Optimal Adaptive Modeling of Hydrogen Polymer Electrolyte Membrane Fuel Cells Based on Meta-Heuristic Algorithms Considering the Membrane Aging Factor

An efficient adaptive modeling criterion for the polymer electrolyte membrane fuel cell (PEMFC) is proposed in this paper, which can facilitate its precise simulation, design, analysis and control. In this work, a number of state-of-the-art algorithms have been adapted to optimize the complex electrochemical PEMFC model. Investigations are carried out not only from the conventional perspective of modeling accuracy but also from a new perspective represented by the impact of process computational time. Here, a novel technique of PEMFC modeling is proposed based on a meta-heuristic optimization algorithm called the wild horse optimizer (WHO). The proposed technique is concerned with the impact of the computational time on dynamic PEMFC modeling. A comprehensive statistical analysis was performed on the results of competing meta-heuristic optimizers that were adapted to a common PEMFC modeling problem. Among them, the proposed WHO approach’s results showed a promising performance in terms of its accuracy and minimum computational time over the other state-of-the-art approaches. For further evaluation of the WHO approach, it was used to optimize additional commercial PEMFC stack models. The results of the WHO approach highlighted its superior performance from the point of view of a high accuracy with a low computational burden, which supports its suitability for online applications.