A novel approach for fuel cell parameter estimation using simple Genetic Algorithm

A new problem formulation for effective identification of fuel cell parameters is proposed. The proposed formulation is solved by applying Genetic Algorithm optimization technique. The algorithm steps are coded in MATLAB and the objective function is solved for PEM fuel cell. In order to estimate the performance of the proposed formulation; extensive simulations are performed with both the proposed and conventional objective functions and the results obtained are compared. Further, a comprehensive evaluation based on objective function value, convergence speed and absolute voltage error value is also made to prove the superiority of the proposed formulation over the conventional curve fitting approach. (C) 2015 Elsevier Ltd. All rights reserved.