Semi-empirical PEM fuel cells model using whale optimization algorithm

The accuracy of the fuel cell model (FC) plays an important role, since it affects the simulation results. Obviously, this can appear in different systems of distributed generations and hybrid microgrids. This article presents a new application of the whale optimization approach (WOA) to obtain unidentified parameters of the proton exchange membrane (PEM) FC model. The ultimate objective of the current work is to develop a precise PEMFC model, which provides true results of modeling and simulation of these FCs. In this sense, the I-V characteristics of the PEMFC are non-linear, including seven unknown parameters due to the lack of data from the manufacturer. This problem can be expressed mathematically as a non-linear optimization problem, where the sum of the error squared between the measured FC voltage and its output voltage model defines the fitness value. WOA is applied directly to minimize the objective function under the predefined constraints. The estimated PEMFC model is certified by the numerical results shown, which are performed under various conditions of temperature and regulating pressures. The value of the proposed WOA-based PEMFC model is appraised by comparing the demonstrated results with the empirical results of several typical PEMFCs such as Ballard Mark V, the AVISTA SR-12 PEM generator, the 250 W and the Horizon H-12 stacks. The results based on WOA are compared with other results based on optimization competing methods. The application of WOA can lead to the creation of a precise PEMFC model.