Accurate Modeling of PEM Fuel Cell With Sensitivity Analysis Using Mirage Search Optimization Algorithm

Abstract

Nowadays, green hydrogen technology is a pivotal innovation for reducing environmental pollution and combating global climate change. In the pursuit of sustainability, proton exchange membrane fuel cells (PEMFCs) are considered a promising solution for optimizing the utilization of green hydrogen and enhancing energy storage capabilities. This article presents a novel application of the mirage search optimization (MSO) algorithm for developing an accurate PEMFC model. Through a comprehensive study of four typical PEMFC stacks, the results demonstrate the superior performance of the proposed MSO algorithm when compared to other optimizers in terms of accuracy and convergence speed. The optimization algorithms used for comparison with MSO include the grey wolf optimizer, whale optimization algorithm, chimpanzee optimization algorithm, and other optimizers from the literature. The enhancement in modeling accuracy by obtaining a better fitness value using MSO over other optimizers is up to 10.7% for NedStack PS6, 7.1% for Ballard Mark 5 kW, 31.5% for BCS 500 W, and 85.39% for Horizon H-500. Furthermore, a sensitivity analysis is carried out to validate the results obtained by MSO and to verify the accuracy of the developed model. Through comprehensive performance assessments, it can be confirmed that MSO is a promising algorithm for accurately estimating the parameters of PEMFC models, as it demonstrates high efficiency and robustness.