论文部分内容阅读
PEMFC电堆是一个包括流动、传热、传质和电化学反应等多种物理化学现象的复杂机体,其建模问题是一个具有挑战性的问题。本文提出一种基于RLSSVM-CPSOSA模型的燃料电池建模方法。CPSOSA算法将CO算法、模拟退火算法和粒子群算法有机结合,以克服粒子群算法早熟收敛的不足。然后利用CPSOSA算法对鲁棒最小二乘支持向量机模型(RLSSVM)进行参数寻优,从而获得RLSSVM-CPSOSA燃料电池模型。以MATLAB平台搭建PEMFC电堆模型并进行仿真研究,结果表明所提出的RLSSVM-CPSOSA模型的有效性和良好的预测精度。为PEM燃料电池实时控制系统奠定了基础。
PEMFC stack is a complex body including many physical and chemical phenomena such as flow, heat transfer, mass transfer and electrochemical reaction. The modeling problem is a challenging one. This paper presents a fuel cell modeling method based on RLSSVM-CPSOSA model. CPSOSA algorithm combines CO algorithm, simulated annealing algorithm and particle swarm optimization algorithm to overcome the shortcomings of PSO premature convergence. Then the robust least squares support vector machine model (RLSSVM) is optimized by using CPSOSA algorithm to get the RLSSVM-CPSOSA fuel cell model. The PEMFC stack model is built by MATLAB platform and the simulation is carried out. The results show the effectiveness of the proposed RLSSVM-CPSOSA model and good prediction accuracy. PEM fuel cell real-time control system laid the foundation.