Multi-verse optimizer for identifying the optimal parameters of PEMFC model

In this paper, a recent optimization algorithm named multi-verse optimizer (MVO) is applied to identify the optimal parameters of the proton exchange membrane fuel cell (PEMFC) under certain operating conditions. Seven parameters to be optimized are ξ₁, ξ₂, ξ₃, ξ₄, λ, R_c, b in order to obtain polarization curves closely converged to those obtained in the manufacture's datasheet. MVO is characterized by simple construction, less controlling parameters and requiring less effort in computation process. Four sets of experimental voltage stack are taken into consideration; two of them are used for optimization process while the others are used for model validation in the presence of two types of parameter constraints. Comparative studies including statistical parameters with two types of methods are performed; the first methods are reported in the literature like SGA, HGA, HABC, RGA and HADE while the second approaches are programmed such as grey wolf optimizer (GWO), artificial bee colony (ABC), mine blast algorithm (MBA) and flower pollination algorithm (FPA). The obtained results reveal that MVO is the best choice among the others since it presents less fitness function and less convergence time.