D.: Development of an artificial vision system that allows non-destructive testing on flat concrete slabs for surface crack detection by processing of digital images in MATLAB. In: IOP Conference Series Material Science Engineering, vol. 844, p. 012058, June 2020. https://doi.org/10.1088/1757-899x/844/1/012058 5. Nakamura, T., et al.: Flow-Induced Vibrations, 2nd Edition. Academic Press (2013) 6. Tarazona, B., Sandoval, C.: Evaluación de discontinuidades tipo grietas y fisuras en estructuras de hormigón empleando un analizador de vibraciones y procesamiento digital de imágenes. In: Entre Cienc. E Ing., vol. 13, p. 85, June 2019. https://doi.org/10.31908/19098367.4018 7. Armijos, L., José, F.: Señales acústicas y de vibración: estudio comparativo para la detección de severidad de fallos en engranes rectos, p. 134 (2019) 8. González, H.Á., Cardona, J. F., Monroy, G.A.: Diseño de un banco de pruebas de desalineamiento y desbalanceo mecanico. Sci. Tech., vol. 2, no. 28, Art. no. 28, ene. (2005). https:// doi.org/10.22517/23447214.6827 9. Olarte, W., Botero Arbeláez, C.M., Cañon Zabaleta, B.: Técnicas de mantenimiento predictivo utilizadas en la industria. Sci. Tech., vol. 2, no. 45, Art. no. 45, ago. (2010). https://doi.org/ 10.22517/23447214.355 10. Bauer, B., Geropp, B., Seeliger, A.: Condition monitoring and predictive maintenance in mining industry using vibration analysis for diagnosis of gear boxes. IFAC Proceedings, vol. 30, no. 18, Art. no. 18, ago. (1997). https://doi.org/10.1016/s1474-6670(17)42529-8 11. Orhan, S., Akturk, N., Celik, V.: Vibration monitoring for defect diagnosis of rolling element - Technische Informationsbibliothek (TIB), vol. 39, no. 4, pp. 293–298 (2006) 12. Hurtado, J.E.R.: Analisis de vibraciones en equipos criticos de la industria azucarera trabajo de graduacion preparado para la facultad de ingenieria y arquitectura ingeniero mecanico, May 2020, Accedido: May 28, 2020. Disponible en. https://www.academia.edu/33077038/ analisis_de_vibraciones_en_equipos_criticos_de_la_industria_azucarera_trabajo_de_grad uacion_preparado_para_la_facultad_de_ingenieria_y_arquitectura_ingeniero_mecanico 13. White, G.: Introduccion-al-analisis-de-vibraciones-azima-dli.pdf, DLI. AZIMA DLI (2010) 14. Sanhueza, F.A.: Lozano, R.A., Duran, R.: Duran_Rivas_Ricardo.pdf, Pregrado, Universidad del Bio-Bio (2014)
Descriptive Study of a Rotary Machine Affected by Misalignment
241
15. Palomino Marin, E., La Medición, E., Análisis, D.: Vibraciones en el Diagnóstico de Máquinas Rotatorias, Cuba (1997) 16. Mosquera, G.: Las Vibraciones Mecanicas y Su Aplicacion Al Mantenimiento Predictivo, p. 205 (2018) 17. Alguindigue, I.E., Loskiewicz-Buczak, R.E., Uhrig, A.: Monitoring and diagnosis of rolling element bearings using artificial neural networks. IEEE Trans. Ind. Electron., 40(2), 2 (1993). https://doi.org/10.1109/41.222642 18. Sandoval Rodríguez, C.L., Barros, A., Herreño, S.: Clasificación automática de patrones de vibraciones mecánicas en maquinaria rotativa afectada por desbalanceo. INGEUAN - Tend. En Ing., vol. 4, no. 7, Art no. 7, October 2013. Accedido: jul. 26, 2020. [En línea]. Disponible en: http://csifesvr.uan.edu.co/index.php/ingeuan/article/view/255 19. Neale, M.: The Tribology Handbook, 2nd Edition (1995) 20. Panteli´c, M., Jovanˇci´c, P., Risti´c, L., Bebi´c, M.: Concrete base influence on the increased vibrations level of the mill drive system elements - a case study. Eng. Fail. Anal., 106, 104178 (2019). https://doi.org/10.1016/j.engfailanal.2019.104178 21. George, M., Balaji, J., Sajan, D., Dominic, P., Philip, R., Vinitha, G.: Synthesis and third order optical nonlinearity studies of toluidine tartrate single crystal supported by photophysical characterization and vibrational spectral analysis. J. Photochem. Photobiol. Chem., 393, 112413 (2020). https://doi.org/10.1016/j.jphotochem.2020.112413 22. Hashemnia, K., Pourandi, S.: Study the effect of vibration frequency and amplitude on the quality of fluidization of a vibrated granular flow using discrete element method. Powder Technol., 327, 335–345, March 2018. https://doi.org/10.1016/j.powtec.2017.12.097 23. Li, Z., He, Z., Zi, Y., Chen, X.: Bearing condition monitoring based on shock pulse method and improved redundant lifting scheme, vol. 79, no. 3, Art. no. 3 (2008). https://doi.org/10. 1016/j.matcom.2007.12.004 24. de la Rosa, J.J.G., Agüera Pérez, A., Palomares Salas, J.C., Sierra Fernández, J.M.: A novel measurement method for transient detection based in wavelets entropy and the spectral kurtosis: an application to vibrations and acoustic emission signals from termite activity, Measurement, 68, 58–69, May 2015. https://doi.org/10.1016/j.measurement.2015.02.044 25. Fan, G., Li, J., Hao, H.: Vibration signal denoising for structural health monitoring by residual convolutional neural networks. Measurement, 157, 107651, June 2020. https://doi.org/10. 1016/j.measurement.2020.107651 26. Randall, R.B.: A history of cepstrum analysis and its application to mechanical problems. Mech. Syst. Signal Process. 97, pp. 3–19 (2017). https://doi.org/10.1016/j.ymssp.2016.12.026 27. Shi, D., He, D., Wang, Q., Ma, C., Shu, H.: Free vibration analysis of closed moderately thick cross-ply composite laminated cylindrical shell with arbitrary boundary conditions. Materials, 13(4), 4 (2020). https://doi.org/10.3390/ma13040884 28. Sandoval Rodriguez, C.L.S., Tarazona, B.E., Arias, C.G.C., Javier Gonzalo Ascanio, V., Jhon Jairo Valencia, N.: Detection of structural alterations in metal bodies: an approximation using Fourier transform and principal component analysis (PCA), Sci. Tech., vol. 25, no. 2, Art. no. 2, June 2020. https://doi.org/10.22517/23447214.23501 29. Liu, Z., Zhang, L., Carrasco, J.: Vibration analysis for large-scale wind turbine blade bearing fault detection with an empirical wavelet thresholding method. Renew. Energy, 46, 99–110 (2020). https://doi.org/10.1016/j.renene.2019.06.094
242
C. L. Sandoval-Rodriguez et al.
30. Sandoval-Rodriguez, C.L., Villabona, J.G.A., Cárdenas-Arias, C.G., Rincon-Quintero, A.D., Tarazona-Romero, B.E.: Characterization of the mechanical vibration signals associated with unbalance and misalignment in rotating machines, using the cepstrum transformation and the principal component analysis. In: IOP Conference Series Material Science Engineering, vol. 844, p. 012057, June 2020. https://doi.org/10.1088/1757-899x/844/1/012057 31. Sandoval Rodríguez, C.L., Cardenas, C.G.C., Rincón-Quintero, A.D., Zanguña, J.A.: Extracción de rangos característicos a las condiciones de desbalanceo y desalineación en máquinas rotativas a partir de coeficientes cepstrum. presentado en V Convención internacional de la ingenieria en Cuba, Cuba, 2018, [En línea]. Disponible en: http://repositorio.uts.edu.co:8080/xmlui/bitstream/handle/123456789/887/Certificado% 20Ponente.pdf?sequence=1&isAllowed=y
Flexible Manufacturing Systems Optimization with Meta-heuristic Algorithm Using Open Source Software Fabian Izquierdo(B) , Edwin Garcia, Byron Cortez, and Luis Escobar Universidad de las Fuerzas Armadas - ESPE, Sangolqu´ı, Ecuador {feizquierdo,elgarcia,bhcortez,lfescobar}@espe.edu.ec
Abstract. This research develops the integration between the simulation of a flexible manufacturing system and the optimization of one of its resulting parameters by applying a meta-heuristic method based on genetic algorithms and discrete event simulation. The project focuses on the application of open-source software because the costs in implementing this integration with commercial software are high for the majority of existing companies in the country. The team used Jaamsim software in the simulation stage. It is a powerful simulator with a programmerfriendly graphical interface, developed in Java. For the construction and validation of the algorithm, the same program was used. The research provides a reference for the application of a tool that improves the productivity and competitiveness of companies in the market, focusing on the analysis of resources, planning, and observation in the behavior of a flexible manufacturing system to evaluate the accumulated times in their processes. Keywords: Flexible manufacturing system · Genetic algorithm Meta-Heuristic methods · Jaamsim · Open source software
1
·
Introduction
In a world where massive data have exploded, competitive indexes have also grown dramatically; therefore, a primary requirement in a manufacturing system setting is the correct process planning to optimize the increment in productivity [7]. In addition to proper planning, manufacturing systems need to be flexible to adapt to the fast changing needs and requirements of current dynamic markets. For instance, distributed manufacturing systems, where 3D printers are used, offer great flexibility [11,15,16] towards fulfilling customers needs [13]. Thus, only companies with excellent management of their processes can remain leaders in their production sectors. There are three points to consider to overcome the preceding from a technical point of view of manufacturing [18]: First: the problem is that each manufacturing case generates its unique production conditions, which entails different planning and flexibility for the manufacturing process to develop within a company. c The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 243–256, 2021. https://doi.org/10.1007/978-3-030-72212-8_18
244
F. Izquierdo et al.
Second: at present, there are software packages that simulate production environments known as discrete event software. However, most of them require an elevated fee. With high implementation costs, not very profitable for most Ecuador companies, where 89.6% are micro-enterprises, 8.2% are small, 1.7% are medium, and only 0.5% are large companies. Although there are also Open Source or Free Software options to fulfill this need, they are not focused on average users, since it requires medium to high knowledge of programming and engineering in manufacturing processes. Likewise, occasional use of the software and training of personnel in charge does not justify costs generated during the implementation for local companies to which this project is focused. Third: the choice of the algorithms to be implemented to try to cover several solutions to the exposed cases since at the programming level of a heuristic optimization algorithm there are several methods, among them: Local search, Swarm intelligence, Greedy Randomized Adaptive Search Procedure (GRASP), Reactive Search, Neuronal Networks, Forward Heuristic (BFH), Mixed Integer Linear Programming Formulation (MILPF), Simulated Annealing, Tabu Search, Petri Networks, it should be noted that the latest research was based on the last six algorithms cited.
2
Production System and Meta-heuristic Optimization Algorithms
Technically, an FMS “Is a group of machines or workstations related to each other and that perform a specific task” [1,4]. The author Groove, 1990 defines FMS as “groups of workstations interconnected employing a control system that can process different orders simultaneously under a numerical control code for each station” [10]. Production systems are related, organized, and interacting elements, whether they are people, materials, machines, management style, or procedures. All of this transforms the materials or information into a product or service dedicated to the sale. Once this cycle is finished and repeated when the systems used for industrial business production are analyzed, the processes can be optimized or transformed to be more efficient in terms of costs, delivery times, and quality [5,8]. When solutions are complex or difficult to find, the application of metaheuristic methods is necessary, with vast search spaces. Systematic exploration of the system is not feasible to find the best solution, so this kind of method eliminates searching paths that generate bad solutions. Based on a user-defined starting point, some heuristic or optimization methods are listed below [3,14]: 2.1
Hill Climbing
The method consists of applying iterative improvement techniques to solve statespace problems. It chooses at each step the successor with the best value based on the selected initial state. There are two types of escalation algorithms:
FMS Optimization with OSS
245
– Simple Climbing – Scales for maximum slope 2.2
Simulated Annealing
In this method, a successor is chosen from among all the possible ones according to a probability distribution. Said choice allows us to explore the space by improving or worsen the current solution. Worse states can be chosen (probabilistic). In this way, partially random steps are taken through the solution space looking for the best solution; this allows the algorithm to be able to leave local optimum. 2.3
Particle Swarm Optimization
This type of method takes inspiration from the social behavior of flocks of birds and the movement of schools of fish. These systems are composed of several particles that move through the search space during the algorithm’s execution. It is composed of simple entities with local actions (including interactions with the environment). The result of the combination of simple behaviors is the appearance of complex behaviors and the ability to achieve good results. The techniques of clouds of particles with intelligence are stochastic methods based on the population used in combinatorial optimization problems. Where the collective behavior of individuals arises from their local interactions with the environment to generate functional global patterns, the application of this method is typical in numerical optimization problems. 2.4
Ant Colony Optimization
It is a probabilistic technique to solve computational problems that can reduce the problem of finding the right paths through graphs. It is inspired by the behavior of ants, which are social insects that live in colonies, directed at group development as a whole rather than individual development. The behavior of ants works on the principle that each ant moves randomly. Deposit pheromones on the path, other ants, detect the main path (the one with the most pheromones) inclining to follow it, all the ants in a colony tend to follow the same path to go from their nest to the food. Considering that if there is an obstacle, they will quickly look for a new path to avoid it, and finally, they will follow the shortest route, through which the ants will pass faster and more times, leaving a more substantial pheromone trail. 2.5
Genetic Algorithms
Genetic algorithms are adaptive methods, generally used in parameter search and optimization problems, based on sexual reproduction and on the principle of survival of the fittest between sequences with structured, albeit random, information exchange, in order to build a search algorithm [9].
