Parametric Optimization of Proton Exchange Membrane Fuel Cell using Successive Trust Region Algorithm

In this paper the parameters of a static voltage model of Nexa-1.2 KW proton exchange membrane fuel cell are identified using nonlinear optimization technique. The model is based on semi-empirical formulae defined by parametrical equations characterizing the voltage current relationship of the fuel cell operation. Trust region optimization algorithm is used successively to minimize the objective functions based on two different performance criterion. The results of the optimized model obtained from simulation conform with the experimental data with very small uniform error. The derived model shows significant reduction in the mean square error as compared to the existing optimized PSO model.