Adaptive and efficient optimization model for optimal parameters of proton exchange membrane fuel cells: A comprehensive analysis

Optimum modeling of the proton exchange membrane fuel cell (PEMFC) has attracted considerable research over the last decades to simulate, control, evaluate, manage, and optimize the performance of PEMFC stacks. The main problem in optimal modeling is that the model parameters are not provided by manufacturers, and the empirical dataset points are not sufficient to accurately model the cell. Therefore, a new approach based on the improved chimp optimization algorithm (IChOA) is proposed to define the uncertain parameters of the PEMFC. A ranking-based updating strategy and a balanced exploration and exploitation strategy (BEES) are employed here within the IChOA. In the first strategy, the unbeneficial solutions in the population are replaced with other solutions covering other regions, which are unreachable by the original one. The second strategy aims at utilizing iteration as much as possible so that, at the beginning, the method maximizes the exploration operator in the first half of the optimization process to ensure the balance between the exploration and exploitation framework; and then, in the second half, the exploitation capability is maximized attempting to find a better solution than the best-so-far. The proposed IChOA is validated by three well-known commercial PEMFCs, namely 250 W stack, Ballard Mark V, and AVISTA SR-12 500 W modular. The best results of the IChOA are compared with 15 nature-inspired metaheuristics algorithms and another one known as gradient-based optimizer under various statistical analyses and under varied operating conditions. The superiority of the IChOA is demonstrated in terms of convergence stability, and final accuracy.