Parameter estimation of PEM Fuel Cell Electrical Equivalent Model using Hybrid Optimization

Fuel Cells (FCs) are the most reliable sources as compared to the sources like photovoltaic (PV) and wind based energy systems. Among various types of FCs, Polymer Electrolyte Membrane Fuel Cells have been widely employed for low and medium power applications due to the advantages of high efficiency and low operating temperature. An equivalent mathematical model of PEMFC with accurate model parameters is essential for analyzing its behavior and characterizing the voltage-current (V-I) relation for grid integration. Traditional optimization techniques for parameter identification possess the limitation of trapping in the local optimum point and high computational time. In this regard, a hybrid optimization technique in which a relatively simpler meta-heuristic method i.e. Artificial Bee Colony (ABC) algorithm is hybridized with a local search technique i.e. Hooke-Jeeves algorithm is proposed in this paper. As compared to existing optimization based approaches, the proposed hybrid method possesses a high probability of global convergence with less number of tuning parameters.