246
F. Izquierdo et al.
A genetic algorithm has a fixed sequence that it must comply with and is specified below [2]: – – – – – –
Generate an initial population Iterate to a stop criterion Evolve each individual in the population Select the parents Apply the crossover and mutation operator to those parents Include the new offspring to form the new generation
3
Design and Construction
For this project, the open source software has been used to find a solution taking in consideration the following aspects: – The development of an FMS station, the design, planning, simulation, and execution, various components are considered, for example, machines, robots, automated vehicles, entrances, exits; they are configured together. They must manufacture products through certain sequences in complex scenarios, which is why obtaining a graphic simulation tool that considers decision-making in the execution of all operations, correctly and understandably to the end-user, is a point that has taken into account. – The simulation software must present the following characteristics: flexibility, ease of use and integration, reasonable cost, animation, and debugging capacity, possess statistical and data management capabilities, have documentation and support. – The result of the simulation of a discrete event system must be evaluated by applying the meta-heuristic algorithm, which optimizes the production of the generated products. 3.1
Jaamsim Modeling
It is a free, open-source simulation package written in Java programming language. It provides a modern graphical user interface (GUI) comparable to commercial software, including drag and drop options (Drag and Drop), model building, an input editor, output viewer, and 3D graphics. Users can create their personalized, high-level object palettes using standard Java and modern programming tools like Eclipse [12]. The simulation in Jaamsim and subsequent evaluation of its performance employing a developed genetic algorithm. For this research, as seen in Fig. 1, an FMS is considered that has [6]: – An input/output station for raw material and finished product, – Four stations with machines for product processing, – A buffer for AGV’s (Automatic Guided Vehicles) that will transport the products between stations.
FMS Optimization with OSS
247
Fig. 1. FMS block diagram
Based on previous points, the need for methodologies and methods to face the problems is clear, for which the following restrictions have been taken into account [6]: – All products have to be processed in the same FMS. – All raw materials are ready for processing and have a constant arrival rate at the entry/exit station. – Each Machine can handle one operation for one job at a time. – All jobs will go through the buffer for distribution among the machines until they are fully processed. – The distances between stations are equidistant. – The sequences of each job and the service times for each product at the stations are predefined. – The first job in the queue will be assigned to the corresponding available station. – Once the operation has started, a machine cannot be stopped. The flow diagram shown in Fig. 2 explains the working methodology of the proposed FMS cell. The manufacturing process of the batch of products begins when a production order is received. The order consists of a variety of elements with their respective sequences and operating times. If this order does not contain the correct parameters, the simulation will reject the order, and the values entered must be verified again to continue with the process. Once the production order is valid, the raw material enters into the central buffer. The AGVs will be in charge of distributing the orders based on the availability of the processing machines and the production sequence assigned to each element. Items that have completed their production sequence are sent to the departure station, where it will confirm whether the entire batch has finished its production. In this last step, the total production time is evaluated as a product of the simulation. For didactic reasons in Fig. 3, the simulated model is divided into seven sections, which describes the relationship presented in the flow diagram in Fig. 2 [17].
248
F. Izquierdo et al.
Fig. 2. Flowchart for simulation in Jaamsim
The description of operations is as follows: 1. The objects, FileToMatrix (File to matrix), will enter three-parameter blocks for the simulation through text files with a txt extension, which contains the sequences assigned to each product. These represent the production sequences, the times assigned to each station, and the distance between stations. 2. SimEntity (Simulation Entity) and EntityGenerator (Entity Generator) are the objects that generate the raw material.
Fig. 3. Jaamsim operating blocks
FMS Optimization with OSS
249
3. By combining blocks: Assign, Discrete Distribution, Branch, and Statistics, the assignment of the elements generated in the previous point is set to be evaluated in the buffer of AGV’s and distributed for production. 4. The AVG’s Buffer is in charge of validating the processed sequences of each product and transferring it to its corresponding station. Check availability in the allocation of operations and dispatch finished products to the exit station. Its operation is similar to that of a server that, through the interaction of Assign, AddTo, Branch, and RemoveFrom blocks, assigns the AVG’s for distribution tasks. 5. They represent the stations where the four machines are, through which the products must pass. It should be noted that, in the simulated model, the input/output process (2/6) is also considered a station. Additionally, an average utilization per station is shown. The blocks that interact in this section are Seize (Receiver), Resource (Resources), and Release (Liberator), where the reception, process, and dispatch of a product are simulated. 6. The exit station shows the end of elements processed in a specific time range. 7. This section reflects statistical data from the simulation by validating properties in the process flow. It is achieved based on Statistics blocks. 3.2
Meta-heuristic Algorithm
Considering that the code was generated from conditions and characteristics of the genetic algorithm, as detailed in the flow diagram in Fig. 4. The same process is also observed in the code programmed in Fig. 5, which is the primary function of the target program to call each of the functions in a predetermined order, as shown here: – – – – – – –
Start population Convert population Individual evaluation Parents Selection Tournament Combination - Mutation Set new population
The main code of the program calls each one of the functions of the genetic code process, such as: create population, adaptability, tournament, duplicate, print, mutation, and optimal sequence selection, as shown in Fig. 5. All this process is sequential. The part of the code “convert individuals” is responsible for transforming the chromosome or the sequence of batches of each individual to a total production time based on a coding format and delivery times per product as detailed in Table 1.
250
F. Izquierdo et al.
Fig. 4. Genetic algorithm flowchart
Fig. 5. Genetic code main function
FMS Optimization with OSS
251
Table 1. Coding for chromosome sequence Chromosome Information
3.3
Representation
Delivery Time
1
Product A
8
2
Product B
10
3
Product C
12
4
Product D
14
System Flow Diagram
The following flow diagram demonstrates the relationship between the Jaamsim modeling sections and the metaheuristic algorithm of the present investigation. In Fig. 6 the process and union of the two processes are presented.
Fig. 6. Metaheuristic flow diagram
Within the Jaamsim environment, a case study was simulated. This simulation reflects a general production time per initial batch. (Makespan). The genetic algorithm will evaluate the batch. This algorithm will produce a new
252
F. Izquierdo et al.
batch, which must be evaluated in the simulation. If the new batch produces a better Makespan, one of the objectives of this research will have been achieved.
4
Validation Methodology
In this section, the metaheuristic method is applied to a Case Study that wishes to produce a batch of 10 units of 4 different types of products. In an FMS with similar characteristics, the routines for the different products are specified, as shown in Table 2. Table 2. Product Specifications Product Operation 1 Machine
Time
Operation 2 Machine
Time
Operation 3 Machine
Time
Operation 4 Machine
Quantity
Time
A
1
2
2
2
3
2
4
2
3
B
2
2
4
4
3
2
1
2
4
C
1
3
2
2
3
3
4
4
2
D
3
5
1
2
4
4
2
3
1
An initial production batch is established according to the quantities requested in the work order, which is considered the initial population to start the tests. The products are coded as follows: – – – –
A=1 B=2 C=3 D=4
The population must be entered in the Input Editor corresponding to the discrete distribution named JobTypeDistribution, where the required production order is entered. This value can be taken as a vector. The ValueList field lists the product types and the ProbabilityList the quantity (Fig. 7).
Fig. 7. Production order entry to model in Jaamsim
FMS Optimization with OSS
253
Fig. 8. Verification of values in simulation
The simulation must be run with the data entered; and verify the quantity produced in a particular time until adjusting the 10 units analyzed for the case study in the JobDeparture object (Fig. 8). As exhibited in Table 3, the sequences generated through the algorithm do not change drastically; This is because our initial population has not presented a significant mutation over time. The sequence is based on the evolution of individuals. In iteration 2, an improvement is achieved, and again in iteration 4, Table 3. Sequences and Evaluation Parameters GA
KP
PP
T ES P T R
IN(OP) [A A A B B B B C C D] [0.3 0.4 0.2 0.1] 1.79
0.00
RD 0.00
IT 1
[A A A A B C B B A D] [0.5 0.3 0.1 0.1] 1.625 −9.22 0.165
IT 2
[A A A A B C B B A D] [0.5 0.3 0.1 0.1] 1.625 −9.22 0.165
IT 3
[A A A A B C B B A D] [0.5 0.3 0.1 0.1] 1.625 −9.22 0.165
IT 4
[A A A B B C B B A D] [0.4 0.4 0.1 0.1] 1.71
−4.47 0.080
IT 5 [A A A B B C B B A D] [0.4 0.4 0.1 0.1] 1.71 The abbreviation used in Table 3 is detailed below: – GA = Genetic Algorithm – KP = Kind of Product – PP = Production Probability – TES = Time Evaluation in Simulator (Hours) – PTR = Percentage of time reduction (%) – RD = Reduced Time (H) – IN(OP) = Initial Population (Order Of Production) – IT = Iteration
−4.47 0.080
254
F. Izquierdo et al.
a worsening of the evaluation is also presented. They are showing the capacity to escape from the local maximum. The production batches are arranged by utilizing the algorithm generating a better makespan than the base batch. The present reduction is not significant. However, considering this large-scale variation, a distinctive difference is seen under this criterion.
5
Results
Within the objectives established in the research, it is determined to use OpenSource Software that allows the user to model a flexible manufacturing system and optimize one or more parameters of its design. It is highlighted that the object of study has fulfilled its purposes, denoting a reduction after the application of the algorithm as an analysis tool; the simulated model is fed back for its evaluation. The results are appreciable in Fig. 9. Figure 9(a) shows the iterations that have been presented as the result of different executions of the algorithm evaluated within the model created in Jaamsim, denoting a reduction in production times, it is clear that in each iteration performed a change in the makespan is present, this improvement is also highlighted when observing the reduction rate of batches per production order since the simulation indicates a reduction of 5.23% in its new iterations compared to the execution time of the initial population is shown in Fig. 9(b), with this it is shown that the algorithm has the ability to leave local minimums and maximums, understanding that a higher number of iteration can optimize the initial sequence of production of manufactured batches.
Fig. 9. Analysis of Results. a) Time evaluation per batch. b) Absolute percentage reduction in production times
Since one of the most significant limitations is the uniqueness of each process, its simulation is unique. The case study presented would show a different trend in its values when there is a change in variation of restrictions, product sequences, and alteration in the arrival rate of raw material. The different treatment prioritizes the queues for each machine or the use of different statistical distributions.
FMS Optimization with OSS
255
Another factor to consider is the number of interactions of the product, or the degree of mutation for the generations of the population, which will determine individuals with better characteristics. The focus of the tests was based on obtaining a sequence with better performance in its makespan. Not always obtaining the initial products, this is interpreted that there will be a product in inventory that will serve to meet new orders. Since the design is flexible, it will be enough to change the desired values’ production order.
6
Closing Remarks
In the present research, an open-source heuristic algorithm was developed. Capable of optimizing production sequences based on genetic algorithms of n individuals with n chromosomes and determining the most optimal sequence for automated manufacturing systems, as well as finding a better solution for combinatorial problems. With the number of iterations evaluated for the initial population, it was possible to determine the best sequences and decrease production times. Where the dominant results were obtained in the crosses of individuals (children), solutions were also presented in the mutated individuals. However, unlike the previous ones, they presented not a better solution, demonstrating the ability to avoid local maximum. Considering that there were random variations, the new populations obtained did not present drastic alterations to the initial sequence; however, they improved in a good percentage the makespan, determining that the program fulfills its function of optimizing. Jaamsim, as a discrete event simulator, turned out to be a powerful tool with a user-friendly interface.
References 1. Bernal, M., Sarmiento, C., Restrepo, J.: Productividad en una celda de manufactura flexible simulada en promodel utilizando path networks type crane. Universidad Distrital Francisco Jos´e de Caldas (2015). https://www.redalyc.org/pdf/2570/ 257036222011.pdf 2. Caceres, R., Rivas, W.: Desarrollar un software planificador de rutas, para encontrar una ruta o ´ptima, mediante algoritmos gen´eticos (jgap) con interfaz desarrollada en java. Unidad Acad´emica de Ingenier´ıa Civil - UTMACH (2017) 3. Caldas, A., Carpente, M., Lorenzo, S.: Aplicaci´ on de algoritmos heur´ısticos para optimizar el coste de doblaje de pel´ıculas. Universidade da Coru˜ na (2014) 4. Cruz, J., Badii, M.: SMED: El camino a la flexibilidad total, vol. 1 (2004) 5. EAE Business School: Tipos de sistemas de producci´ on industrial y sus caracter´ısticas (2018). https://retos-operaciones-logistica.eae.es/tipos-de-sistemas-deproduccion-industrial-y-sus-caracteristicas/ 6. Escobar, L.F.: Adaptive scheduling of flexible manufacturing systems with material handling constraints using GA and Colored Timed Petri Net. Jap´ on (2015). http:// www.f.waseda.jp/t-murata/Alumni.htm
