Accurate modeling of various proton exchange membrane fuel cell models using an effective algorithm based on artificial gorilla troops optimizer

Fuel cells are one of the most important alternative energy sources at the present time, and using them is a good way to decrease the dependency on traditional power plants. Between several FC stacks, the most widely used type of fuel cell is the proton exchange membrane fuel cell (PEMFC). This article uses a modified artificial gorilla troops optimizer based on the convex Lens imaging learning strategy, Tent-chaos and Cauchy mutation (CTGTO) algorithm to find the best solution for determining the PEMFC model’s unknown parameters. With this adjustment, the standard GTO demonstrates improved exploration capabilities, allowing it to avoid being trapped in local optima and to find the global optimum across the entire solution space more effectively. Furthermore, a Tent-chaos and Cauchy mutation makes the suggested CTGTO approach more random with good optimization results. A deeper analysis using seven benchmark data sets is also performed to confirm the superiority of the proposed algorithm concerning Chaos Game Optimization (CGO), weighted mean of vectors (INFO) algorithm, artificial hummingbird algorithm (AHA), gradient based optimization (GBO), and original GTO. In addition, five different types of PEM fuel cells: 250 W FC stack, BCS 500 W, Ballard Mark V, Horizon H-12 stack, and NedSstack PS6, were employed to prove the superiority of the CTGTO. To approve the superiority of CTGTO, the results were compared with those obtained using different algorithms. Furthermore, a statistical evaluation of the optimization outcomes has been carried out to verify the resilience and stability of the suggested CTGTO method in attaining the best possible outcome for the estimation problem of PEMFC parameters. In addition, the accuracy of the determined values for the unknown parameters is determined through investigation and assessment of the dynamic behavior of several PEMFC stacks operating at varying pressure and temperature levels. When compared to other contemporary optimization techniques and the traditional GTO, the simulation results confirmed the improved technique’s efficacy and dependability in determining the optimum values of the PEMFC unknown parameters.