Improved genetic algorithm based on reinforcement learning for electric vehicle front-end structure optimization design

doi:10.1007/s40436-024-00495-z

Advances in Manufacturing ›› 2024, Vol. 12 ›› Issue (3): 556-575.doi: 10.1007/s40436-024-00495-z

• • 上一篇

Improved genetic algorithm based on reinforcement learning for electric vehicle front-end structure optimization design

Feng-Yao Lyu¹, Zhen-Fei Zhan¹, Gui-Lin Zhou¹, Ju Wang², Jie Li², Xin He²

1. Chongqing Jiaotong University, Chongqing, 400074, People's Republic of China;
2. Changan Automobile Co. Ltd., Chongqing, 404100, People's Republic of China

收稿日期:2023-10-08 修回日期:2023-11-26 发布日期:2024-09-07
通讯作者: Zhen-Fei Zhan,E-mail:zhenfeizhan@cqjtu.edu.cn E-mail:zhenfeizhan@cqjtu.edu.cn
作者简介:Feng-Yao Lyu is currently pursuing a master’s degree at the school of Mechatronics and Vehicle Engineering, Chongqing Jiaotong University Chongqing, China;
Zhen-Fei Zhan received the Ph.D. degree in mechanical engineering from the Shanghai Jiao Tong University, Shanghai, China. He is the assistant Dean and a professor at the school of Mechatronics and Vehicle Engineering, Chongqing Jiaotong University Chongqing, China. His current research interests include AI-based engineering applications, big data analytics, automotive safety, etc. He was a data scientist and research engineer at Ford North America;
Gui-Lin Zhou is currently pursuing a Master’s degree at the school of Mechatronics and Vehicle Engineering, Chongqing Jiaotong University Chongqing, China;
Ju Wang received the Master’s degree from Chongqing University. She’s now chief of security at Changan Automobile. Her research interests include automobile crash safety, automobile structure optimization;
Jie Li received the Ph.D. degree from the Ruhr-University of Bochum, Germany. He’s a researcher at the State Key Laboratory of Vibration and Noise. His research interests include AI applications in automobiles;
Xin He received the Ph.D. degree in mechanical engineering from the Technical University of Berlin, Germany. He’s now a researcher at Changan Automobile. His research interests include automotive intelligent safety.
基金资助:
This paper is supported by the Open Fund of National Key Laboratory of Intelligent Vehicle Safety Technology (Grant No. IVSTSKL-202305), and Chongqing Jiaotong University-Yangtse Delta Advanced Material Research Institute Provincial-level Joint Graduate Student Cultivation Base (Grant No. JDLHPYJD2021008).

Improved genetic algorithm based on reinforcement learning for electric vehicle front-end structure optimization design

Feng-Yao Lyu¹, Zhen-Fei Zhan¹, Gui-Lin Zhou¹, Ju Wang², Jie Li², Xin He²

1. Chongqing Jiaotong University, Chongqing, 400074, People's Republic of China;
2. Changan Automobile Co. Ltd., Chongqing, 404100, People's Republic of China

Received:2023-10-08 Revised:2023-11-26 Published:2024-09-07
Contact: Zhen-Fei Zhan,E-mail:zhenfeizhan@cqjtu.edu.cn E-mail:zhenfeizhan@cqjtu.edu.cn
Supported by:
This paper is supported by the Open Fund of National Key Laboratory of Intelligent Vehicle Safety Technology (Grant No. IVSTSKL-202305), and Chongqing Jiaotong University-Yangtse Delta Advanced Material Research Institute Provincial-level Joint Graduate Student Cultivation Base (Grant No. JDLHPYJD2021008).

摘要/Abstract

摘要： The structural optimization of electric vehicles involves numerous design variables and constraints, making it a complex engineering optimization task over the past decades. Many population-based evolutionary algorithms encounter issues such as converging to local optima and lacking population diversity when tackling such optimization problems. Consequently, the solutions obtained for the optimization may be flawed or suboptimal. To address these problems, an improved genetic algorithm (GA) based on reinforcement learning is proposed in this paper. The proposed method introduces a population delimitation method based on individual fitness ranking. The population is divided into two parts: the excellent population and the ordinary population, and different selection and cross-mutation methods are applied to each part separately. More efficient crossover and mutation methods are then applied to the ordinary population to enhance the generation of excellent individuals. Furthermore, the proposed approach replaces the traditional fixed crossover and mutation rates with a dynamic selection method based on reinforcement learning to enhance optimization efficiency. A markov decision process model is constructed based on GA environment in this context. The population state determination method and reward method are designed for reinforcement learning in the GA environment, dynamically selecting the most appropriate genetic parameters based on the current state of the population. Finally, the uncertainty in the manufacturing process is introduced into the optimization problem and the case study results demonstrate that the proposed reinforcement learning-based GA significantly outperforms other evolutionary algorithms when applied to solving the structural optimization of electric vehicles.

