Numerical study and prediction of water transfer in gas diffusion layer of proton exchange membrane fuel cells under vibrating conditions
Yang Yu, Sheng Chen
Abstract
Proton exchange membrane fuel cells (PEMFCs) operate with vibrations that have a significant effect on the transfer of water and gas in the gas diffusion layer (GDL), as well as the conduction of electrons. However, the available research on the effects of vibration on PEMFC is sparse. The main purpose of this study is to investigate the transfer characteristics of water in GDL under different vibration conditions and to optimize the water content of GDL using computational fluid dynamics (CFD) and machine learning method. The results show that the transfer of water in the GDL can be divided into two stages: water accumulation and drainage. In the accumulation stage, vibration has only a small effect on saturation; during the drainage stage, vibration causes saturation to fluctuate regularly. In addition, vibration has a large effect on the water content of GDL. Vibrations with a high frequency and a large amplitude will accelerate the accumulation of water in the GDL. Moreover, horizontal vibration will cause more flow paths, while vertical vibration will generate a relatively stable flow path. In addition, the established genetic algorithm-back propagation neural network (GA-BPNN) data-driven model has good predictive performance for the saturation of water in the GDL. Compared to the physical simulation approach, the data-driven model has faster computational efficiency and less computational cost and can be used to optimize the water content of GDL.