Parameter Estimation of Hydrogen Fuel Cell using Metaheuristic Algorithm

This study presents an advanced approach of modelling proton exchange membrane fuel cells (PEMFCs) for electric vehicles (EVs) by employing adaptive particle swarm optimization (APSO) for parameter estimation. PEMFCs, known for their solid electrolytes, operational flexibility, fast startup, and rapid response to electrodynamic changes, pose significant modelling challenges due to complex interactions involving mass and heat transfer, electrochemical reactions, and energy losses. To overcome these challenges, APSO is utilized which adapts its search behaviour to achieve optimal results. This enhances the accuracy of the PEMFC models compared to conventional methods. Even in situations with parametric noise, the results show that APSO performs better than other optimization techniques in the estimation of the six primary PEMFC parameters. This demonstrates its resilience, sustainability for analysing noisy realworld data, and dependability in enhancing PEMFC modelling for electric vehicle applications.