Parameter estimation of proton exchange membrane fuel cell using a novel meta-heuristic algorithm

In recent years, proton exchange membrane fuel cells (PEMFCs) have been known to be a viable method for meeting the electrical energy needs, thereby enhancing the overall reliability of renewable energy systems. PEMFCs demonstrate various promising attributes like pollution-free, totally sustainable, non-self-discharging. These need hydrogen as fuel, and air for their operation, while the final product is pure water only. Thus, under varying operating conditions, the appropriate modeling and parameter optimization of PEMFCs have gained considerable importance in recent times. The evolutionary optimization approaches had been utilized in recent past for estimating PEMFCs parameters as exact modeling of the same does not exist in the literature. For the evaluation of PEMFCs performance criteria, a newly proposed algorithm is developed in this manuscript i.e. black widow optimization (BWO). Firstly, the performance of this proposed algorithm is checked by complex benchmark results. After that, this proposed algorithm is applied to extract the parameters of PEMFCs models under different operating temperatures. The parameter optimization results are obtained using BWO and are further compared with those obtained with five other algorithms, i.e., particle swarm optimization (PSO), multi-verse optimizer (MVO), sine cosine algorithm (SCA), whale optimization algorithm (WOA), and grey wolf optimization (GWO). The complete error analysis is carried out for the two data sheets of the PEMFCs to establish the superiority of BWO. It has been observed that the developed proposed algorithm gives better results when compared to those obtained with rest of the algorithms considered in this work. After calculating the error, non-parametric test is performed which suggests that the BWO is better than the rest of the compared algorithms.