Standard and Quasi Oppositional bonobo optimizers for parameter extraction of PEM fuel cell stacks

The accuracy of the mathematical model used to simulate the electrochemical characteristics of fuel cell (FC) stacks has a significant impact on the quality of the electrochemical simulation. Due to the lack of manufacturers' information, a set of unidentified parameters must be properly estimated for establishing an accurate representation of proton exchange membrane FCs (PEMFCs). In this research, the Bonobo Optimizer (BO) and the modified Quasi Oppositional BO (QOBO) are applied for determining unknown design parameters of various typical PEMFCs. The obtained results in the form of polarization curves are compared with the corresponding experimental ones to validate the appropriateness and adequacy of the used algorithms. The findings obtained using the original BO and modified QOBO algorithms are compared to those acquired using other current optimization approaches, revealing that the QOBO algorithm outperformed the conventional BO algorithm in solving the optimization problem. Further, a statistical analysis is carried out that guaranteed the robustness and reliability of the developed QOBO.

Automated summary

The accuracy of the mathematical model used to simulate the electrochemical characteristics of fuel cell stacks is crucial. In this research, the Bonobo Optimizer and a modified Quasi Oppositional Bonobo algorithm are applied to determine unknown design parameters of different types of proton exchange membrane fuel cells. The results are validated by comparing with experimental data, and the modified algorithm outperforms the traditional one in solving the optimization problem.