Optimal parameter extraction of PEM fuel cells by meta-heuristics
Guiju Zhang, Caiyuan Xiao, Navid Razmjooy
Abstract
This paper represents a new technique for optimal modelling and simulating a proton exchange membrane fuel cell (PEMFC) system to assure dependable modelling. The main idea is to utilise a newly developed meta-heuristic, called Chaos Owl Search Algorithm (COSA) to optimal selecting of the model parameters of the PEMFC stacks by minimising the Sum of Squared Error (SSE) between the estimated and the measured output voltage for two different case studies. By applying 50 independent runs with the algorithm, it is analysed and compared with some literature meta-heuristics including Bat Algorithm (BA) Firefly algorithm (FFA), and Multi-verse optimiser (MVO) in terms of convergence speed and minimum SSE. The final results declare that the proposed method achieves the best convergence speed in comparison with others. The results also determine the high efficiency of the presented method.