Abstract

This study proposes an optimal energy management strategy for dual‐source fuel cell hybrid electric vehicles (FCHEV) utilizing the support vector machine (SVM) classifier. The goal is to optimize power distribution between fuel cells and batteries to enhance vehicle's performance and efficiency. The SVM classifier is trained using a dataset of driving conditions and corresponding optimal power distribution (OPD) values obtained through simulation. The trained classifier predicts real‐time OPD based on driving conditions. In comparison to existing literature, this study conducts a comparative analysis of energy management control strategies like model predictive control (MPC), fuzzy, equivalent consumption minimization strategy (ECMS), proportional‐integral (PI) control, and state machine control (SMC) strategy for FCHEVs using the MATLAB/SIMULINK platform and real‐world driving dataset. The proposed strategy is then tested in a real‐time EV simulator to verify its efficacy. Additionally, this study introduces the SVM classifier technique for selecting the optimal energy management strategy for FCHEVs. Performance analysis using SVM reveals that the MPC control strategy offers the highest efficiency compared to other techniques based on selected features, achieving an average performance of 95% through cross‐validation. This analysis demonstrates the most cost‐effective and fuel‐efficient utilization of electricity flow in a modern energy‐efficient environment.

This paper presents an optimal energy management strategy for dual‐source fuel cell hybrid electric vehicles (FCHEV) using support vector machine (SVM) classifier, aiming to enhance vehicle performance and efficiency. Through comparative analysis, the study identifies model predictive control (MPC) as the most efficient strategy, achieving 95% performance on average, demonstrating its cost‐effectiveness and fuel efficiency in modern energy‐efficient environments.image