Parametric Optimization of Proton Exchange Membrane Fuel Cell Using Chaotic Swarm Intelligence Technique

The proton exchange membrane fuel cell (PEMFC) converts chemical energy into electricity without pollution or noise. PEMFC has a stable electrolyte, reliable, fast start-up, and lightweight portability. Optimal PEMFC parameter prediction is difficult. Due to inaccurate estimation, many meta-heuristic models are inaccurate. Particle swarm, Ant colony, and grey wolf algorithms are tested for fuel cell parameter estimation. These traditional algorithms benefit from chaotic maps for exploitation and exploration. Analyzing V–I and P–I polarization curves, hybrid approaches estimate unknown parameters. Results are compared with others to support the identifying strategy. This project examined SR-12 and NedSstack PS-6 PEMFC stacks. The objective function was shown by the sum of square errors between the experiments, and once an appropriate map is identified, the single chaotic map inclusion method can perform better than the multiple chaotic map scheme, and the “Iterative” approach in chaotic swarm is the most promising for multidimensional challenges.