Battery box design of electric vehicles using artificial neural network-assisted catch fish optimization algorithm

Abstract

In engineering and other fields, metaheuristic algorithms are increasingly used to solve challenging optimization problems. High-dimensional and multimodal issues are complex for traditional optimization techniques, which has led to the development of hybrid metaheuristics that are improved by artificial neural networks (ANNs). To improve search efficiency and solution accuracy, this work presents an ANN-assisted Catch Fish Optimization Algorithm (MCFOA), which draws inspiration from conventional fishing methods. Numerous engineering applications, such as the optimal design of the side profile of an electric vehicle battery box, shell and tube heat exchanger, industrial gear optimization, and welded beam cost minimization, show off the efficacy of MCFOA. For the novel battery case problems, the modified optimizer realized a 20 % improvement in the design compared to the initial design, as well as a 4.5 % improvement compared to the Starfish optimizer. Moreover, for the engineering design problems, the modified optimizer realized 4-10 % better results in terms of the best values of the fitness function. This shows the applicability and implementation of the proposed optimizer for the optimization of real-world engineering problems.