256
F. Izquierdo et al.
7. Flores, K., Medina, P.: Desarrollo de un software acad´emico para programar la producci´ on en los sistemas de manufactura flexible. Universidad Tecnol´ ogica de Pereira Facultad de Ingenier´ıa Industrial (2013). http://recursosbiblioteca.utp.edu. co/tesisd/textoyanexos/0053F634.pdf 8. Garc´ıa, E., Escobar, L.: Desarrollo de un simulador de sistemas de manufactura con interfaz gr´ afica basado en redes de petri. Revista iberoamericana de ingenier´ıa mec´ anica (2018). https://dialnet.unirioja.es/servlet/articulo?codigo=6484717 9. Gestal, M., Cebri´ an, D., Rabu˜ nal, J., Pazos, A.: Introducci´ on a los algoritmos gen´eticos y la programaci´ on gen´etica, 1st edn. Universidade da Coru˜ na, Servizo de Publicaci´ ons, Espa˜ na (2010) 10. Groover, M.: Introducci´ on a los algoritmos gen´eticos y la programaci´ on gen´etica, 1st edn. Prentice Hall Press Upper Saddle River, New Jersey (2002) 11. Huang, J., Segura, L.J., Wang, T., Zhao, G., Sun, H., Zhou, C.: Unsupervised learning for the droplet evolution prediction and process dynamics understanding in inkjet printing. Addit. Manuf. 35, 101197 (2020) 12. JaamSim Software Inc: JaamSim User Manual (2017). https://jaamsim.com/docs/ JaamSim%20User%20Manual%202017-10.pdf 13. Matt, D.T., Rauch, E., Dallasega, P.: Trends towards distributed manufacturing systems and modern forms for their design. Procedia cirp 33, 185–190 (2015) 14. Rabanal, P., Rodr´ıguez, I., Rubio, F.: Algoritmos heur´ısticos y aplicaciones a m´etodos formales. Universidad Cumplutense de Madrid Facultad de Inform´ atica (2010). https://eprints.ucm.es/12027/1/T32515.pdf 15. Segura, L.J., Zhao, G., Sun, H., Zhou, C.: Gaussian process tensor responses emulation for droplet solidification in freeze nano 3D printing of energy products. In: International Manufacturing Science and Engineering Conference, vol. 58745, p. V001T01A024. American Society of Mechanical Engineers (2019) 16. Segura, L.J., Zhao, G., Zhou, C., Sun, H.: Nearest neighbor gaussian process emulation for multi-dimensional array responses in freeze nano 3d printing of energy devices. J. Comput. Inf. Sci. Eng. 20(4), (2020) 17. Teruel, E., Arag¨ u´es, R.: Aprendiendo simulaci´ on de eventos discretos con jaamsim. pp. 522–527. Servicio de Publicaciones de la Universidad de Oviedo (2017). https:// dialnet.unirioja.es/servlet/articulo?codigo=6591655 18. Valtierra, J., Sausedo, J.: Reconfiguraci´ on aut´ onoma de sistemas de manufactura mediante la optimizaci´ on de funciones de desempe˜ no del proceso. 12th Latin American and Caribbean Conference for Engineering and Technology (2014). http:// www.laccei.org/LACCEI2014-Guayaquil/RefereedPapers/RP016.pdf
Estimation of the Energy Consumption of an Electric Utility Vehicle: A Case Study Gianina Garrido-Silva1 , Jessica Gissella Maradey-Lazaro1 , Arly Dario Rincón-Quintero2,3(B) , Omar Lengerke-Pérez3 , Camilo Leonardo Sandoval-Rodriguez2,3 , and Carlos Gerardo Cardenas-Arias2,3 1 Universidad Autónoma de Bucaramanga UNAB, Bucaramanga Santander 680003, Colombia
{ggarrido,jmaradey}@unab.edu.co 2 University of the Basque Country UPV/EHU, 48013 Bizkaia, Spain
{arincon,csandoval,ccardenas}@correo.uts.edu.co 3 Unidades Tecnológicas de Santander UTS, Bucaramanga Santander 680005, Colombia
[email protected]
Abstract. The development of electric vehicles (EV’s) has been growing in last decade since is a promising technology that will optimize the use of energy. The estimation of energy consumption is a key task to design and select components of power train, control strategies and predict lifecycle. Some factors such as road slope, temperature, type of route, driver’s behavior directly affect the energy consumption. This article provides an easy methodology to estimate energy consumption for an electric utility vehicle (EUV) using Green Race Software, which allows different types of routes, estimate a road slope, energy consumption and percentage of regeneration. From this analysis, is possible to decide the best route for harvesting cocoa and sizing the powertrain of the vehicle before purchasing materials. The most important advantage of the proposal method is that can be used in early stage of design and assembly of EV’s. Keywords: Energy consumption · Electric vehicle · Vehicle autonomy
1 Introduction The design of electric vehicles constitutes one of the main areas of current research in the mobility sector, especially in urban roads [1]. Similarly, to replace fossil fuels with renewable energy sources, making efficient use of energy, zero emissions, and including smart technologies for estimation, prediction and simulation of systems are global goals [2, 3]. In this regard, an accurately energy consumption estimation for EV’s is a need and critical aspect to minimize the anxiety state of the drivers [4], to localize and manage charging stations and to select the best options to the components such as: batteries, motors, sensors, gears, among others [5]. Besides, the successful development of an electric vehicle model is achieved by means of the combination of items: power, energy consumption, weight, autonomy, and cost [1]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 257–272, 2021. https://doi.org/10.1007/978-3-030-72212-8_19
258
G. Garrido-Silva et al.
Some factors that affect directly energy consumption in EV’s such as: driving conditions (i.e. road slope and gradients, type of route, topography, temperature, wind speed, traffic), driving behaviour (i.e. age, gender, emotional state, habits, driving style), vehicle operation factors (i.e. rolling resistance, aerodynamic drag coefficient, forces, powertrain, gear-box) and auxiliary systems (i.e. air-conditioning and heating systems) [4–6]. Figure 1 shows a schematic map of mentioned aspects. So, other aspects that can be considered are state of charge of battery (SOC), quantity, degradation, type, the influence of the temperature, energy storage [6] and the efficiency of braking energy recovery [1, 5]. An example of study related to influence of the environmental aspect, specially temperature, is reported to [7, 8], however in many studies these effects are not taken into account due to insufficient travel data [5].
Fig. 1. Factors that affect energy consumption in EV.
Commonly, official driving cycles (DC) have been used to determine the energy consumption, driving range, SOC and optimize the energy management system (EMS). However many authors have reported differences between official driving cycles and real driving cycles which based on road information [9, 10]. Every driving cycle is unique according to the study region and the mechanical configuration of the EV is different of internal combustion engine vehicles (ICEV’s) and the purposes too. Some limitations in the development of cycles is the amount of data that must be collected so that it is representative of how people drive and the precise and stable instrumentation necessary [11]. Eco-driving concept appear how as a logical way to improve energy and mechanical efficiency through the use of several sensors able to evaluate the current condition of the vehicle and can determine the current optimal state, inform the driver how to operate the actuator such as an optimal gear and improve the driver’s experience [12]. Some driver’s habits can generate high energy consumption such as: late braking, incorrect shifting and Failure to comply with the prescribed average driving speed for a given route. Then, the principal rules of eco-driving are shift up as soon as possible, maintain a steady speed, anticipate traffic flow and decelerate smoothly [11]. To sum up, eco-driving can be favored by improving consistency between vehicle dynamics, longitudinal road profile and the succession of speed limits [13]. In addition, there are exist models of simulation, estimation, and prediction of energy consumption of EV’s, which are based on road information and could be off-line or online too. The road information can be extracted though the Shuttle Radar Topography Mission (SRTM) and OpenStreetMap (OSM) data [14]. Table 1 illustrates a summary of the models found in literature. Consequently, the prediction of range is more complex
Estimation of the Energy Consumption of an Electric Utility Vehicle
259
because depends on stochastic factors (i.e. driving behaviour, traffic, topography, vehicle characteristics and environmental conditions) [5]. An accurately prediction of driving speed and road information can improve the estimation of energy consumption and range or autonomy [4]. ADVISOR, AUTONOMIE, PAMVEC are a package software to modeling energy consumption based on vehicle characteristics which are input variables and parametric analytical expressions from road load equations. The error of these tools is less than dynamic simulation models that are more expensive in terms computationally. Microsimulations can be made using VISSIM, PARAMICS and AIMSUN software. The advantage is that include traffic information but do not considered a road slope [3]. Nevertheless, in general, the main limitation of EV modelling is a suitable calibration model from real-data information. Although there exist databases such as spritmonitor.de, fuelly.com and honestjohn.co.uk not all consider air velocity and temperatures but can be used to reference [3, 15]. For activities in the agricultural sector, utility vehicles, which are used for harvesting and transporting fruits in rural areas. These vehicles are subjected to uneven terrain, steep slopes and where the power requirement is greater. Also, they are continuously subjected to acceleration and braking cycles, which causes higher energy consumption. Another point is that these types of vehicles do not have a gearbox. Considering that in these areas there are no electric charging stations, it was decided to design a battery electric vehicle. To design the electric utility vehicle, it is important to know the capacity of the motor and the batteries, that is, the powertrain and since there is no real data it is necessary to use simulation models for this task. This article shows an easy methodology to select the powertrain of an electric utility vehicle (EUV) for cocoa harvesting using the Green Race Software, when there is no road data or the opportunity to have real data second by second, since the design starts from scratch. The results are a comparison of routes considering the road slope, temperature, energy consumption and percentage of energy regeneration.
2 Methodology The methodology developed is based in the range estimation of the utility electrical vehicle, which depend of the road condition, specially road slope and topography. These factors have influence on cocoa harvest task in the selected zone for study. Also, is necessary to study of energy consumption of the batteries, which feed the motor of the electric vehicle as well as defines future works to localize a enough fast charging stations along the route and select a specific route to transport the fruit to its storage place, guaranteeing, at the same time, a good harvest done and timely, avoiding damage and loss of care, contributing to the reduction of sources of inoculum, thereby maintaining the health of the farm [22]. Comparisons taking into account different roads must be considered. Therefore, the stages of the methodology are: Scope, Data collection and Assessment, as shown in Fig. 2. 2.1 Scope Cocoa is a Colombian traditional peasant economy crop planted in small or mediumsized plots with productive units of 3.3 ha on average. This sector of the industry presents
260
G. Garrido-Silva et al.
low productivity levels due to the fact that scarcely 450 dry kilograms per hectare are produced per year, this is generated since the cocoa bean production system is characterized by areas cultivated under conditions of intensive use of family labor, little technology and areas of difficult access such as the non-interconnected zone (i.e. ZNI) which prevent access to constant electricity. In accordance with the above, an electric utility vehicle was designed which allows sowing and harvesting cocoa in areas suitable for cultivation with slopes of up to 29° inclination and mountainous terrain (see Fig. 3). Table 1. A summary of energy consumption models Author
Year Energy Hight lights consumption model
Petterson et al. [16]
2019 Estimation
A statistical operating cycle of the physical properties of road using stochastic models
Milesich et al. [11]
2018 Estimation
A hierarchy of neural network model based on CAN bus and GPS sensor. Eco-driving
Luin et al. [17]
2017 Estimation
Road grate estimate using satellite elevation data (OPenStreet Map). Vehicle Specific Power (VSP) model sensitivity analysis
Wang et al. [5]
2017 Estimation
Based on GPS observations. Traditional linear regression and multilevel linear regression Temperature behaviour and relation with energy consumption
Jimenez et al. [18]
2017 Estimation
Based on-board information. Characterization of the behavior of drivers using only the sensors in a smartphone. Electric Vehicle Consumption Model includes events validation
Mcnerney et al. [19] 2017 Estimation
TripEnergy. Reconstructs driving behavior across USA and simulate vehicle performance for different driving conditions
Fiori et al. [20]
A simple EV energy model using second-by-second vehicle speed and roadway grade data. The Virginia Tech Comprehensive Power-based EV Energy Consumption Model ( VT-CPEM). Instantaneous braking energy regeneration. 5.9% error vs empirical data
2016 Estimation
(continued)
Estimation of the Energy Consumption of an Electric Utility Vehicle
261
Table 1. (continued) Author
Year Energy Hight lights consumption model
Wang et al. [21]
2016 On-line Prediction
An algorithm to adjust the energy consumption prediction during driving including vehicle parameter estimation and driving behavior correction. Verified by 21 driving tests, in different type of roads. The error was within 5% for every test
Duel et al. [10]
2014 Estimation
A novel framework shows environmental effects including dynamic traffic assignment model
Fig. 2. Methodology proposed.
Fig. 3. A) A mountainous terrain photo with a high slope and representation of batches 1, 2 and 3. B) Cocoa cultivation planted in ridges.
In addition, Cocoa cultivation is generally sown using the ridged technique, which allows the leaves that fall from the trees, plus those generated by pruning, to be stored and serve as fertilizer to the land (see Fig. 3). The cocoa harvesting route is carried out for four batches, which can be seen in Figs. 3 and 5; and whose technical characteristics are defined in Table 2. So, energy consumption varies in accordance with road type due to the frequency and range of acceleration and deceleration in a travel distance. Finding the most economical route can be found by simulation [23] and is possible too, make comparison between candidate routes. The problem of route planning and schedule have been extensively studied [24]. Then, using of the Green Race 5.2 Software, the route is traced according to the longitude and latitude of the area in which the batch crops are located, as evidenced
262
G. Garrido-Silva et al.
in Fig. 4b. It is important to take into account that they are located 8 indicators, which can be seen in Fig. 4b, to establish the work routes and thus be able to collect a greater amount of data and thus observe a better performance of energy consumption for each incline slope detected by the software. Some mathematical model of energy consumption and dynamic programming could be included in the analysis. Once the work routes have been defined, the parameters requested by the software are selected to determine the autonomy of the utility vehicle. These parameters are described below:
Fig. 4. A) A batch 4 photo. B) Collection route for the four batches. Table 2. Technical characteristics of batch crop Batch
Number of plants
Sown area [m2 ]
Anual harvest
Batch 1
673
6057
March-April
Batch 2
660
5940
Batch 3
603
5427
Batch 4
1523
13707
September-November
– Dead-weight: the mass of the car without its passengers is defined, given in Kg. – Aerodynamic drag coefficient: for its calculation, equation A is used in which the drag force is due to the forces generated on the vehicle in the orientation parallel to the direction of advance and in the opposite direction.
Fz =
1 × d × V 2 × A × Cx 2
(1)
where, d = Air density kg/m3 2 A = Front surface m V = Work speed (m/s) Cx = Aerodynamic drag coefficient – Air density: According to the Mining and Energy Planning Unit (UPME) in its Annex 4 of the Atlas of Wind and Wind Energy of Colombia, the air density ranges between
Estimation of the Energy Consumption of an Electric Utility Vehicle
– –
– –
– –
263
0, 7 kg/m3 over high mountain areas and 1, 2 kg/m3 in places located at sea level without showing significant changes in its seasonal part. Therefore, it is estimated that there may be variations ±0, 1 kg/m3 depending on the time of year, although in principle this variation casn be neglected. For the case study, the value of 0, 7 kg/m3 is estimated on high mountain areas. Working speed: the maximum working speed is defined, given in (m/s) Front surface: the vehicle was parameterized using the SolidWorks software in which measurement values can be viewed as surfaces. For the case study, its front surface is 3.54 (m/s). Figure 5 shows the elevation of the vehicle, whose surface depends on the height, width, and shape of that elevation (including tires, mirrors and everything that is exposed to the air in the direction of travel). To determine the aerodynamic coefficient, the SolidWorks software is used again and the air flow simulation is carried out, using the Finite Volume method of the vehicle with the Flow Simulation software. To do this, the parameters summarized in Table 3. Must be considered. In the case of temperature and humidity in the case study area, the values are determined according to the data provided by NASA’s EOSWEB program, which are referenced in Table 3. Battery system energy: the technical characteristics of the batteries used are defined, highlighting their energy capacity. Performance of the traction chain: according to the performance of the AC-50 motor used and the electrical system, including the discharge of the batteries and the control chain, the average value is estimated to represent this performance. Regenerated energy: a percentage of excess energy that can be recovered during braking is estimated. Rolling resistance coefficient: from the values in Table 5, the rolling resistance coefficient is selected according to the state of the ground.
