Modelling of PEM fuel cell for parameter estimation utilizing clan co-operative based spotted hyena optimizer

The Proton Exchange Membrane Fuel Cell (PEMFC) is considered as one of the most promising energy conversion technologies of the future due to its enormous advantages like low operating temperature, high efficiency, and longevity. PEMFC steady-state model is conventionally established when set of seven unknown parameters are identified. However, due to the considerable nonlinearity, multivariate data, and increased number of parameters; modelling PEMFC characteristics pose difficulties. Due to this reason, herein, a new optimization method based on Spotted Hyena hunting behaviour has been presented for optimum PEMFC model parameter identification problem. The effectiveness of this SHO method in optimizing the PEMFC model is experimentally verified for five distinct PEMFC stacks, including the Ballard Mark V, BCS Stack, NedStack PS6, Temasek, and WNs stacks. A comprehensive assessment with 25 well -established modelling methodologies in literature is carried out to demonstrate the method ' s dependability and effectiveness. Three different multi -attribute decision -making methods such as TOPSIS, MOORA, and COPRAS are used to rank the metaheuristic algorithms in terms of performance. The achievements show that the suggested SHO method ranks first with lowest SSE of 0.0000104, 0.00211, 0.1308, 0.0085, and 0.0146 for the tested five fuel cell stacks. The best objective function efficiency of 98.6034 % is attained by the SHO algorithm hence, asserting its capability in determining the unknown PEMFC model parameters effectively.