Robust parameter identification strategy of solid oxide fuel cells using bald eagle search optimization algorithm

Enhancing the solid oxide fuel cell (SOFC) is an attractive topic for academic research. One of the investigated solutions is to enhance the modeling accuracy. This paper presents a robust strategy based on a bald eagle search optimization (BES) algorithm to determine the best parameters of the solid oxide fuel cell (SOFC) model. Six unknown parameters to be determined are E, A, I_o, R_Ω, B, and I_max. The sum of mean squared error (SMSE) between the measured and estimated stack voltages is minimized. Two scenarios of the SOFC operation mode are considered. The first is a steady-state model, and the second is the transient/dynamic-state model. In a dynamic state-based model, a demand disturbance is implemented, and the performance of the constructed model is examined. Furthermore, a comparison with previously reported approaches and other programmed algorithms has been performed to approve the proposed strategy. The high performance of the BES algorithm in terms of accuracy and robustness makes it an excellent algorithm for this application. To approve the performance of the proposed strategy, it will be investigated for several operating temperatures with multiple algorithms. The relevant results are auspicious where the final fitness, which expresses the identification accuracy, has been minimized by the BES to 2.6906 × 10⁻⁶ for 1073 K, 6.3942 × 10⁻⁶ for 1213 K 1.0401 for 1273 K.