Fig. 5. A view of the utility electric vehicle.
2.2 Data Collection The study of the behavior of the energy consumption of each route is carried out using the Green Race 5.2 software to determine the number of times the vehicle must recharge the powertrain batteries and thus indicate a good management of fast charging to maintain the life cycle of the batteries and their energy autonomy. The software makes an
264
G. Garrido-Silva et al. Table 3. Parameters Definition using Flow Simulation software Parameters Thermodynamics
Speed
Humidity
Pressure ([Pa])
Direction X (m/s) Relative (%)
Temperature ([K]) Direction Y (m/s) Direction Z (m/s)
estimate every 10 [m] at a certain slope of energy consumption at different heights. The consumption value may increase according to the selected rolling coefficient, which is defined by the type of terrain; and additionally, to the value of the slope by selected route. However, the software generates a table in which it estimates the distance traveled, the critical slopes, the energy consumption per slope, the regeneration, and the autonomy per route. 2.3 Assessment The above data allow to collect enough information to be able to estimate according to the defined route, the distance, the energy consumption of the batteries, the autonomy and the number of necessary laps that the vehicle can make before recharging. It is important to highlight the type of charge to be carried out, since the generic chargers that come with this type of system are programmed for different levels of charge: slow or fast, by means of control algorithms. The control algorithms make it possible to guarantee the equalization of the 330 cells of each battery module, which, for this case, corresponds to 1,980 cells, since there is a set of 6 battery modules in the powertrain. This equalization determines the amount of charge that a cell requires to be dissipated or transferred [26]. For the battery configuration carried out in the powertrain, which corresponds to a parallel series set (3p2s), a full charge analysis is performed in three hours as shown in Fig. 6, where an algorithm is evidenced slow charge control starting with high current loads (8.5 (A)), which slowly decreases, until reaching a little more than the threshold voltage (114 (V)) with a charging current of 3 (A). Finally, it delivers a high current of up to 6.5 (A), causing the voltage to rise rapidly until it reaches approximately its maximum working voltage (122 (V)) (Table 4). With the previous data, it is possible to estimate the theoretical value of the vehicle’s autonomy, according to Eq. 2. Autonomy =
(Total distance × Total battery consumption) (Total route consumption)
Where, Autonomy (km) Total Route (km). Total battery consumption (kWh) Total route consumption (kWh).
(2)
Estimation of the Energy Consumption of an Electric Utility Vehicle
265
Table 4. Average temperature and humidity for study zone ◦ Monthly temperature C Jul
Aug Sep Oct
Average temperature Nov Dec ◦ C
Jan
Feb Mar Apr May Jun
32,9
35,8 35,2 31,9 29,8 30,0 31,6 32,5 30,9 29,0 28,9 29,7 31,52
Monthly relative humidity (%) Jul
Aug Sep Oct
Average Nov Dec relative humidity (%)
Jan
Feb Mar Apr May Jun
65,6
59,7 66,0 76,5 80,1 76,3 69,9 69,0 75,8 81,8 81,5 76,1 73,3
Table 5. Rolling coefficient values for different terrains [25] Nature and state of the soil Rolling resistance coefficient (k)1 Road in good condition
0,02 a 0,04
Affirmed dirt road
0,03 a 0,05
Dirt road
0,04 a 0,06
Waste ground
0,06 a 0,10
Dry stubble
0,08 a 0,10
Tilled land
0,10 a 0,20
Sand and very loose soil
0,15 a 0,30
Fig. 6. Charging behavior of the battery configuration 3p2s.
3 Results The tests are carried out using Green Race 5.2 software, the parameters necessary for simulating the behavior of the electric vehicle are determined according to the defined route.
266
G. Garrido-Silva et al.
– Deadweight: A deadweight of 400 [kg] is determined without passengers or cargo. – Aerodynamic drag coefficient: to determine this value, the density of air over high mountain areas is estimated for its calculation with a value of 0, 7 kg/m3 , a working speed of 30 (km/h) or its equivalent to 8, 33 (m/s) and a front surface of the vehicle of 3, 5389 m2 . However, to find the drag force, an average temperature of ◦ 31, 52 C ∼ = 304, 67 (K) and a relative humidity of 73.3 (%) are considered. These data allow simulating the behavior of air speed to determine the drag force exerted on the electric utility vehicle and thus be able to determine the aerodynamic coefficient. The simulation describes the behavior of the air velocity in the z axis and, in addition, shows a value of the drag force of 81.54 (N), as well as its behavior in Fig. 7.
Fig. 7. Behavior of air speed in the electric utility vehicle at a speed of 30 (km/h).
Thus, in this case the air resistance is higher when moving and, therefore, it does not take advantage of the power provided by the engine. According to the previous data and considering Eq. (1), a value of the aerodynamic drag coefficient of 0.948 is calculated. – Battery system energy: According to the data sheet of the Tesla Smart Lithium-Ion 18650EW batteries, the energy capacity per module is 3 (kWh), which represents a total energy of the battery system for the case study of 18 (kWh). – Traction chain performance: according to the technical characteristics of the HPEVS AC-50 motor, its performance is 85 (%). For calculation purposes, a value of 75 (%) is taken which includes the discharge of the batteries, the control chain, slow engine turning at start-up or on steep slopes and in the absence of a gearbox. – Maximum speed: this speed corresponds to an estimated maximum working value of 80 (km/h), since a higher speed is not required, due to the terrain and the weight per load of the vehicle; since it must be guaranteed that the fruit reaches its storage in good conditions. – Regenerated energy: regenerated energy represents 30 to 50 (%) of the system’s electrical supply [27]. For the case study, a value of 30 (%) of recovered energy is taken, during braking. – Accessories that demand power: No additional accessories are incorporated, since the secondary circuit of the vehicle is powered by photovoltaic solar panels. – Rolling resistance coefficient: according to table D, a value of 0.05 is selected for dirt roads, which corresponds to the working terrain. According to the previous results, Table 6. Summarizes the variables obtained for the simulation of the system. Each label in Fig. 4b indicates the route of the vehicle, which is divided into seven sections of the route: A-B, B-C, C-D, D-E, E-F, F-G and G-H. For this case, only
Estimation of the Energy Consumption of an Electric Utility Vehicle
267
Table 6. Variables obtained for simulation by Green Race Software Variable
Value
Dead-weight
400 (kg)
Aerodynamic drag coefficent
0, 948
Energy of batteries system
18 (kWh)
Traction chain performance
75 (%)
Máxima velocidad de trabajo
80 (km/h)
Regenerated energy
30 (%)
Accesories that demand power 0 (W ) Rolling resistance coefficient
0, 05
the route with critical slopes is indicated with to determine the behavior of the energy consumption of the batteries, whether positive or regenerated. On the other hand, the GreenRace 5.2 software with the vehicle characteristics and the values indicated above, shows that the utility vehicle can achieve a range of 239.03 (km), a total distance traveled of 9.68 (km), an energy consumption of 7.53 (kWm), every 100 (km) at an altitude of up to 1200 (m). The study of the energy consumption behavior of each route is carried out with the GreenRace 5.2 software, taking for each one, 200 data energy consumption derived from the vehicle’s behavior according to the established routes, their distances and work slopes, with in order to determine the times that the vehicle must recharge the batteries and thus indicate a good management of fast charge to maintain the life cycle of the batteries and their autonomy. The software makes an estimate every 10 (m) at a certain slope of energy consumption at different heights. Next, the behavior of energy consumption and the regeneration of the utility vehicle is studied according to each section of the route established. – Route A-B: on the travel made on this route, there are critical slopes of up to 24.4 (°) of inclination, with an average consumption of 4,56 (Wh). The total travel distance of the route is 1,22 (km) and the energy consumption is 0,49 (kWh), with a regeneration of up to 0,0019 (kWh). In Fig. 8 can be seen that, if the terrain maintains steep slopes, energy consumption increases and is maintained in certain sections due to negative slopes, but its regeneration does not weigh consumption to a great extent, therefore despite remain, although critical gradient appears, it will continue to increase considerably. – Route B-C: this route has critical slopes of up to 27, 5 (◦ ) incline. This route presents a value of 0.958 (km), with a total energy consumption of 0,33 (kWh) and a regeneration of 0,0044 (kWh). In Fig. 9, a behavior similar to route A-B can be observed, where the critical slope highlights a pronounced value in energy consumption, which tends to decrease with the negative slopes of the terrain. – Route C-D: this route, due to the fact that it has slopes less than 8 (◦ ) of inclination and, in addition, negative, helps the vehicle to perform better. Although regeneration is still low with a value of 0,0155 (kWh), the energy consumption of the vehicle when
268
G. Garrido-Silva et al.
working with lower slopes is also slight, reaching a value of 0,10 (kWh). In Fig. 10, the previous behavior can be observed. – Route D-E: this route has the same behavior due to the negative slopes present in the C-D case, although a maximum slope of 24.4 (°) of inclination stands out. The route of the route is 0.89 (km), which represents a low energy consumption of approximately 0.07 (kWh) and a slightly lower but more optimistic regeneration of 0.0398 (kWh). In Fig. 11, the behavior described above can be analyzed, in which a terrain with many negative slopes can be seen, which estimates a positive energy regeneration for the system in part of the route, as well as a low and stable consumption, despite to reach your maximum work slope. – Route EF: This route has critical slopes less than 8, 5 (◦ ) incline, a route of 0,97 (km), which dates a low consumption of 0,11 (kWh) and a regeneration of 0,0024 (kWh). In Fig. 12, a particular behavior can be reviewed, since, when working on the same elevation, energy consumption increases in a linear way, but its curve stabilizes when negative slopes appear along the way. Then, here is when regeneration plays an important role in this type of system.
Fig. 8. Accumulated energy and behavior of the slopes of the route A-B.
Fig. 9. Accumulated energy and behavior of the slopes of the route B-C.
– Route F-G: The behavior largely represents negative slopes, therefore, regeneration dates a value of 0,12 (kWh) and a consumption of 0,19 (kWh). The maximum slope is 20(◦ ) which represents the previous consumption value. In Fig. 13, a varying behavior is observed on the slopes of the terrain, which influences a quasi-stable behavior
Estimation of the Energy Consumption of an Electric Utility Vehicle
269
of the total energy consumption, since the regeneration reaches to compensate the consumption due to the positive slopes. – Route GH: This is mostly a descending terrain, the ascents do not have a degree of inclination greater than 6.4 (°), as shown in Fig. 14. The travel of the route is 2.25 (km), but due to its behavior already described, consumption is low and corresponds to a value of 0.13 (kWh) and regeneration is 0.01 (kWh).
Fig. 10. Accumulated energy and behavior of the slopes of the route C-D.
Fig. 11. Accumulated energy and behavior of the slopes of the route D-E.
Fig. 12. Accumulated energy and behavior of the slopes of the route E-F.
The road slope generates a change in forces acting on EV affecting the energy consumption. Then, positive values correspond to uphill stages and negative values represent downhill stages. Naturally, while uphill stages the increase of the road slope will increase the energy consumption too, equivalent to sinusoidal component. Therefore, the total electric motor power consumption will increase respectively [14]. The impact of
270
G. Garrido-Silva et al.
Fig. 13. Accumulated energy and behavior of the slopes of the route F-G.
Fig. 14. Accumulated energy and behavior of the slopes of the route G-H.
change in the road slope is often neglected or accounted for through the use of operation modes: cruise, idle time, acceleration, deceleration and braking [3]. According to the characterization of each route with the respective analysis of the behavior of the slopes and regeneration, it can be determined that the total route of the vehicle is 9.83 (km), the consumption per lap is 1.41 (kWh) and the theoretical autonomy of the vehicle corresponds to 125.69 (km). To sum up, it can be deduced that the number of laps that the vehicle manages to make with the general route outlined is 13 times without the need to recharge its batteries fully, which implies a positive management in the useful life of load cells, since it could be estimated if one or more electric charging stations are required on the route.
4 Discussion and Conclusions Implementing new environmentally friendly technologies creates paradigms in the different socio-economic sectors of the country and more so for a sector such as agriculture. Migrating to an electric vehicle technology is a slow process since the cost of an electric kit is around U$20,000 without counting the labor involved in installing it, besides having availability of a sufficient number of charging stations. For our case study, given the difficulties to measure and define the energy consumption of an electrical vehicle for rural sector for the cocoa harvest, we proposed a methodology using the Green Race V5 software, can be established that for a given condition, a 160 [m] route, the energy consumed by the vehicle reaches 52.31 [Wh]; This implies a low consumption due to the short distance traveled and the negative slopes of the study field (which establishes a very low consumption); on the other hand, for slopes up to 31.54 % of incline, the battery consumption reaches up to 1.15 [Wh] for a
Estimation of the Energy Consumption of an Electric Utility Vehicle
271
short distance of up to 10 [m]. Also, it is enough to clarify that if the vehicle achieves a section of up to 100 [km] with similar terrain conditions, the energy consumption of the batteries can reach a value of up to 33.23 [kWh] which implies that the vehicle’s autonomy is 162.5037 [km] at a working speed of 30 [km/h]. Regenerative brakes represent a totally electric system in which no type of brake intervenes mechanical friction, therefore, all energy is reverted to the battery. In the case study, recovery energy per turn is 1.16%, although it represents a low value, can really be said that it is adding to the autonomy of the vehicle since this energy accumulates directly in the batteries. The presented methodology is a key tool which could be used for vehicle designers, that need to develop powertrain and estimate the range of an electric vehicle (i.e. autonomy) and when there is no real data, that is, for special applications. Then, simulation study offers a reliable alternative to achieve this purpose, allowing accurately justify the selection of powertrain components without focusing only on characteristic curves of engine behavior at a given voltage and current; as well as the technical sheets of the batteries and their respective calculations for their selection. Factors such as dead weight, drag coefficient, battery system energy, drive chain performance, energy regenerated, rolling resistance coefficient and accessories with energy demand were considered. The performance and energy consumption could be analyzed, especially for each selected route. As a result, this type of vehicles can improve processes such as to collect a cocoa harvest of 4 batches that are equivalent to approximately 86,475 cocoa fruits, which can be taken the fruits to the center storage 13 times without the need to fully recharge the batteries. This process takes currently manually and collection may take more than a day to complete. Future works should be focused on validation test with real data and instrument the vehicle and include other approaches considering the change of the temperature and road slope for selected route.
