Genetic algorithm‐based higher‐order model reduction of proton exchange membrane fuel cell

Abstract

This study investigates the performance and simulation analysis of a larger-scale proton exchange membrane fuel cell model using a lower-order model. With model order reduction studies, the present work aims to demonstrate the effectiveness of the proposed Genetic algorithm technique in terms of various reduced order transfer functions. The proposed methodology can assist the higher-order model in meeting its requirements, including stability, computation complexity, and the issue of local optima entrapment. This study demonstrates the efficacy of the proposed method by comparing it to other reduction techniques, such as the factor division method, the stability method, and the truncation method, utilizing a variety of performance metrics, including error indices, non-parametric tests, through in frequency, and time-domain analyses. The proposed method for model reduction offers advantages in terms of perturbation, stability, and parameter uncertainty.