Multi-objective optimization of gradient porosity gas diffusion layer of PEMFC with metal foam flow fields by LBM and AI-MOO

Abstract

A gas diffusion layer (GDL) with a porosity gradient has been widely recognized as an effective approach to enhancing the performance of proton exchange membrane fuel cells (PEMFCs). However, research on gradient porosity GDL structures for PEMFCs with metal foam flow field (MFFF) remains insufficient. In this study, lattice Boltzmann methods (LBMs) are used to investigate liquid water and oxygen transport. The influence of porosity gradient GDLs on multiphase transport and PEMFC performance is comprehensively evaluated. It is observed that the current density and uniformity index exhibit a non-monotonic dependence on the porosity gradient of GDL. Furthermore, artificial intelligence-based multi-objective optimization (AI-MOO) is implemented to synergistically optimize the structural parameters of the gradient porosity GDL. Compared with unoptimized structures, the optimized GDL increases current density by 3.9469% and reduces the uniformity index by 75.5979%. This study provides theoretical guidance on the gradient-structure design of GDLs in PEMFCs with MFFFs.