Engineering Applications of Modern Metaheuristics PDF
Preview Engineering Applications of Modern Metaheuristics
Studies in Computational Intelligence 1069 Taymaz Akan Ahmed M. Anter A. Şima Etaner-Uyar Diego Oliva Editors Engineering Applications of Modern Metaheuristics Studies in Computational Intelligence Volume 1069 Series Editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland The series “Studies in Computational Intelligence” (SCI) publishes new develop- ments and advances in the various areas of computational intelligence—quickly and with a high quality. The intent is to cover the theory, applications, and design methods of computational intelligence, as embedded in the fields of engineering, computer science, physics and life sciences, as well as the methodologies behind them. The series contains monographs, lecture notes and edited volumes in computational intelligence spanning the areas of neural networks, connectionist systems, genetic algorithms, evolutionary computation, artificial intelligence, cellular automata, self- organizing systems, soft computing, fuzzy systems, and hybrid intelligent systems. Of particular value to both the contributors and the readership are the short publica- tion timeframe and the world-wide distribution, which enable both wide and rapid dissemination of research output. This series also publishes Open Access books. A recent example is the book Swan, Nivel, Kant, Hedges, Atkinson, Steunebrink: The Road to General Intelligence https://link.springer.com/book/10.1007/978-3-031-08020-3 Indexed by SCOPUS, DBLP, WTI Frankfurt eG, zbMATH, SCImago. All books published in the series are submitted for consideration in Web of Science. · · Taymaz Akan Ahmed M. Anter · A. S¸ima Etaner-Uyar Diego Oliva Editors Engineering Applications of Modern Metaheuristics Editors Taymaz Akan Ahmed M. Anter Department of Medicine Egypt-Japan University of Science Louisiana State University and Technology (E-JUST) Health Sciences Center Alexandria, Egypt Shreveport, USA Faculty of Computers and Artificial Department of Software Engineering Intelligence ˙Istanbul Topkapı University Beni-Suef University Istanbul, Turkey Beni Suef, Egypt A. S¸ima Etaner-Uyar Diego Oliva Department of Software Engineering Depto. de Innovación Basada en la Fatih Sultan Mehmet Vakıf University Información y el Conocimiento Istanbul, Turkey Universidad de Guadalajara, CUCEI Guadalajara, Jalisco, Mexico ISSN 1860-949X ISSN 1860-9503 (electronic) Studies in Computational Intelligence ISBN 978-3-031-16831-4 ISBN 978-3-031-16832-1 (eBook) https://doi.org/10.1007/978-3-031-16832-1 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 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 Empirical Comparison of Heuristic Optimisation Methods for Automated Car Setup ........................................... 1 Berna Kiraz, Shahriar Asta, Ender Özcan, Muhammet Köle, and A. S¸ima Etaner-Uyar Metaheuristic Algorithms in IoT: Optimized Edge Node Localization ....................................................... 19 Farzad Kiani and Amir Seyyedabbasi JAYA Algorithm Versus Differential Evolution: A Comparative Case Study on Optic Disc Localization in Eye Fundus Images .......... 41 J. Prakash and B. Vinoth Kumar Minimum Transmission Power Control for the Internet of Things with Swarm Intelligence Algorithms ................................ 51 Ahmet Cevahir Cinar An Enhanced Gradient Based Optimized Controller for Load Frequency Control of a Two Area Automatic Generation Control System ........................................................... 79 Nabil Anan Orka, Sheikh Samit Muhaimin, Md. Nazmush Shakib Shahi, and Ashik Ahmed A Meta-Heuristic Algorithm Based on the Happiness Model ........... 109 Aref Yelghi and Shirmohammad Tavangari Application of Metaheuristic Techniques for Enhancing the Financial Profitability of Wind Power Generation Systems ......... 127 Prasun Bhattacharjee, Rabin K. Jana, and Somenath Bhattacharya Optimization of Demand Response .................................. 149 Altaf Q. H. Badar, Rajeev Arya, and Diego Oliva v vi Contents Fitting Curves of Ruminal Degradation Using a Metaheuristic Approach ......................................................... 167 Muhammed Milani Optimizing a Real Case Assembly Line Balancing Problem Using Various Techniques ................................................ 179 Nima Mirzaei and Mazyar Ghadiri Nejad Multi-circle Detection Using Multimodal Optimization ................ 