Optimal Model Parameter Estimation of PEM Fuel Cell Using Mountaineering Team-Based Optimization

In this paper, a recently developed meta-heuristic (MH) algorithm, namely mountaineering team-based optimization (MTBO) has been applied to estimate seven unknown model parameters of proton exchange membrane (PEM) fuel cell. The mathematical model of the PEM fuel cell is considered as complex and nonlinear, and it contains several unknown parameters. Identification of these parameters is difficult using classical approaches. The MH algorithms are capable of solving complex problems. Therefore, MTBO has been used to identify the values of these parameters. A sum of squared error (SSE)-based objective function is considered and minimized in this study. The error is the difference between the experimental and estimated value of voltages. The polarization characteristics such as I-V and I-P curves have been used to evaluate the accuracy of the MTBO algorithm. Results obtained using MTBO have been compared with four competitive algorithms, namely Gray wolf optimization (GWO), Harris hawk optimization (HHO), differential evolution (DE), and whale optimization algorithm (WOA). The best value of SSE obtained by the MTBO algorithm is 1.1698E−02. Moreover, convergence curves and box plot study have been used to check the speed and reliability of the algorithm. Moreover, statistical study including mean, min, and standard deviation values is also performed to check the robustness of MTBO algorithm. It is concluded that MTBO performs better than other competitive algorithms in solving the problem of parameter estimation of PEM fuel cells.