Parameter Identification of PEMFC Model Using Improved Dung Beetle Optimization Algorithm

A proton exchange membrane fuel cell (PEMFC) is a complex system with multiple inputs and outputs, nonlinearity and strong coupling, and the establishment of an accurate model is the basis for evaluating the performance of PEMFC and developing control strategies. As the majority of the current intelligent algorithms tend to become stuck in local optimum when attempting to determine the PEMFC model’s parameters, resulting in low accuracy of parameter identification and poor model generalization ability, we propose an Improved Dung Beetle Optimization (IDBO) algorithm to identify the PEMFC model’s best parameters. To evaluate the IDBO algorithm’s performance, we identify the model optimal parameters of two typical commercial stacks, BCS 500 W and NedStack PS6, and the self-developed 3 kW PEMFC system, with the minimization of the sum of squared errors between the experimental output voltages and the model output voltages as the objective function. The verification results indicate that the IDBO algorithm has better convergence performance and higher parameter identification exactitude than the DBO algorithm. The robustness and applicability of the IDBO algorithm in addressing the issue of parameter identification of the PEMFC models are verified.