- https://doi.org/10.1109/icesa66763.2025.11281021
PEM Fuel Cell Parameter Estimation Using a Modified Particle Swarm Optimization Algorithm
- Oct 22, 2025
- Muhammad Ali Mughal +2 more
The simulation, control and analysis of the systems based on Proton Exchange Membrane Fuel Cells (PEMFC) demand a precise model with accurately defined parameters. This study introduces a modified swarm intelligence optimization approach, the Modified Particle Swarm Optimization (MPSO) algorithm, to estimate the parameters of the fuel cell model. The MPSO algorithm is an enhanced version of the standard particle swarm optimization (SPSO) technique, designed to improve performance by incorporating stagnation and aggressive mutation strategies, supplementing in avoiding premature convergence and amplifying the exploitation. In this method, a mathematical framework is established to guide the parameter optimization process. An objective function, based on the sum of squared errors (SSE) between the experimental and estimated voltage values, is constructed to facilitate optimal parameter estimation. The MPSO algorithm is then employed to determine the parameters for Nedstack PS6 fuel cell stack. The results are compared with several well-known optimization algorithms from the literature and the standard PSO. The comparative analysis demonstrates that the proposed MPSO algorithm outperforms others in terms of accuracy, search efficiency, and convergence speed. The MPSO has achieved an SSE of 2.0655 with a lower standard deviation of 0.6137 for the thirty runs, comparatively.