References 1. El Amrani, S.: Chennani, M., Belkhayat, D.: Comparative study of electric vehicle energy consumption between trunk roads and highways. In: Proceedings 2019 7th International Renewable and Sustainable Energy Conference (IRSEC), pp. 1–7 (2019) 2. Goł¸ebiewski, W., Lisowski, M.: Theoretical analysis of electric vehicle energy consumption according to different driving cycles. In: IOP Conference Series: Materials Science and Engineering (Vol. 421, No. 2, p. 022010). IOP Publishing (2018) 3. Levin, M.W., Duell, M., Waller, S.T.: Effect of road grade on network wide vehicle energy consumption and ecorouting. Transp. Res. Rec. 2427, 26–33 (2014) 4. Wang, J., Besselink, I., Nijmeijer, H.: Electric vehicle energy consumption modelling and prediction based on road information. World Electr. Veh. J. 7(3), 447–458 (2015) 5. Wang, J., Besselink, I., Nijmeijer, H.: Battery electric vehicle energy consumption prediction for a trip based on route information. Proc. Inst. Mech. Eng. D J. Automob. Eng. 232(11), 1528–1542 (2018) 6. Sweeting, W.J., Hutchinson, A.R., Savage, S.D.: Factors affecting electric vehicle energy consumption. Int. J. Sustain. Eng. 4(3), 192–201 (2011)
272
G. Garrido-Silva et al.
7. Liu, K., Wang, J., Yamamoto, T., Morikawa, T.: Exploring the interactive effects of ambient temperature and vehicle auxiliary loads on electric vehicle energy consumption. Appl. Energy 227(August), 324–331 (2018) 8. Iora, P., Tribioli, L.: Effect of ambient temperature on electric vehicles’ energy consumption and range: Model definition and sensitivity analysis based on Nissan Leaf data. World Electr. Veh. J. 10(1), 1–15 (2019) 9. Huertas, J.I., Quirama, L.F., Giraldo, M., Díaz, J.: Comparison of three methods for constructing real driving cycles. Energies 12(4), 665 (2019) 10. Duell, M., Levin, M., Waller, S.T.: Urban vehicle energy consumption for policy evaluation: Impact of electric vehicles. In: 19th International Conference of Hong Kong Society for Transportation Studies: Transportation and Infrastructure, HKSTS 2014, pp. 271–278 (2014) 11. Bˇrezina, T.: Mechatronics 2017 - preface. Adv. Intell. Syst. Comput. 644, 5 (2018) 12. Ye, F., Wu, G., Boriboonsomsin, K., Barth, M.J.: A hybrid approach to estimating electric vehicle energy consumption for ecodriving applications. In: 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC), pp. 719–724 (2016) 13. Coiret, A.L., Vandanjon, P.O., Deljanin, E., Ortiz, M., Lorino, T.: Management of road speed sectioning to lower vehicle energy consumption. Transp. Res. Procedia 45, 724–731 (2020) 14. Sagaama, I., Kchiche, A., Trojet, W., Kamoun, F.: Impact of road gradient on electric vehicle energy consumption in real-world driving. In: International Conference on Advanced Information Networking and Applications, pp. 393–404 (2020) 15. Rincón-Quintero, A.D., et al.: Manufacture of hybrid pieces using recycled R-PET, polypropylene PP and cocoa pod husks ash CPHA, by pneumatic injection controlled with LabVIEW Software and Arduino Hardware. In: IOP Conference Series: Materials Science and Engineering, Vol. 844, No. 1, p. 012054 (2020) 16. Pettersson, P., Johannesson, P., Jacobson, B., Bruzelius, F., Fast, L., Berglund, S.: A statistical operating cycle description for prediction of road vehicles’ energy consumption. Transp. Res. Part D Transp. Environ. 73, 205–229 (2019) 17. Luin, B., Petelin, S., Al Mansour, F.: Modeling the impact of road network configuration on vehicle energy consumption. Energy 137, 260–271 (2017) 18. Jiménez, D., Hernández, S., Fraile-Ardanuy, J., Serrano, J., Fernández, R., Alvarez, F.: Modelling the effect of driving events on electrical vehicle energy consumption using inertial sensors in smartphones. Energies 11(2), 412 (2018) 19. McNerney, J., Needell, Z.A., Chang, M.T., Miotti, M., Trancik, J.E.: Tripenergy: estimating personal vehicle energy consumption given limited travel survey data. Transp. Res. Rec. 2628(1), 58–66 (2017) 20. Fiori, C., Ahn, K., Rakha, H.A.: Power-based electric vehicle energy consumption model: model development and validation. Appl. Energy 168, 257–268 (2016) 21. Wang, J., Besselink, I., Nijmeijer, H.: Online prediction of battery electric vehicle energy consumption. World Electr. Veh. J. 8(1), 213–224 (2016) 22. Suárez, Y.J., Hernández, F.A.: Manejo de las enfermedades del cacao (2010) 23. Yao, E., Yang, Z., Song, Y., Zuo, T.: Comparison of electric vehicle’s energy consumption factors for different road types. Discret. Dyn. Nat. Soc. 2013 (2013) 24. Fotouhi, A., Shateri, N., Shona Laila, D., Auger, D.J.: Electric vehicle energy consumption estimation for a fleet management system. Int. J. Sustain. Transp. 15(1), 40-54 (2019) 25. Guo, Q., Zhao, Z., Shen, P., Zhan, X., Li, J.: Adaptive optimal control based on driving style recognition for plug-in hybrid electric vehicle. Energy 186, 115824 (2019) 26. Song, L., Liang, T., Lu, L., Ouyang, M.: Lithium-ion battery pack equalization based on charging voltage curves. Int. J. Electr. Power Energy Syst. 115, 105516 (2020) 27. Xu, W., Chen, H., Zhao, H., Ren, B.: Torque optimization control for electric vehicles with four in-wheel motors equipped with regenerative braking system. Mechatronics 57, pp. 95–108 (2019)
Artistic Creations Supplied by Renewable Energy Located in the Most Attractive Mountains of Azuay. Case Study: Cultural Heritage of Quingeo Daniel Icaza1,2,3,4(B) , Santiago Pulla Galindo1,3,4 , Carlos Flores-Vázquez1,2,3,4 , and Fabián Sangurima Paute3,4 1 Grupo de Radiación Visible y Prototipado GIRVyP, Cuenca, Ecuador
{dicazaa,gpullag,cfloresv}@ucacue.edu.ec 2 Grupo de Investigación en Optimización Energética del Sistema, Unidades de Transporte
Urbano, Cuenca, Ecuador 3 Carrera de Ingeniería Eléctrica, Cuenca, Ecuador
[email protected] 4 Universidad Católica de Cuenca Ecuador, Cuenca, Ecuador
Abstract. The purpose of this research is focused on carrying out the design, modeling and simulation of a photovoltaic system and then implement it in the image of the Virgin Mary in the Curiquinga hill of the Quingeo Parish, Cuenca canton, Azuay province. For the feasibility analysis, data taken by the meteorological station located in said sector was processed, which consists of solar radiation and existing temperature during a period of one year. Subsequently, a mathematical model was built to perform simulations with MATLAB for different input values that allow the design of a photovoltaic system that covers the electricity demand. The simulations are carried out in different operating conditions of the photovoltaic system and then compared with the experimental results, reaching the conclusion that we have favorable conditions for the production of electrical energy in Quingeo through this isolated photovoltaic system that drives tourism in the region. Given the fervent Catholicism in the sector, it is possible to contribute to the maintenance of this religious culture through this project. Keywords: Art and culture · Renewable energy · Lighting · Solar panel · Tourism · Modeling · Cultural products
1 Introduction Currently several areas of knowledge are complementing each other, there are no longer unique areas, that is why working in multidisciplinary teams becomes a reality and a latent need when conducting research, especially when it is required to innovate processes and achieve something new. In our case, after a great period of investigation regarding the lighting system, the production of adequate and continuous energy applied to the works of art, a monument to © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 273–287, 2021. https://doi.org/10.1007/978-3-030-72212-8_20
274
D. Icaza et al.
the Virgin Mary was achieved in the Cerro Curiquinga in Azuay and that has awakened an important interest not only for what this study means but for the impact it has produced on Catholic parishioners, which will be the object of another investigation. For these reasons renewable energies are seen as an opportunity to guarantee electric power service, without relying on the public conventional network, a plant of this type either from a renewable energy source or the provision of a hybrid system will become a guarantee of supply, especially a source that does not pollute the environment, avoids having storage of fuels such as diesel or gasoline, which in some cases require quite special spaces due to the high risk of explosion that can be caused, comparatively, wind energy or solar energy do not present imminent risks and are rather friendly to the environment [17–23]. Energy generation processes using renewable sources to supply energy to telecommunications systems are systems that have been addressed in depth and have been very successful in different countries of the world, these analyzes have been addressed by Valerdi [1] and Zhou [2] to supply energy through renewable sources to base stations. More recent studies were carried out by Akella A. K. [3] and Enslin J. H. R. [4] carried out studies regarding the economic impact of applying rural electrification systems in areas where there is no public electricity grid. These investigations reinforce the methodologies for including renewable energies in the tourism field and their benefits will be diverse [12–17]. The results obtained in remote places are surprising, and the interest in visiting these places where, above all, cultural elements are implemented and that are aligned with the sacred have a very important impact. It is possible to take advantage of these forms of generating electricity to mainly give a colorful and pleasing appearance to the eye and, on the other hand, visitors are provided with security since in these isolated places having lighting systems becomes a primary need. Year after year we can notice substantial changes in different parts of the world with the application and utility of renewable energy systems that is clean, all these actions increase penetration rates in the rural sector which boosts the economy [5–7]. Then, Fig. 1 shows the conception of the solar module, which must be located just above the figure of the Virgin, where there are no shadows. The new technologies in photovoltaic solar energy are modernized, for this reason, there has been an improvement in recent times in the capture of the resources provided by the sun, for the generation of electrical energy the use of conventional energies generated is being reduced especially fossil fuels [7–11]. Solar energy, the same that has a range of electromagnetic radiation (light, heat and ultraviolet rays), is what reaches the Earth from the Sun displaced by vacuum, where it has been generated by a nuclear fusion process since transforms part of the mass of the sun into energy. The use of solar energy can be done in two ways: by high temperature thermal conversion (photothermic system) and by photovoltaic conversion (photovoltaic system), in so-called solar or photovoltaic cells.
2 Peak Solar Hour It is the amount of solar energy it receives in specific hours for each square meter with reference to an orientation α and an inclination β, with the number of hours in a day with
Artistic Creations Supplied by Renewable Energy Located
275
Fig. 1. Design of the solar energy supply system for the Virgin Mary monument. (project in DIALUX).
a fictitious irradiance of 1000\,W/m2 , which would have the same total irradiance as the real irradiation of that day, said energy is not the same, this depends on the location, in this case, the closer it is to the equator, the greater it will be, as well as in winter times it will have a variation in solar time peak, unlike in summer time. With the following formula the HSP(α, β) of one day will be obtained; in this case, the irradiance of the day measured in Wh/m2 is divided by 1000 W/m2 : HSP (α,β) =
Gdm (α, β) 1000 W/m2
(1)
Where: HSP(α, β) : number of peak solar hours for a photovoltaic module with an inclined orientation (α, β). Gdm (α, β): average daily monthly value of the global irradiation on the plane of the photovoltaic module with an orientation α and a slope β, expressed in W/m2 . The average value of the HSP is 3,768 for the Cerro Curiquinga of the Quingeo parish, the results were obtained from the data of the meteorological station and verified with the data from the NASA website. In Fig. 2 we can see the HSP values averaged between the months of July 2018 to June 2019 generated in the study sector.
276
D. Icaza et al.
Fig. 2. Direct normal irradiation of the PSM (Physical Solar Model) -Pico Solar Hour-July 2018– June 2019.
2.1 Losses of Solar Radiation Due to Orientation and Inclination When looking for solar capture with photovoltaic modules, it will not be possible, because several factors influence such as physical imperatives, shadows, resistance to wind, etc. Due to its orientation, in this case to the northern hemisphere, towards the South (α = 0o) with an inclination angle βopt . The irradiance factor certifies the losses with respect to the optimal position. According to the IDEA Technical Specifications, it defines it as the percentage of incident radiation for an orientation and inclination generator (α, β) with respect to the corresponding one for an optimal orientation and inclination (α = 0o, βopt ), and The irradiation factor is calculated through the following expression: FI = 1 − POP
(2)
And to calculate the solar radiation losses for a position different from the optimal position, it is done with the following equations: pot = 1.2 ∗ 10−4 ∗ (β − βopt )2 + 3.5 ∗ 10−5 ∗ α2
(3)
For 15o < β < 90o Pot = 1.2 ∗ 10−4 ∗ (β − βopt )2 ; for β ≤ 15o
(4)
Artistic Creations Supplied by Renewable Energy Located
277
α: azimuth angle of the surface, in sexagesimal degrees. β: Angle of inclination of the surface, in sexagesimal degrees. βopt : Optimal tilt angle, in sexagesimal degrees. In Fig. 3 the intensity-voltage characteristic graph or the I-V graph can be estimated and in Fig. 4 the electrical power supplied by a photovoltaic panel and its electrical voltage or P-V curve is included. Where Maximum Power Point (PMPP). The IMPP and VMPP parameters represent the panel current and voltage at the maximum power operating point.
Fig. 3. Current-voltage curve of a photovoltaic module under standard measurement conditions
The incident solar irradiance is what supplies the electrical power of a photovoltaic module. The higher the irradiance, the higher the power, the lower the irradiance, the lower the power. With the following expression, the maximum power of a photovoltaic module can be obtained. PMPP,G = PMPP ∗
G GSTC
(5)
PMPP, G : maximum power of the photovoltaic module, for a solar irradiance G, in W. G: solar irradiance received by the photovoltaic module in W/m2 . GSTC : solar irradiance under standard measurement conditions (1000 W/m2 ). Through the following expressions, the short-circuit current, the open-circuit voltage and the power at the maximum power point can be calculated for a different working temperature under standard measurement conditions: γ ∗ (T − T STC ) PMPP,T = PMPP ∗ 1 + (6) 100
278
D. Icaza et al.
Fig. 4. Power voltage curve of a photovoltaic module under standard measurement conditions.
