Optimized Particle Swarm Fuzzy Controller for Energy Management Control of Fuel Cell Battery Electric Vehicles

The particle swarm optimization (PSO) algorithm has garnered substantial attention for its straightforward implementation, rapid convergence, and low computational demands in fuel cell battery electric vehicles (FCBEVs) energy system control. To address the challenge of PSO being prone to local optima, this study investigates two enhanced optimization techniques-adaptive PSO-fuzzy logic control (FLC) and simulated annealing (SA) PSO-FLC-and further proposes a hybrid PSO-FLC energy management strategy (EMS) that combines the strengths of both approaches. A co-simulation platform integrating AVL Cruise for vehicle dynamics and MATLAB/Simulink for energy management control is established to evaluate performance under the Worldwide Harmonized Light Vehicles Test Cycle (WLTC). Results demonstrate that the hybrid PSO-FLC strategy maintains the state-of-charge (SOC) fluctuation within 2% and enhances fuel economy by 3.76%. Additionally, it achieves a 37.04% reduction in convergence time compared to the traditional PSO-FLC, confirming its effectiveness and computational efficiency for real-time EMS applications in FCBEVs.