A reliable starfish optimization algorithm for establishing equivalent circuit of proton exchange membrane fuel cell

Abstract The proton exchange membrane fuel cell (PEMFC) can be effectively simulated, managed, and controlled via accurate identification of its model’s unknown parameters. The nonlinear properties of fuel cells (FCs) and the absence of complete data in the datasheet cause certain obstacles in this procedure. This study suggests a novel approach employing the starfish optimization algorithm (SFOA) for designing the equivalent circuit of PEMFC by evaluating its parameters utilizing experimental data. This algorithm is highly efficient as well as avoiding local optima due to exploitation and exploration balance. Five distinct FCs are analyzed: Horizon_H12, Horizon H-1000XP PEMFC stack, Temasek_1kw, SR-12PEM 500-W, and BCS-500W PEMFC stacks. The suggested SFOA is evaluated in comparison with previously published techniques of seagull optimization algorithm (SOA), escape optimizer (ESC), gold rush optimizer (GRO), tunicate swarm algorithm (TSA), beluga whale optimization (BWO), success history intelligent optimizer (SHIO), enhanced transient search optimizer (ESTO) and Newton–Raphson-based optimizer (NRBO). Decreasing the sum square error (SSE) between the calculated and measured output voltages is the targeted goal. The minimum values of SSE obtained through the suggested SFOA are 4.662696E−04, 3.948527E−01, 5.4322160E−01, 1.056370, and 1.169778E−02 for Horizon_H12, Horizon H-1000XP PEMFC stack, Temasek_1kw, SR-12PEM 500-W, and BCS-500W PEMFC stacks, respectively. Additionally, the PEMFC dynamic simulation model is constructed in Matlab/Simulink, while step load disruption is employed for analyzing the model’s performance. The acquired results indicate the success of the suggested SFOA in constructing dependable models of different PEMFCs.