α ∗ (T − T STC ) ISC,T = ISC ∗ 1 + 100 β VOC,T = VOC ∗ 1 + ∗ (T − T STC ) 100
(7) (8)
PMPPT : maximum power that the photovoltaic module can deliver at a working temperature T. PMPP : maximum power that the photovoltaic module can deliver under standard measurement conditions. ISC,T : short-circuit current at a working temperature T. ISC: short-circuit current under standard measurement conditions. VOC : open circuit voltage under standard measurement conditions. T: working temperature of the photovoltaic module cells. TSTC : working temperature of the module cells under standard measurement conditions (25 °C). α: temperature coefficient of the short-circuit current. β: temperature coefficient of the open circuit voltage. γ: temperature coefficient of maximum power. The Nominal Operating Cell Temperature (TONC) is the temperature reached by the cells of the photovoltaic module as the ambient temperature is 20 °C, with an irradiance of 800 W/m2 and a wind speed of 1 m/s. The value of TONC can be roughly calculated with the following expression: T = TA + G ∗
TONC − 20 800
(8)
Artistic Creations Supplied by Renewable Energy Located
279
The power output of a photovoltaic solar panel arrangement is based on solar irradiance and ambient temperature. The power output in this model is calculated as: Ppv = ηpvg Apvg Gt
(9)
Where ηpv is the photovoltaic generation efficiency, Apv is the area of the photovoltaic generator (m2 ), and Gt is the solar irradiation in the inclined plane of the module (W/m2 ). ηpvg Also, it is defined as: ηpvg = ηr ηpc [1 − β(Tc − Tcref )]
(10)
Where ηpc is power conditioning efficiency that is equal to one when MPPT is used, and β is the temperature coefficient ((0.004–0.006) per °C), and ηr is the efficiency of the reference module and Tcref the temperature of the reference cell in °C. The reference temperature Tcref can be obtained by the relation: NOCT − 20 Gt Tc = Ta + (11) 800 Where Ta is the ambient temperature in °C, NOCT is the nominal temperature of the operating cell in °C, and Gt is the solar irradiation in the inclined module plane (W/m2 ).
3 System Modeling The solar cell, the building block of the solar panel, is basically a P-N junction semiconductor capable of producing electricity due to the photovoltaic effect. Photovoltaic cells are interconnected in a series-parallel configuration to form a photovoltaic array. Using the ideal single diode as shown in Fig. 5, for an array with cells connected in series Ns and cells connected in parallel Np, the current of the array can be related to the voltage of the array as: q(V + IRs ) −1 (12) I = Np Iph − Irs exp AKTNs
Fig. 5. PV array electrical circuit.
280
D. Icaza et al.
The simulation of the solar panel is observed in Fig. 6 where once the solar panel with which we are going to work has been parameterized, the irradiance and temperature data are entered and later from the output m we will have an ammeter and voltmeter to
Fig. 6. Solar Panel Simulation using Matlab/Simulink.
Fig. 7. V-I results at different irradiation values.
Artistic Creations Supplied by Renewable Energy Located
281
view the V-I graph and then using a multiplier to observe the V-P graph. The results of the simulations are displayed in the graphs of Fig. 7 and Fig. 8 and will serve to contrast with the experimental data.
Fig. 8. V-P results at different irradiation values.
Next in Fig. 9 we present the solar irradiation profile in the Curiquinga hill, the same ones that are a reference to enter the system used in simulink.
Fig. 9. Profile of solar irradiation in Curiquinga hill.
282
D. Icaza et al.
In Table 1 we can identify the characteristic data month by month in the period under study where the data were obtained by means of an on-site meteorological station. Table 1. Monthly average of Quingeo station irradiation. MONTHS
Jul-18
Aug-18
Sep-18
Oct-18
Nov-18
Dec-18
Jan-19
Feb-19
Mar-19
Apr-19
May-19
Average
416,99
396,54
401,87
341,62
363,61
424,57
371,17
372,34
341,34
451,56
451,56
Jun-19 529,88
MAX
946,71
914,00
1104,10
890,55
987,31
1012,77
1014,40
1021,82
1017,71
1029,65
1149,9
1068,0
MIN
12,86
12,34
11,15
10,67
13,02
13,06
1,46
14,20
13,00
12,00
13,5
15,1
4 Results and Discussions We proceed to solve Eqs. (1)–(11) taking into account that the total power may not necessarily be simultaneous and for data validation purposes later, this numerical model was used for the simulation in MATLAB. In addition, the intention of this project is to guarantee the necessary energy so that the monument of the Virgin Mary is illuminated at night. Under this objective, the contrast between the simulated curve and the resulting curve adjusted to the experimental data in the field (Fig. 10) was performed when the image shown in Fig. 11 is installed on site.
Fig. 10. Curves corresponding to the mathematical model and experimental curve.
5 Constructive Aspect Initially, for the construction of the figure of the Virgin Mary located in Quingeo, it is constituted by a pedestal of metal rods and pipe as shown in Fig. 12.
Artistic Creations Supplied by Renewable Energy Located
283
Fig. 11. (a) Appreciation of the panel and light on the Virgin Mary monument (b) Side appraisal.
Subsequently, several layers of reinforced cement and plaster were used for the coating. In each covering of the sacred image of the Virgin Mary, the essential details are given, such as the dress, the hand, etc. In this way we seek that the image is resistant to the elements both in the rain and on very sunny days as shown in Fig. 13. It is important in this phase to also consider the passage of the wiring that will go internally with the purpose other than visible and accessible to the eye. The location of the equipment as modeled in dialux and displayed in Fig. 1 will be placed in the lower part, keeping the respective security and that access is only for persons authorized for the respective maintenance. To calculate the performance (Performance Ratio) the following parameters established in Table 2 have been used: It is a photovoltaic solar installation with the following consumptions distributed by months throughout the year. See table 3. These types of projects where communities are involved and have a purpose to promote the sectors have a very strong impact. In Fig. 14 it can be identified that groups of people from the place and from other latitudes can visit these places that are precious and that only require citizen initiatives to give added value. With this, tourism is promoted,
284
D. Icaza et al.
Fig. 12. Structural assembly of the monument to the Virgin Mary.
Fig. 13. General views of the final coating of the masterpiece.
the Catholic faith, use of clean energies very friendly with the environment, art and sacred culture are also present and are integrated with areas such as energy.
Artistic Creations Supplied by Renewable Energy Located
285
Table 2. Specification of the performance of the photovoltaic solar system in artistic creation. Battery loss percentage
5%
Battery self-discharge percentage
0.5%
Battery discharge depth
60%
DC/AC conversion loss percentage 7% Wiring loss percentage
5%
System autonomy
3d
General performance
80.93%
Table 3. Consumption distributed by months throughout the year. Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
%month
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
Consumption (W)
185
185
185
185
185
185
185
185
185
185
185
185
Fig. 14. Visitors to the great top of the Curiquinga hill where the sacred artistic creation of the Virgin Mary is located.
6 Conclusions The production of electricity from renewable energy sources in our environment are having greater application in different areas, it is no less true that in the near future they will be widely used. The artistic and cultural field does not escape integrating with these new technologies, with innovation it is possible to have various artistic creations that can
286
D. Icaza et al.
have a very important impact. Above all, it lends itself to preserve the Catholic religion that is rooted in our Ecuadorian families. On the other hand, it is also possible to promote community tourism, in places that are very attractive to develop activities where crowds of people are immersed. These actions of social benefit are of great interest and are based on integration with renewable energies. In the town of Curiquinga hill where this study was carried out, there is an important benefit, actually the public electricity distribution lines are several kilometers away, furthermore the wiring affects the aesthetics of the place, with these studies where they are Immersed in renewable energies, the supply of energy becomes very simple, in these cases it is much cheaper to install these systems compared to the supply of energy from the public network. Photovoltaic solar energy systems are friendly to the environment, they are not polluting and do not obey too strict regulations, compared to fuels that exist caution and state regulations in the purchase, where the amounts should not be higher due to presumption of contraband. A different proposal has been presented in the application of renewable energies, which in this case is applied to art and culture. In Ecuador this field is not yet exploited, the energy area is not yet integrated with artistic creations and here we give a perspective that can be expanded with new research and that can reach quite interesting levels of application. According to the research presented here, the monument to the Virgin Mary is guaranteed to be supplied with energy permanently, in fact it has been 3 months since its implementation in the Curiquinga hill and the operation of the system has been permanently monitored and no rather, the impact has been great on citizens, something that was not expected and that leads us to focus on new projects of this type where artistic creation and innovation are required.
References 1. Valerdi, D., et al.: Intelligent energy managed service for green base stations. In: GLOBECOM Workshops (GC Wkshps), 2010 IEEE, pp. 1453–1457. IEEE (2010) 2. Zhou, J., et al.: Energy source aware target cell selection and coverage optimization for power saving in cellular networks. In: Proceedings of the 2010 IEEE/ACM International Conference on Green Computing and Communications & International Conference on Cyber, Physical and Social Computing, pp. 1–8. IEEE Computer Society (2010) 3. Akella, A.K., Saini, R.P., Sharma, M.P.: Social, economical and environmental impacts of renewable energy systems. Renew. Energy 34(2), 390–396 (2009) 4. Enslin, J.H.R.: Renewable energy as an economic energy source for remote areas. Renew. Energy 1(2), 243–248 (1991) 5. Infield, D., Freris, L.: Renewable Energy in Power Systems. Wiley (2020) 6. ØSTERGAARD, Poul Alberg, et al. Sustainable development using renewable energy technology. 2020. 7. Jurasz, J., Canales, F.A., Kies, A., Guezgouz, M., Beluco, A.: A review on the complementarity of renewable energy sources: Concept, metrics, application and future research directions. Sol. Energy 195, 703–724 (2020)
Artistic Creations Supplied by Renewable Energy Located
287
8. Leonard, M.D., Michaelides, E.E., Michaelides, D.N.: Energy storage needs for the substitution of fossil fuel power plants with renewables. Renewable Energy 145, 951–962 (2020) 9. SINSEL, Simon R.; RIEMKE, Rhea L.; HOFFMANN, Volker H. Challenges and solution technologies for the integration of variable renewable energy sources—a review. renewable energy, 2020, vol. 145, p. 2271–2285. 10. Alam, MM., Murad, Md.W.: The impacts of economic growth, trade openness and technological progress on renewable energy use in organization for economic co-operation and development countries. Renew. Energy 145, 382–390 (2020) 11. Inês, C., Guilherme, P.L., Esther, M.G., Swantje, G., Stephen, H., Lars, H.: Regulatory challenges and opportunities for collective renewable energy prosumers in the EU. Energy Policy 138, 111212 (2020) 12. Rani, P., Mishra, A. R., Mardani, A., Cavallaro, F., Alrasheedi, M., Alrashidi, A.: A novel approach to extended fuzzy TOPSIS based on new divergence measures for renewable energy sources selection. J. Cleaner Prod. 257, 120352 (2020) 13. Sen, ¸ Z.: Solar energy fundamentals and modeling techniques: atmosphere, environment, climate change and renewable energy. Springer, London (2008) 14. Khan, I.: Impacts of energy decentralization viewed through the lens of the energy cultures framework: Solar home systems in the developing economies. Renew. Sustain. Energy Rev. 119, 109576 (2020) 15. Ero˘glu, H.: Effects of Covid-19 outbreak on environment and renewable energy sector. Environ. Dev. Sustain. 1–9 (2020) 16. Khan, M.J., Mathew, L.: Comparative study of optimization techniques for renewable energy system. Arch. Comput. Methods Eng. 27(2), 351–360 (2020) 17. Herman, K.S.: Attracting foreign direct investment the Chilean government’s role promoting renewable energy. In: 2013 International Conference on Renewable Energy Research and Applications (ICRERA), pp. 37–41. IEEE (2013) 18. Ramirez, E.: Model to make electricity generation projects viable by using renewable energy in rural areas to promote its sustainable development. In: 2015 CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON). IEEE, 2015. p. 501–509. 19. Elizondo, J.L., Rivera, M., Wheeler, P.: Wind energy development and technology in the world: a brief overview. In: 2019 IEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON), pp. 1–5. IEEE (2019) 20. González-Andrés, F., Martínez-Morán, O., Sánchez-Morán, M.E., Gómez-Barrios, X.A., Morán, A., Urbano-López-de-Meneses, B.: Evaluation of an innovative teaching methodology for engineering involving companies and ICTs in a flipped classroom. Revista Infancia, Educación y Aprendizaje 3(2), 536–543 (2017) 21. Escapa, A., Morán, A., Tartakovsky, B., Heidrich, E.S.: Microbial electrochemical technologies for renewable energy production from waste streams. Front. Energy Res. 7, 104 (2019) 22. Lata-García, J., et al.: Sizing optimization of a small hydro/photovoltaic hybrid system for electricity generation in Santay Island, Ecuador by two methods. In: 2017 CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON), pp. 1–6. IEEE (2017) 23. Amutha, W.M., Rajini, V.: Techno-economic evaluation of various hybrid power systems for rural telecom. Renew. Sustain. Energy Rev. 43, 553–561 (2015)
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy F. Endara, C. Pérez, J. Rodriguez, D. Ortiz-Villalba(B) , and J. Llanos Universidad de las Fuerzas Armadas ESPE, Sangolqui, Ecuador [email protected]
Abstract. In recent years unmanned aerial vehicles (UAV) have been used to perform some tasks such as inspection, surveillance, military applications, among others. The performance of an UAV is evaluated based on 3 characteristics: planning, control, and autonomy. The autonomy of an UAV is limited by the amount of energy available in its batteries. Several research works have focused to estimate and increasing the flight time. In this work, the increase in the flight autonomy is evaluated considering the installation of photovoltaic cells in the UAV fuselage in order to increase the flight time using solar energy. The proposed methodology includes mechanical analysis, UAV mathematical model that allows calculating the energy balance for determining the flight time. The results show that the flight time increases about 1 h 46 min considering the installation of a set of solar cells over the UAV fuselage under study. Keywords: Autonomous aerial vehicle · Solar energy · Batteries
1 Introduction Currently, unmanned aerial vehicles (UAVs) have become the focus of attention of several institutions and companies, this market segment is fast-growing with a potentially bright future. UAVs have been applied in several fields such as inspection, surveillance, among others [1]. A fundamental variable in the development of unmanned aerial vehicles is the amount of energy available in the battery, the flight autonomy of the UAV depends on it. Therefore the battery capacity plays a crucial role in the development of the UAV [2]. There are several types of UAV models, some of them use fossil fuels to convert the thermal energy in mechanical energy, other models use batteries. During the last decade, the development of photovoltaic technology has allowed solar cells to be included in the UAVs fuselage in order to receive clean energy from the sun. As mentioned before, technological advances in the field of solar cells manufacturing industry and materials have allowed a significant increase in its efficiency. If the energy obtained by solar irradiance is stored in a battery and this is greater than the energy consumed during the flight, it will result in a long and uninterrupted flight [3]. Solar energy might be used in two manners, passively where heat gains and losses are favored and actively obtaining electrical energy through photovoltaic cells taking © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 288–299, 2021. https://doi.org/10.1007/978-3-030-72212-8_21
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy
289
advantage of the incidence of sunlight. Commercially, there are some types of solar cells such as monocrystalline, polycrystalline, and amorphous silicon. It is worth noting that the maximum energy consumption does not necessarily match with the maximum irradiance level, therefore the use of a battery bank is recommended for storing electrical energy [4]. A collection of research for the development of UAVs using solar energy are summarized below. In [5], the authors used the software Simulink and SimElectronics for modeling and simulation of a photovoltaic system where an equivalent circuit from a 3W solar cell is implemented, for the case study a combination of 11 solar cells is considered. Simulation parameters are compared with a built photovoltaic system using monocrystalline cells. Simulation results corroborate the behavior of a built photovoltaic system. The authors of [6], show the UAV design considering a set of photovoltaic cells for environmental monitoring applications. The system has a wingspan of 4m, with two sets of 48 photovoltaic cells (C60) connected as follows: 24 cells are connected in series and then two sets of 24 are connected in parallel. The photovoltaic cells have an efficiency of about 22,6% the photovoltaic system generates electrical energy about 96.96 Wh. When considering a minimum irradiance, an autonomy of 6 h of interrupted flight has reached the system it was evaluated in the period of 9 AM and 3 PM at a height of 1000 m.a.s.l. In [6] a UAV model the development of solar cells is taken into consideration. Photovoltaic cells were selected and modeled, flight conditions were also taken into account. Obtaining the result of a system that allows studying and evaluating the behavior of the battery and useful charge separately. Plecs and Matlab computational tools were used to run the simulations. In [7], the authors present the analysis performance of three types of solar cells for UAVs, a module with thin cells and a module with standard solar cells. All of them are applied to high-altitude flights considering the same technical specifications. Nevertheless, the weight and thickness of the solar cells are different in the performed evaluation. A detailed analysis of the interconnect tape designed to select the material, thickness and width of the tape-based on considerations of mass is carried out. Moreover, power loss and, reliability are also evaluated. The results show the total power loss is lower than 3,5% considering 1,5 mm wide conductors. The study concludes that thin solar cells produced the lowest losses. The authors in [8] show the design of a solar energy management system for UAVs. The evaluations performed demonstrates that the variations of sunlight incidence angle over the solar cells from 0 to 45°, might drive a reduction up to 30% of the power injected from solar cells. This in return depends on the load conditions and can reduce up to 30%. Therefore, changes in the altitude of the aircraft might affect the power obtained from the solar cells and must be considered for an optimal design as well as the completion flight route. This paper presents a comparative analysis between a conventional UAV and a solar UAV. Solar UAV takes into account the same conventional model but considers the installation of solar cells over the fuselage of conventional UAV. The analysis includes environmental conditions such as air density, wind direction, irradiance level, temperature, and height (meters above sea level). The UAV model takes into account technical