The full text can be downloaded at https://link.springer.com/article/10.1007/s40436-024-00495-z

关键词: Structure optimization, Genetic algorithm (GA), Q-learning

Abstract: The structural optimization of electric vehicles involves numerous design variables and constraints, making it a complex engineering optimization task over the past decades. Many population-based evolutionary algorithms encounter issues such as converging to local optima and lacking population diversity when tackling such optimization problems. Consequently, the solutions obtained for the optimization may be flawed or suboptimal. To address these problems, an improved genetic algorithm (GA) based on reinforcement learning is proposed in this paper. The proposed method introduces a population delimitation method based on individual fitness ranking. The population is divided into two parts: the excellent population and the ordinary population, and different selection and cross-mutation methods are applied to each part separately. More efficient crossover and mutation methods are then applied to the ordinary population to enhance the generation of excellent individuals. Furthermore, the proposed approach replaces the traditional fixed crossover and mutation rates with a dynamic selection method based on reinforcement learning to enhance optimization efficiency. A markov decision process model is constructed based on GA environment in this context. The population state determination method and reward method are designed for reinforcement learning in the GA environment, dynamically selecting the most appropriate genetic parameters based on the current state of the population. Finally, the uncertainty in the manufacturing process is introduced into the optimization problem and the case study results demonstrate that the proposed reinforcement learning-based GA significantly outperforms other evolutionary algorithms when applied to solving the structural optimization of electric vehicles.

The full text can be downloaded at https://link.springer.com/article/10.1007/s40436-024-00495-z

Key words: Structure optimization, Genetic algorithm (GA), Q-learning

Feng-Yao Lyu, Zhen-Fei Zhan, Gui-Lin Zhou, Ju Wang, Jie Li, Xin He. Improved genetic algorithm based on reinforcement learning for electric vehicle front-end structure optimization design[J]. Advances in Manufacturing, 2024, 12(3): 556-575.

