Moth flame-random search optimization of a zero-dimensional model of a proton exchange membrane fuel cell

Modelling of proton exchange membrane fuel cell (PEMFC) is important for better understanding, simulation, and design of high-efficiency fuel cell systems. PEMFC models are often multivariate with several nonlinear elements. Metaheuristic algorithms that are successful in solving nonlinear optimization problems are good candidates for this purpose. This study proposes a new metaheuristic algorithm called MFORS that uses the advantages of the moth-flame optimization algorithm in global search and the non-deterministic properties of the random search algorithm to identify the optimal parameters of the PEMFC model. The sum of squared errors between the estimated and measured voltage is a quality of fit criterion. To show the effectiveness of the proposed algorithm, two case studies of zero-dimensional models of SR-12 Modular PEM Generator and Ballard Mark V fuel cell are considered. The sum of squared errors for the parameter estimated of electrical PEMFCs by the proposed MFORS algorithm is compared with recent works. The results showed that by the MFORS algorithm, the minimum sum of absolute errors of the actual stack voltage and the simulated stack voltage in the two PEMFC are 0.095037 and 0.018019, compared with the second-best algorithm results giving 0.09681 and 0.8092, respectively.