Enhancing PEMFC parameter estimation: A comparative literature review and application of the Walrus optimization algorithm and its hybrid variants

This paper presents a novel approach for the estimation of equivalent circuit parameters of proton exchange membrane fuel cells using the Walrus optimization algorithm (WaOA) and its hybrid variants: SaWaOA, combined with simulated annealing and CWaOA, incorporating chaotic sequences. The evaluation was conducted on two types of fuel cells, Ballard-Mark-V 5 kW and BCS 500 W. To ensure an objective assessment, validation of the results was performed using statistical measures: root mean square error (RMSE), sum of squared errors (SSE), mean absolute error, and mean absolute percentage error. The testing was carried out by implementing the recommended optimization algorithm across eight different variants, where the boundaries of the parameters lambda and R-c were varied during estimation. Results indicate that WaOA and its hybrid variants achieve high precision in parameter estimation, evidenced by reduced RMSE and SSE values compared to methods from the literature. Additionally, the influence of changing all parameters on estimation precision was analyzed, as well as the effects of varying operating temperatures and pressures on the output characteristics of the tested fuel cells. The proposed WaOA can significantly enhance the accuracy of parameter estimation for fuel cells in various operating conditions, opening possibilities for its broader application in industrial and transport systems.