A novel educational competition optimizer for precise parameter optimization in proton exchange membrane fuel cells

M Aljaidi and P Jangir and Arpita and SP Agrawal and SB Pandya and A Parmar and AF Alkoradees and R Jangid and TR Agrawal, IONICS, 31, 10777-10803 (2025).

DOI: 10.1007/s11581-025-06568-8

Accurate parameter optimization of proton exchange membrane fuel cells (PEMFCs) is essential for maximizing their performance and long-term reliability; however, it remains a complex challenge due to the highly nonlinear nature of their voltage-current characteristics and diverse operating regimes. This study introduces the educational competition optimizer (ECO), a novel metaheuristic algorithm inspired by competitive dynamics within educational systems. ECO utilizes a structured three- phase progression-elementary, middle, and high school stages-that facilitates a dynamic transition from global exploration to local exploitation, improving search efficiency and convergence stability. The algorithm is applied to the parameter estimation of six widely studied PEMFC stacks under static operating conditions, where manufacturer- provided polarization data is available. Comparative evaluations against nine state-of-the-art optimization algorithms (ROA, MVO, SCA, HHO, AOA, SSA, WOA, GWO, and ALO) reveal that ECO consistently achieves the lowest sum of squares error (SSE), demonstrating superior modeling accuracy, robustness, and computational efficiency. These findings underscore the efficacy of ECO in addressing nonlinear static PEMFC optimization problems. Although this work focuses on steady-state modeling, the integration of ECO with dynamic PEMFC models-capable of capturing load transients and environmental fluctuations-is identified as a critical direction for future research, with the aim of enhancing real-world applicability in advanced fuel cell control systems.

Return to Publications page