Accurate key parameters estimation of PEM fuel cells using self-adaptive bonobo optimizer

The present study introduces an efficient Self-Adaptive Bonobo Optimizer (SaBO) for identifying the unknown parameters of the proton exchange membrane fuel cell (PEMFC). A comparative analysis between recent robust approaches, such as Gradient-based Optimizer (GBO), Bald Eagle Search Algorithm, and Rime-Ice algorithm (RIME), has been introduced. The basic concept is to minimize the mean bias error between the measured and predicted stack voltage. The main results show that although the techniques were close, in contrast, the SaBO optimizer provides a better superiority than GBO, BES, and RIME for an optimum forecast of the PEMFCs model. Moreover, the best fitness was achieved with the SaBO at 0.0367 (V) for the Heliocentris FC-50, and 0.1150 (V) for Nexa (R) 1200, also, with the minimum deviation of 0.0027 & 0.0172, and high efficiency. These achievements denote that the SaBO algorithm is more stable and robust for PEMFC parameter estimation.