The optimization of an innovative interdigitated flow field proton exchange membrane fuel cell by using artificial intelligence

Abstract

A suitable flow field is vital for the efficient performance of proton exchange membrane fuel cells. This study proposes an innovative interdigitated flow field configuration to enhance mass transfer, resulting in improved performance. A wave pattern is repeated in the flow field, and the flow field is called the wave interdigitated flow field. A three-dimensional, multiphase CFD model is employed to simulate the fuel cell. The wave interdigitated flow field is compared with serpentine, simple interdigitated, tapered interdigitated, and spiral interdigitated flow fields. Mass transfer, reactant distribution, and water management are investigated. The wave interdigitated has the most convective and diffusive transfer of reactants, resulting in the highest reaction rate. Its power density is 27.3 %, 19.7 %, 17.5 %, and 14.2 % higher than that of the spiral interdigitated, tapered interdigitated, simple interdigitated, and serpentine flow fields. It brings the most uniform water saturation, reducing the possibility of flooding. Saturated reactants can increase the generated current and exacerbate water saturation. Water saturation and high current density were selected as objectives. A code was developed to generate and optimize a surrogate model based on artificial intelligence. The optimum conditions result in a water saturation of 0.015 and a power density of 3 W cm−2.