193 Aydin Cetin, Somaiyeh Rezai, and Taymaz Akan Empirical Comparison of Heuristic Optimisation Methods for Automated Car Setup Berna Kiraz, Shahriar Asta, Ender Özcan, Muhammet Köle, and A. S¸ima Etaner-Uyar Abstract Tuning a race car to improve its performance by adopting an effective setup is crucial and an extremely challenging task. The Open Racing Car Simulator, referred to as TORCS, is a well-known simulator in which a race car requires a con- figuration of twenty two real-valued parameters for an optimal setup. In this study, various modern (meta)heuristic techniques, such as, evolutionary algorithms, swarm intelligence algorithm and selection hyper-heuristics, are evaluated using TORCS to solve the car setup optimisation problem across a range of tracks. An in-depth performance comparison and analysis of those techniques on the car setup optimisa- tion problem are provided with a discussion on their strengths and weaknesses. The empirical results indicate the success of Covariance Matrix Adaptation Evolutionary Strategy for the car setup optimisation problem. · · Keywords Evolutionary computation Heuristic algorithms Particle swarm · optimization Simulation A. S¸. Etaner-Uyar worked at the Computer Engineering Department of Istanbul Technical University when she pursued this study. B B. Kiraz ( ) · A. S¸. Etaner-Uyar Department of Computer Engineering, Fatih Sultan Mehmet Vakif University, Istanbul, Turkey e-mail: [email protected] A. S¸. Etaner-Uyar e-mail: [email protected] S. Asta · E. Özcan School of Computer Science, University of Nottingham, Nottingham NG8 1BB, UK e-mail: [email protected] M. Köle Department of Computer Engineering, Istanbul Technical University, Istanbul, Turkey © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1 T. Akan et al. (eds.), Engineering Applications of Modern Metaheuristics, Studies in Computational Intelligence 1069, https://doi.org/10.1007/978-3-031-16832-1_1 2 B.Kirazetal. 1 Introduction The Open Racing Car Simulator (TORCS) [1] based car setup optimisation prob- lem deals with finding the best (preferably optimal) configuration for the parameters that determines performance of a race car. A high quality configuration increases the performance of the car throughout the race. The car itself together with the racing environment is simulated in TORCS. The simulator, initially a car racing game, employs a real-world and powerful physical engine which produces a racing environment with realistic physical effects such as traction, aerodynamics and fuel consumption [2]. Various racing tracks with a variety of distinguishing features are embedded in TORCS. For instance, tracks can differ in length, shape, width or sur- face properties (rough, dirty and etc.). Consequently, various tracks expose racing cars to different levels of difficulty. A good parameter configuration for the racing car is hence a successful setting on a wide range of different tracks. The simulator also provides a non-graphical mode and several vital interfaces. For instance, while running an experiment involving evolutionary algorithms, one can use the communi- cation interface to pass the candidate solution to the simulator for objective function evaluation. The same interface can also be used to collect the feedback from the simulator. This type of interaction between the objective function and optimisation algorithms is vital and is often used when objective function and optimisation algo- rithm are separated (for another example see [3]). In TORCS, the fuel usage can be incorporated into the evaluations through the use of a flag. Due to fuel consumption of the racing car over fitness evaluations, TORCS can be considered as a dynamic optimisation problem. Metaheuristics [4, 5] provide guidelines for heuristic optimisation. Existing approaches for dynamic optimisation problems are mostly a single population based metaheuristics, in particular evolutionary algorithms [6]. The multi-population based evolutionary algorithms to solve dynamic optimisation problems have also been growing in the last years. Some recent studies that comprehensively analyse the chal- lenging issues for multi-population approaches for dynamic optimisation problems are presented in [7–9]. Besides population-based methods, there are some single- solution based approaches for dynamic environments [10, 11] as well. The car setup optimisation problem can be viewed as a global optimisation prob- lem to explore the best parameter setting for a race car. In this problem, all decision variables can be real-valued or transformed to real-valued parameters from the dis- crete space. Moreover, the problem addressed in this paper is dynamic, where prob- lem features are time varying. This non-stationary characteristic poses a challenge as it renders optimal or sub-optimal values discovered for specified time period of little use in future. In this paper, it is aimed to compare the performance of different modern (meta)heuristic optimisation techniques including evolutionary algorithms, swarm intelligence algorithm and selection hyper-heuristics which represent meta- heuristics that search the heuristic space formed by a predefined set of heuristics [12, 13] for optimising the car setup in TORCS environments.
The list of books you might like