290
F. Endara et al.
parameters such as weight, aspect ratio, wing size, space available for the implementation of photovoltaic cells, cruise speed, as well as the characteristics of the battery. In order to check the percentage increase in autonomy during a flight. The mathematical model is implemented in Matlab software, considering all aforementioned characteristics. Air resistance, the aerodynamic effect in the UAV, and the characteristics of the solar cells are also considered into the model. Two flight simulations were performed: the first considers the conventional UAV model, the second, implementing photovoltaic cells in the infrastructure available in the commercial UAV under study. Once the results were obtained, a performance comparative analysis is carried out. The document is structured as follows: Sect. 2 explains the proposed methodology in detail; in Sect. 3, the characteristics of the UAV under study are described, as well as the environmental conditions considered during the flight time. In Sect. 4, the results are presented and discussed. Finally, in Sect. 5 the conclusions are established.
2 Methodology The proposed methodology is depicted in Fig. 1, each stage is explained in detail in the next paragraphs. 2.1 UAV Mechanical Modeling In order to implement the proposed methodology of this research work, a commercial UAV model is required. This model is modified with the purpose of installing solar cells over the available surface of the conventional UAV. For the mechanical modeling stage, main technical characteristics are considered as follows: aspect ratio to be calculated dividing the wingspan by the middle chord, or in cases where the middle chord is difficult to determine, dividing the wingspan square by the total area [10] (dimensionless), UAV reference area (UAV area), weight, fuselage material among others. Conventional UAV is modeled in Solidworks software [11]. 2.2 Solar UAV Model Once the conventional UAV is modeled, solar UAV is designed (implementing solar cells over its fuselage); for this stage of the proposed methodology technical characteristics of the solar cells are required. For instance, the type of solar cell as well as the available area of the wings, weight of the solar cells, among others. Once the solar cells are selected keeping in mind their characteristics, they are installed specifically, over the effective area available in the wings of the conventional UAV model, the main purpose of this stage is to obtain the feasible area for the calculation and analysis. 2.3 Flight Simulation Technical parameters that were obtained in the previous stages of the proposed methodology are used with the purpose to implement the mathematical model that represents the UAV behavior. The simulations carried out consider conventional UAV and solar UAV. In
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy
291
Fig. 1. Proposed methodology
order to implement the mathematical model, additional parameters are required: cruise speed, drag coefficient, and the Oswald constant. These parameters are described in detail in the next paragraphs. In addition, the simulation carried out considering the Solar UAV model, this also requires other technical aspects. That being said the efficiency of the photovoltaic cells, as well as battery technical features: discharge time in hours, but also the battery capacity is recorded in ampere-hours. The irradiance level is also considered, in order to quantify the amount of energy that the photovoltaic cells provide to the batteries in order to increase the flight time. The irradiance level depends on the area under study. In addition, the air density is considered into the model, this is directly proportional to the UAV flight height. Technical parameters are implemented into the mathematical model reported in [12]. This model was used for the development of this research work. The Eq. 1, represents the power required for the UAV flight (P) where ρ represents the air density ( mkg3 ) considering the altitude of the flight, V is the cruising speed ( ms ) which is considered quasi-stationary, i.e. the forces are neglected of inertia when establishing the balance forces, S is the reference area of the UAV (m2 ), the drag coefficient with zero elevation C D quantifies the resistance of the UAV in mid air (dimensionless), W is the weight of the UAV (kg). P = 0, 5.ρ.V 3 .S.CDo + k.
2.W 2 ρ.V 2 .S
(1)
292
F. Endara et al.
k is presented in Eq. 2, where AR is the aspect ratio, and e is the Oswald efficiency factor, its value depends on the air distribution along the wingspan (dimensionless) [11]. k=
1 π.AR.e
(2)
The power generated by the solar cells is calculated using Eq. 3 where S cell is the effective area of the photovoltaic cells (m2 ), ηcell represents the efficiency of the cells (dimensionless), and Irr the solar irradiance value ( mw2 ) that depends. on the area under study during the UAV flight. Pcell = Scell .ηcell .Irr
(3)
Figure 2 depicted the energy balance, which is calculated using Eq. 4 where Pnet represents the net propulsion power which is the power required for the flight considering the efficiency of the propulsion system and Psolar represents the generated power by the solar cells connected to each other. Equation 5 is applied to determine the battery discharge time, where Rt represents the battery discharge time in hours, whilst C is the battery capacity in ampere-hours, and n is the Peukert exponent. Pbat = Pnet − Psolar
(4)
Fig. 2. Energy balance
Equation 5 allows determining the total discharge time of the UAV battery considering all conditions mentioned in the previous stages of the proposed methodology. It is worth mentioning that for the simulation of the conventional model, Psolar = 0, because solar cells are not considered. t = Rt 1−n
Vbat ∗ Cn Pnet − Psolar
(5)
2.4 Analysis Results The previous section provides the battery discharge time. But also, the power required for each UAV model considered in this research process. As well as the power generated
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy
293
by the set of solar cells, the mathematical model was implemented in Matlab software. The simulation results are compared from the flight time point of view, moreover, in order to show the irradiance effect, the performance of the solar UAV under study is evaluated considering different schedules. The results show that the use of solar cells increases the available power in the battery of UAV which implies an increase in terms of flight time.
3 Case Study This section presents the main UAV characteristics under study to perform the simulations using the mathematical model described in the previous section. The UAV model chosen is AeroVironment RQ-11 Raven [12], which is depicted in Fig. 3. This UAV model is one of the most widely applied in the military field in the United States. In this field, the most important applications are summarized as follows: recognition, surveillance, and object search among others. This UAV model might include a camera installed at the bottom of its structure. The conventional UAV is small and light, therefore, the aircraft is launched by human momentum, once the UAV is launched, it is propelled by an electric motor.
Fig. 3. AeroVironment RQ-11 Raven
Table 1 summarized the main technical characteristics of the conventional UAV, the characteristics are implemented in the mathematical model in order to obtain the battery discharge time. Table 1. Technical Features AeroVironment RQ-11 Raven Parameter
Value
Wingspan
140 cm
Length
90 cm
Wing area
2487 cm2 c/u
Weight
1, 9 kg
Operating altitude 30 m−152 m Speed
km 32 km h −81 h
Battery duration
60−90 min
294
F. Endara et al.
The solar UAV model considers the implementation of monocrystalline photovoltaic cells due to their high efficiency (up to 20% efficiency), including large useful life, estimated over 25 years. In addition, monocrystalline photovoltaic cells are characterized by being made of high purity silicon. Due to its efficiency and long life, this type of photovoltaic cells is more expensive than polycrystalline cells. [15]. Taking into account the wings dimensions of conventional UAV, the solar cells are installed over the UAV wings. The conventional UAV model is converted to solar UAV taking into account the largest effective area available. The Solar UAV model is depicted in the Fig. 4.
Fig. 4. Solar AeroVironment RQ-11 Raven
Three types of monocrystalline photovoltaic cells are considered with the purpose of taking advantage of all UAV available area. Table 2 summarizes the main technical characteristics of the photovoltaic cells and the quantity of solar cells implemented over the UAV structure. [13, 17, 18]. Table 2. Technical Features of Solar Cells Cell 1 Cell 2
Cell 3
Width [mm]
125
156, 75 158.75
Length [mm]
125
156, 75 158.75
Thickness [mm] 0, 2
0, 2
0, 2
Weight [gr]
12
12
12
Efficiency [%]
18, 5
20
22.1
Voltage [V]
0, 525 0, 522
0.571
Current [A]
5, 378 8, 870
9.70
Power [W]
2, 86
4, 89
5.54
2
2
Number of Cells 5
Taking into account the size of the solar cells and the available area in the fuselage of the UAV, 9 monocrystalline cells are installed, this increases the weight of the commercial
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy
295
UAV about 108 g. The main aim is to decrease the battery discharge rate during the flight, it is important to mention that a regulator is also installed to get the solar cells to work in MPPT (maximum power point tracking) and ensure maximum use of the power generated by the cells. The additional weight in the solar UAV is shown in Table 3 Table 3. Technical Features of AeroVironment RQ-11 Raven with solar cells Parameter
Value
Cells Area
1755 cm2
Solar UAV weight 2, 124 kg
The technical characteristics of the UAV battery are shown in Table 4. Table 4. Technical Features of the battery AeroVironment RQ-11 Raven Parameter
Value
Voltage
25 V
Capacity
1, 3 Ah
Discharge time 1 h
In order to evaluate the performance of the solar UAV, the Atacama Desert is the ideal place for this study. The Atacama desert is located in the north of Chile, in the regions of Arica, Parinacota, Tarapacá, Antofagasta, Atacama and the northern of Coquimbo region, in this whole region there is the highest irradiance level on the planet, direct normal irradiance values are high over 3300 km h is reached. The solar projects have significantly increased in the last decade, an extensive database of solar irradiance is available in order to drive new solar projects analysis. The irradiance results are based on global reanalysis data to force a clear-sky radiative transfer model and an empirical model based on satellite data for cloud conditions. The average percentage error of the horizontal global irradiance time series is only 0,73%, considering clear and cloudy days. [19]. The irradiance level was obtained by calculating the average values from 11:00 h to 13:00 h, since in that time interval there is the highest level of irradiance in the area under study, the irradiance level calculated is 955, 16 mw2 .
4 Analysis Results This section presents the results obtained by applying the proposed methodology. A conventional UAV and a solar UAV are evaluated considering the mathematical model presented in Sect. 2, with the purpose to evaluate the flight time of two UAV models under study.
