A review and hybrid metaheuristic approach for PEMFC parameter estimation

The primary objective of this study is to review state-of-the-art metaheuristic optimization algorithms and highlight their key contributions to parameter estimation in Proton Exchange Membrane Fuel Cells (PEMFCs). Accurate parameter estimation is crucial for reliable PEMFC modeling; however, the complex, nonlinear, and multivariate nature of PEMFC models makes this task particularly challenging. This paper provides a comprehensive and systematic review of PEMFC parameter estimation, covering fundamental concepts, mathematical formulations, optimization problem definitions, and the various solution methodologies reported in the literature. The study focuses on techniques for estimating both linear and nonlinear parameters, with special emphasis on identification algorithms, which form a foundation for developing effective global energy management strategies. Initially, different PEMFC models with diverse classifications and objectives are discussed. Subsequently, parameter estimation is conducted for three widely studied semi-empirical PEMFC models—STD 250 W, 500 W BCS PEMFC, and NedStack PS6 PEMFC—using a novel hybrid Artificial Lemming Algorithm (ALA) combined with the Dung Beetle Optimizer (DBO), referred to as DALA. The statistical significance of DALA’s performance is validated using the Wilcoxon rank-sum and multiple comparison tests. The obtained results demonstrate that DALA is a robust and highly effective approach for PEMFC system identification, showing strong potential for applications in digital twin development, advanced control systems for automotive applications, and the broader advancement of renewable energy technologies.