参考文献

1. Chen X, Wang J, Zhao K et al (2022) Electric vehicles body frame structure design method:an approach to design electric vehicle body structure based on battery arrangement. Proc Inst Mech Eng Part D J Automob Eng 236(9):2025-2042
2. Mallick PK (2020) Materials, design and manufacturing for lightweight vehicles. Woodhead Publishing, Cambridge
3. Qiu D, Wang Y, Hua W et al (2023) Reinforcement learning for electric vehicle applications in power systems:a critical review. Renew Sust Energy Rev 173:113052. https://doi.org/10.1016/j.rser.2022.113052
4. Ahmed M, Zheng Y, Amine A et al (2021) The role of artificial intelligence in the mass adoption of electric vehicles. Joule 5(9):2296-2322
5. Altun F, Tekin SA, Gürel S et al (2019) Design and optimization of electric cars:a review of technological advances. In:The 8th international conference on renewable energy research and applications (ICRERA). IEEE, Brasov Romania
6. Li Q, Wu L, Chen T et al (2021) Multi-objective optimization design of B-pillar and rocker sub-systems of battery electric vehicle. Struct Multidisc Optim 64:3999-4023
7. Jeong MH, Park GJ (2023) Nonlinear dynamic structural optimization of electric vehicles considering multiple safety tests. Int J Auto Tech 24(2):573-583
8. Wang S, Wang D (2021) Crashworthiness-based multi-objective integrated optimization of electric vehicle chassis frame. Arch Civ Mech Eng 21(3):103. https://doi.org/10.1007/s43452-021-00242-2
9. De Gaetano G, Mundo D, Maletta C et al (2015) Multi-objective optimization of a vehicle body by combining gradient-based methods and vehicle concept modeling. Case Stud Mech Syst Signal Process 1:1-7
10. Stabile P, Ballo F, Gobbi M et al (2021) Multi-objective structural optimization of vehicle wheels. In:International design engineering technical conferences and computers and information in engineering conference. ASME, Missouri
11. De S, Singh K, Seo J et al (2019) Structural design and optimization of commercial vehicle hassis under multiple load cases and constraints. In:IAA Scitech Forum. San Diego
12. Bertolini A, Martins MS, Vieira SM et al (2022) Power output optimization of electric vehicles smart charging hubs using deep reinforcement learning. Expert Syst Appl 201:116995. https://doi.org/10.1016/j.eswa.2022.116995
13. Ryberg AB, Bäckryd RD, Nilsson L (2015) A metamodel-based multidisciplinary design optimization process for automotive structures. Eng Comput 31:711-728
14. Qin H, Guo Y, Liu Z et al (2018) Shape optimization of automotive body frame using an improved genetic algorithm optimizer. Adv Eng Softw 121:235-249
15. Shi K, Ruan Z, Jiang Z et al (2018) Improved plant growth simulation and genetic hybrid algorithm (PGSA-GA) and its structural optimization. Eng Comput 35(1):268-286
16. Chen R, Yang B, Li S et al (2020) A self-learning genetic algorithm based on reinforcement learning for flexible job-shop scheduling problem. Comput Ind Eng 149:106778. https://doi.org/10.1016/j.cie.2020.106778
17. Kimbrough SO, Koehler GJ, Lu M et al (2008) On a feasible-infeasible two-population (fi-2pop) genetic algorithm for constrained optimization:distance tracing and no free lunch. Eur J Oper Res 190(2):310-327
18. Köksal AE, Li Z, Veeravalli B et al (2022) Reinforcement learning-enabled genetic algorithm for school bus scheduling. J Intell Transp S 26(3):269-283
19. Karafotias G, Smit SK, Eiben AE (2012) A generic approach to parameter control. In:European conference on the applications of evolutionary computation. Springer, Berlin Heidelberg
20. Karafotias G, Hoogendoorn M, EibenÁE (2014) Parameter control in evolutionary algorithms:trends and challenges. IEEE T Evolut Comput 19(2):167-187
21. Haasdijk E, Eiben AE, Karafotias G (2010) On-line evolution of robot controllers by an encapsulated evolution strategy. In:IEEE congress on evolutionary computation. IEEE, Barcelona
22. Katoch S, Chauhan SS, Kumar V (2021) A review on genetic algorithm:past, present, and future. Multimed Tools Appl 80:8091-8126
23. Lambora A, Gupta K, Chopra K (2019) Genetic algorithm-a literature review. In:International conference on machine learning, big data, cloud and parallel computing (COMITCon), Faridabad, India
24. Haldurai L, Madhubala T, Rajalakshmi R (2016) A study on genetic algorithm and its applications. Int J Comput Sci Eng 4(10):139-143
25. Zhang L, Wong TN (2015) An object-coding genetic algorithm for integrated process planning and scheduling. Eur J Oper Res 244(2):434-444
26. Dunjko V, Briegel HJ (2018) Machine learning&artificial intelligence in the quantum domain:a review of recent progress. Rep Prog Phys 81(7):074001. https://doi.org/10.1088/1361-6633/aab406
27. Chen Q, Huang M, Xu Q et al (2020) Reinforcement learning-based genetic algorithm in optimizing multidimensional data discretization scheme. Math Probl Eng 2020:1-13
28. Kober J, Bagnell JA, Peter J (2013) Reinforcement learning in robotics:a survey. Int J Robot Res 32(11):1238-1274
29. Baker BC, Nolan JM, O'Neill B et al (2008) Crash compatibility between cars and light trucks:benefits of lowering front-end energy-absorbing structure in SUVs and pickups. Accid Anal Prev 40(1):116-125
30. Rajasekaran M, Ram VH, Subramanian M (2016) Multi-objective optimization of material layout for body-in-white using design of experiments. Int J Veh Struct Syst 8(1):17-22
31. Jiang Z, Chen W, Fu Y et al (2013) Reliability-based design optimization with model bias and data uncertainty. SAE Int J Mater Manuf 6(3):502-516
32. Shi L, Yang RJ, Zhu P (2013) An adaptive response surface method for crashworthiness optimization. Eng Optimiz 45(11):1365-1377
33. Zhan Z, Fu Y, Yang R et al (2013) On stochastic model interpolation and extrapolation methods for vehicle design. SAE Int J Mater Manuf 6(3):517-531
34. Fang H, Rais-Rohani M, Liu Z et al (2005) A comparative study of metamodeling methods for multi-objective crashworthiness optimization. Comput Struct 83(25/26):2121-2136
35. Yang J, Zhan Z, Zheng K et al (2016) Enhanced similarity-based metamodel updating strategy for reliability-based design optimization. Eng Optimiz 48(12):2026-2045
36. Ding R, Lin DK, Wei D (2004) Dual-response surface optimization:a weighted MSE approach. Qual Eng 16(3):377-385

[1]	Jiang Guo, Bin Wang, Zeng-Xu He, Bo Pan, Dong-Xing Du, Wen Huang, Ren-Ke Kang. A novel method for workpiece deformation prediction by amending initial residual stress based on SVR-GA[J]. Advances in Manufacturing, 2021, 9(4): 483-495.
[2]	Jia-Hua Zhang, Ai-Ping Li, Xue-Mei Liu. Hybrid genetic algorithm for a type-II robust mixed-model assembly line balancing problem with interval task times[J]. Advances in Manufacturing, 2019, 7(2): 117-132.
[3]	Vedat Savas, Cetin Ozay, Hasan Ballikaya. Experimental investigation of cutting parameters in machining of 100Cr6 with tangential turn-milling method[J]. Advances in Manufacturing, 2016, 4(1): 97-104.

Improved genetic algorithm based on reinforcement learning for electric vehicle front-end structure optimization design

Improved genetic algorithm based on reinforcement learning for electric vehicle front-end structure optimization design

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价