Robust Parameter Identification of PEM Fuel Cell Models Using Huber Loss and an Improved Gray Wolf Optimizer
International Journal of Energy Research · 2026-01 · 2026 (1)
DOI: 10.1155/er/7671334 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Energy Research, online erschienen am 13.11.2026. Verfasst von Badr M. Al Faiya und 6 weiteren Autor:innen. Die Arbeit nennt 94 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jazan University.
Abstract (Originalsprache)
Accurate parameter identification is essential for reliable proton exchange membrane fuel cell (PEMFC) modeling because of the strong nonlinearities associated with electrochemical, ohmic, and mass‐transport phenomena. This study investigates an improved gray wolf optimizer (IGWO) for estimating seven unknown parameters of a semiempirical PEMFC model using the Huber loss function as the optimization objective. The proposed framework is evaluated on three commercial PEMFC stacks, namely Nedstack PS6, Horizon 500 W, and Ballard BCS 500 W, and benchmarked against GWO, ECO, WOA, and SSA. Model performance is assessed using the sum of squared errors (SSE), root mean square error (RMSE), mean absolute error (MAE), convergence behavior, pointwise voltage errors, and repeated‐run statistics. For the Nedstack PS6 stack, IGWO achieved the lowest reported SSE of 0.00996 V 2 with an RMSE of 0.025768 V, closely followed by ECO. For the Horizon 500 W and BCS 500 W stacks, ECO produced the lowest SSE values of 0.0012 and 0.01159 V 2 , respectively, while IGWO remained highly competitive. The results therefore demonstrate that optimizer performance is stack dependent and that no single method consistently provides the best fit across all operating cases. Huber‐threshold analysis further showed that the best‐fit residuals remain within the quadratic region at δ = 1 V, whereas a +2 V perturbation test confirmed the reduced sensitivity of Huber loss to large residuals compared with SSE. Overall, IGWO provides a competitive and effective solution for nonlinear PEMFC parameter identification, particularly when combined with robust error‐based evaluation.
Förderung
- Jazan University
Zitieren
APA
Faiya, B. M. A., Alaas, Z. M., Mansour, H. S. E., Omar, A. I., Smaili, I. H., Moustafa, G., & Elmetwaly, A. H. (2026). Robust Parameter Identification of PEM Fuel Cell Models Using Huber Loss and an Improved Gray Wolf Optimizer. International Journal of Energy Research, 2026(1). https://doi.org/10.1155/er/7671334
Vancouver
Faiya BMA, Alaas ZM, Mansour HSE, Omar AI, Smaili IH, Moustafa G, et al. Robust Parameter Identification of PEM Fuel Cell Models Using Huber Loss and an Improved Gray Wolf Optimizer. International Journal of Energy Research. 2026;2026(1). doi:10.1155/er/7671334
BibTeX
@article{Faiya2026,
author = {Faiya, Badr M. Al and Alaas, Zuhair M. and Mansour, Hany S. E. and Omar, Ahmed I. and Smaili, Idris H. and Moustafa, Ghareeb and Elmetwaly, Ahmed Hussain},
title = {Robust Parameter Identification of PEM Fuel Cell Models Using Huber Loss and an Improved Gray Wolf Optimizer},
journal = {International Journal of Energy Research},
year = {2026},
volume = {2026},
number = {1},
doi = {10.1155/er/7671334}
}
EUVIMED