Honey badger optimizer for extracting the ungiven parameters of PEMFC model: Steady-state assessment

Abstract

In this work, a novel attempt is performed to optimally identify the seven ungiven parameters of the proton exchange membrane fuel cells (PEMFCs) steady-state model. A fitness function is adapted to reduce the sum of quadratic errors (SQEs) between the experimentally measured voltages and the corresponding computed values. A honey badger optimizer (HBO) is utilized to minimize the SQEs, exposed to a group of inequality bounds. Three test cases of well-known commercial PEMFCs units as benchmarking are elucidated and discussed over different steady-state operating conditions. Substantial comparisons to the other up-to-date optimizers published in the art-of-literature are provided to appraise the HBO’s viability. It’s worth highlighting that the values of maximum percentage biased voltage deviations for Ballard Mark, SR-12 and 250 W stacks are equal to 2.696%, −0.016% and 1.595%, respectively. Besides, several statistical measures are applied to indicate the proposed HBO robustness and accurateness. Furthermore, a sensitivity study based on SOBOL indicators is performed at which the influence of small deviations of the seven extracted parameters on the PEMFC's model, is comprehensively illustrated. It can be confirmed that the HBO asserts its capability to tackle this task effectively rather than others.