296
F. Endara et al.
The parameters considered into the mathematical model are presented as follows: considering the air density, and the height at which the UAV develops the cruising speed, it is possible to obtain the atmospheric pressure in pascals. Another parameter is the average temperature in kelvin degrees, this value is calculated using the average temperatures in the time interval in which the flight is carried out, also the gas constant for dry air is applied. These parameters are considered in Eq. 1. The UAV model selected to evaluate the proposed methodology is the AeroVironment RQ-11 Raven since compared to the rest of conventional UAVs, these have a light weight and an effective area that allows the assembly of solar cells, the model includes the parameters, cruise rate, drag coefficient, Oswald constant among others. It is worth mentioning that in the simulation it is considered a high irradiance which implies that a considerable power generated is obtained by the solar cells, thus the results show that the flight time increases considerably. Figure 5 depicted UAV power consumption in the y-axis and the flight speed in the x-axis, as can be seen when the UAV flies at a higher speed, a higher amount of energy is consumed, therefore the flight time is reduced. Moreover, if the flight is carried out considering an average speed, UAV power consumption is lower and the flight time will be longer. It is worth mentioning that for this study, a cruising speed of 13 m/s is considered, it is due to at this speed the battery power consumption is minimized; the evaluation of the energy consumption is carried out from the moment the UAV reaches its height and cruising speed since the take-off of this type of UAV is through human impulse, the energy consumption in take-off and landing is neglected. Technical data of the conventional UAV is implemented in the mathematical model, the amount of energy injected by the battery system is 26, 30 W, considering efficiency of the system about 85, 5%, the amount of energy injected to the UAV propulsion system is 22.49 W. The battery discharge is produced after 1.35 h of flight time at cruising speed. In Table 5, the installed capacity of the solar cells is 33, 37 W, this amount of solar cells implies an increase in the flight autonomy to 3,13 h, it represents an increase of 131% over time flight of the conventional UAV, considering the irradiance levels of the Atacama desert. Figure 6 shows the battery behavior for the conventional UAV (blue line in a) presents the total discharge time, this occurs after 1,35 h of use, while the solar UAV discharges its battery at 3,13 h (red line in d) considering the irradiance levels of the typical noon of the Atacama Desert. The purple line in c represents the battery discharge considering the irradiance of the sunset, finally, the black curve in b represents the battery discharge considering the typical irradiance levels that occur in the Atacama desert at the morning. In all scenarios under study, an increase in the time of battery discharge is observed, therefore we can conclude that regardless of the irradiance levels to which the solar UAV is exposed, an increase in autonomy occurs, for instance considering the typical irradiance levels at the morning (204, 35 W/m2 ), there is an increase about 65% flight autonomy. Figure 7 shows the battery voltage during the discharge considering the conventional UAV model (red line) and the solar UAV model (blue line). As can be seen, the solar UAV model has a lower slope, which results in a delay in the battery discharge, increasing the autonomy of the solar UAV. It is worth mentioning that the battery manufacturer
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy
297
Table 5. Solar cells installed capacity in the solar UAV Power Cell 1
14, 3 W
Cell 2
9, 78 W
Cell 3
11, 08 W
Total installed capacity 35, 16 W
38 36
Power[W]
34 32 30 28 26 24 22
1 [m/s] 8
10
12
14
16
18
20
22
24
Speed [m/s]
Fig. 5. Speed and UAV Power consumption
recommends that the minimum voltage at which battery must be discharged is 21V in order to preserve the battery lifespan.
5 Conclusions This paper presents a practical approach in order to evaluate the performance of a conventional UAV against to solar UAV. The solar UAV model is obtained using the conventional UAV model, considering the installation of solar cells over the available area on the UAV fuselage. The main aim of solar cells is to increase the flight autonomy, the energy injected from the solar cells delays the battery discharge. The set of solar cells allows to increase the flight time by a total of 3,13 h, which represents an increase of 131% in its autonomy compared to the conventional model. The UAV under study has 33,37 W of solar cells installed capacity, the modified model is able to fly at cruising speed of 13 m/s during the mission with the purpose to maximize its autonomy. The results obtained show that the installation of solar cells over the UAV fuselage increases the flight autonomy, due to a reduction in the battery discharge, the amount of energy injected from the solar cells depends on the irradiance.
298
F. Endara et al. a
200
b
200
Discharge morning
Power [W]
Power [W]
Discharge UAV
150 100
150 100 50
50
0
0 0
0.5
1
0
1.5
1
1.5
2.5
d
200
Discharge noon
Discharge sunset
150
Power [W]
Power [W]
2
Time [h]
Time [h] c
200
0.5
100
150 100 50
50 0
0 0
0.5
1
1.5
2
2.5
0
0.5
Time [h]
1
1.5
2
2.5
3
Time [h]
Fig. 6. a) Conventional UAV. b) Solar UAV at the morning. c) Solar UAV at the sunset. d) Solar UAV at the noon. 25 Solar UAV UAV
24.5
Voltage [V]
24 23.5 23 22.5 22 21.5 21 0
0.5
1
1.5
2
2.5
3
3.5
Time [h]
Fig. 7. Voltage droop during battery discharge
The proposed methodology might be implemented in any type of conventional UAV with the purpose to evaluate the performance of the conventional UAV converted into a solar UAV.
Analysis of Unmanned Aerial Vehicle (UAV) Based on Solar Energy
299
References 1. Guerrero-Castellanos, J.F., Marchand, N., Hably, A., Lesecq, S., Dela-mare, J.: Bounded attitude control of rigid bodies: Real-time experimentation to a quadrotor mini-helicopter. Control Eng. Pract. 19(8), 790–797 (2011) 2. Boucher, R.J.: Sunrise, the world’s first solar powered airplane. J. Aircraft 22(10), 840–846 (1985) 3. Noth, A.: Design of solar powered airplanes for continous flight. PhD thesis, ETH Zurich (2008) 4. Domizio, M.: Aplicaciones de la energıa solar fotovoltaica: una opcion a nuestro alcance, Revista de la Universidad de Mendoza (1999) 5. Garcia, M., Grano, C., Guerrero, J.F., Ambrosio, R.C., Moreno, M., Guerrero, W.F., Mino, G., Gonzalez, V.R.: Modeling and simulation of a photovoltaic array for a fixed-wing unmanned aerial vehicle. In: 2016 IEEE 43rd Photovoltaic Specialists Conference (PVSC), pp. 2682– 2687. IEEE (2016) 6. Thipyopas, C., Sripawadkul, V., Warin, N.: Design and development of a small solar powered UAV for environmental monitoring application. In: 2019 IEEE Eurasia Conference on IOT, Communication and Engineering (ECICE), pp. 316–319. IEEE (2019) 7. Martinez, V., Defay, F., Salvetat, L., Neuhaus, K., Bressan, M., Alonso, C., Boitier, V.: Study of photovoltaic cells implantation in a long-endurance airplane drone. In: 2018 7th International Conference on Renewable Energy Research and Applications (ICRERA), pp. 1299–1303. IEEE (2018) 8. Nowlan, M.J., Maglitta, J.C., Darkazalli, G., Lamp, T.: Ultralight photovoltaic modules for unmanned aerial vehicles. In: Conference Record of the Twenty Sixth IEEE Photovoltaic Specialists Conference-1997, pp. 1149–1152. IEEE (1997) 9. Shiau, J., Ma, D., Yang, P., Wang, G., Gong, J.: Design of a solar power man- agement system for an experimental uav. IEEE Trans. Aerosp. Electron. Syst. 45(4), 1350–1360 (2009) 10. Simons, M.: Model aircraft aerodynamics. Nexus Special Interests Hemel Hemp- stead, UK (1999) 11. SOLIDWORKS. https://www.solidworks.com, August2006. (undefined 1/7/2020 10:59) 12. de Carvalho Bertoli, G., Pacheco, G., Adabo, G.: Extending flight endurance of electric unmanned aerial vehicles through photovoltaic system integration. In: 2015 International Conference on Renewable Energy Research and Applications (ICRERA), pp. 143–147. IEEE (2015) 13. Torabi, H., Sadi, M., Varjani, A.: Solar power system for experimental un- manned aerial vehicle (UAV); design and fabrication. In: 2011 2nd Power Electronics, Drive Systems and Technologies Conference, pp. 129–134. IEEE (2011) 14. Raven, R.: Qc rq-11a/b (2016). https://www.avinc.com 15. Zhang, S., Pan, X., Jiao, H., Deng, W., Xu, J., Chen, Y., Altermatt, P., Feng, Z., Verlinden, P.: 335-w world-record p-type monocrystalline module with 20.6% efficient perc solar cells. IEEE J. Photovol. 6(1), 145–152 (2015) 16. Allesun new energy (Vietnam). https://cdn.enfsolar.com/Product/pdf/Cell 17. 17.“All sun new energy (Vietnam). https://cdn.enfsolar.com/Product/pdf/Cell 18. Monocrystalline solar cell. https://cdn.enfsolar.com/Product/pdf/Cell 19. Molina, A., Falvey, M., Rondanelli, R.: A solar radiation database for Chile. Sci. Rep. 7(1), 1–11 (2017)
Blackberry (Rubus Glaucus) Natural-Fiber Reinforced Polymeric Composites: An Overview of Mechanical Characteristics Enrique Mauricio Barreno-Avila(B) , Morayma De Los Ángeles Balladares-Pazmiño, Alex Francisco Barreno-Avila, and Segundo Manuel Espín-Lagos Faculty of Civil and Mechanical Engineering/Technology and Transfer Center (CTT), Universidad Técnica de Ambato, Av. los chásquis, 180207 Ambato, Ecuador [email protected]
Abstract. The commonly known Andean blackberry or Mora de Castilla (Rubus glaucus) can be grown in the highlands of Latin America, from Mexico to Bolivia. However, the major productions are in rural areas of Colombia and Ecuador contributing a source of income to their population. For instance, the fruit crop is of economic importance for farmers in the Province of Tungurahua (Ecuador). This study is based on the composite material’s experiment of a polymeric matrix reinforced with blackberry green fiber. The proposed natural fiber is chosen for its vast planted area residues of 3,673 hectares that is being pruned and burned monthly, producing large amounts of CO2, hence polluting the environment. Unfortunately, there is no post usage application for production wastage. Why not use this fiber as a cars’ reinforcing material? Specimens were manufactured under ASTM D3039, ASTM D7268, ASTM D5628-10 standards for traction, bending and impact, respectively. It was determined from the statistical analysis that group 6 composed of: 70% polymeric matrix and 30% blackberry stem, presented better mechanical characteristics. This specimens’ group achieved a tensile strength = 18.62 MPa, flexural strength = 46.9 MPa and absorbed energy = 1.2156 J. Thus, not only the hypothesis in this investigation is accepted, but also the composite material meets the minimum mechanical properties required in the internal parts of the buses’ bodies. Keywords: Rubus Glaucus · Buses’ body parts · Natural fiber · Minimum mechanical characteristics
1 Introduction In the last 20 years, European car manufacturers and suppliers have been encouraged to use green fiber composites with both thermoplastic and thermoset matrices for applications such as package trays, backseats [1], dashboards door panels [2], headliners, and interior parts [3, 4]. Thus, green fibers are gradually replacing synthetic fibers in various automobiles parts due to their low-cost, lightweight, and environmental convenience [5]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 M. Botto Tobar et al. (Eds.): CIT 2020, LNEE 763, pp. 300–315, 2021. https://doi.org/10.1007/978-3-030-72212-8_22
Blackberry (Rubus Glaucus) Natural-Fiber Reinforced Polymeric Composites
301
The commonly used plant fibers are: sisal [6], kenaf [7], hemp [8], coir [9], bamboo [10], groundnut shell [11], pineapple [12], jute [13], etc. Nevertheless, there are insufficient studies regarding to the blackberry fiber’s mechanical characteristics for bus walls parts. The bus bodywork industry in Ecuador began in the 1950s together with the buses seats’ upholstery, the batteries and tires manufacturing.[14] Subsequently, AYMESA was founded in 1973, launching on the market the first local manufactured car in the country, “The Andino car” [15]. The first car production was 144 units/year. Excessive weight on the buses bodywork has been a drawback to be solved, since it limits acceleration time and increases fuel consumption, affecting the environment with greenhouse gas emissions. Car manufacturers considered the use of high-strength steels or aluminum in construction, but carbon fiber composite materials have been used in the industry for other items. This is a synthetic hydrocarbon origin fiber obtained from the heat treatment of Polyacrylonitrile base material which acrylic is made from. Carbon fiber alone is useless, it needs two materials such as hardener resins or catalysts to form composite materials [16]. The use of vegetable fiber in bodywork is very low. In Brazil, the Technological Institute of Aeronautics (ITA) has developed the first jute fiber car [17]. As there is no study of composite materials with fiber from the stem of the blackberry plant (Rubus Glaucus), this study and tests will be carried out to characterize its mechanical properties to comply with the minimum requirement to be used in internal car body parts [18].
2 Methodology and Materials 2.1 Method The descriptive-experimental methodology was applied to describe the physical and mechanical properties of the composite material, by executing the proposed destructive tests and observing the results using technical sheets and demonstrating, its applicability in bus bodywork interior parts of the Tungurahua industry. The methodology is illustrated in the flow chart in Fig. 1. 2.2 Materials The pruned blackberry plant stems were collected from lands located in the northern part of Ambato where they were discarded. This can be seen in Fig. 2. Fiber Extraction. The fiber extraction was carried out manually by extracting the bark from the stems with the help of a utility knife. This material is considered agricultural waste, as such, it needs drying. For this case, the fibre was placed in an oven set at 130 °C for three hours to obtain a dehydrated fiber. As can be seen in Fig. 3. 2.3 Population and Sample Population. The population had specimens of polymeric matrix reinforced with fiber from the blackberry plant stem (Rubus Glaucus), which are evaluated using configurations with different percentages of matrix and reinforcement detailed in Table 1, shown below.
302
E. M. Barreno-Avila et al.
Fig. 1. Methodology flowchart
Fig. 2. Blackberry plant stems
Sample. For this research, through the analysis with destructive tests, 152 specimens were prepared; which 120 were used. For tensile (ASTM D3039 standard), bending
Blackberry (Rubus Glaucus) Natural-Fiber Reinforced Polymeric Composites
303
Fig. 3. Blackberry stem fiber extraction Table 1. Volume fraction, fiber type, fibers’ orientation and manufacturing method. Volumetric fraction
Fiber type
Fiber orientation
Manufacturing method
Type (A): 80% polymer resin, 20% fiber from the stem of the blackberry plant
Long fiber (>10 cm)
Longitudinal to axis
Manual stratification Compression molding
Short fiber (10 cm)
Longitudinal to axis
Manual stratification Compression molding
Short fiber (10 cm)
Longitudinal to axis
- Manual stratification
7
7
5
- Compression molding
7
7
5
Short fiber (10 cm)
Longitudinal to axis
- Manual stratification
7
7
5
- Compression molding
7
7
5
Short fiber (