Optimization of Fuel Cell Hybrid System Configuration via Modified Multi-Objective Particle Swarm Algorithm

• Published in: 2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) pp. 499 - 503
• Main Authors: Liu, Yingfang, Sun, Zhendong, Chen, Zonghai
• Format: Conference Proceeding
• Language: English
• Published: IEEE 21.07.2024
• Subjects: configuration optimisation; Convergence; Costs; Energy management; fuel cell hybrid power systems; Fuel cells; Manufacturing; Mechanical power transmission; MOPSO based on competitive mechanisms; Optimization; Particle swarm optimization; Resource management; Systems simulation
• DOI: 10.1109/CCSSTA62096.2024.10691839

This paper presents a novel approach to the capacity allocation problem in fuel cell hybrid vehicles. It introduces a multi-objective evolutionary algorithm nested Dynamic Programming (DP) strategy aimed at minimizing both manufacturing and operating costs. The outer loop employs a fast-converging Multi-Objective Particle Swarm Optimization (MOPSO) algorithm based on competitive mechanisms for parameter matching, while the inner loop employs DP for energy management. The effectiveness of the proposed method is validated through simulation under specific operating conditions. Comparative analysis with traditional MOPSO demonstrates superior performance in terms of solution set diversity and convergence, affirming the efficacy of the proposed approach.