AI-Assisted Flow Field Design for Proton Exchange Membrane Fuel Cells: Progress and Perspective

Tongxi Zheng,Fanyu Meng,Wenxuan Fan,Mingxin Liu,Dafeng Lu,Yang Luan,Xunkang Su,Guolong Lu,Zhenning Liu
DOI: https://doi.org/10.1007/s42235-024-00607-2
IF: 2.995
2024-11-27
Journal of Bionic Engineering
Abstract:Bipolar plate is one of the key components of Proton Exchange Membrane Fuel Cell (PEMFC) and a reasonable flow field design for bipolar plate will improve cell performance. Herein, we have reviewed conventional and bionic flow field designs in recent literature with a focus on bionic flow fields. In particular, the bionic flow fields are summarized into two types: plant-inspired and animal-inspired. The conventional methodology for flow field design takes more time to find the optimum since it is based on experience and trial-and-error methods. In recent years, machine learning has been used to optimize flow field structures of bipolar plates owing to the advantages of excellent prediction and optimization capability. Artificial Intelligence (AI)-assisted flow field design has been summarized into two categories in this review: single-objective optimization and multi-objective optimization. Furthermore, a Threats-Opportunities-Weaknesses-Strengths (TOWS) analysis has been conducted for AI-assisted flow field design. It has been envisioned that AI can afford a powerful tool to solve the complex problem of bionic flow field design and significantly enhance the performance of PEMFC with bionic flow fields.
robotics,engineering, multidisciplinary,materials science, biomaterials
What problem does this paper attempt to address?