An Innovative Approach to Design and Optimize Flow Field Pattern of Polymer Electrolyte Membrane Fuel Cell
Takayuki Tsukamoto,Tsutomu Takayama,Yuhei Miyamoto,Keisuke Komiyama,Masakazu Yoneda
DOI: https://doi.org/10.1149/ma2022-02401453mtgabs
2022-11-24
ECS Meeting Abstracts
Abstract:To increase power density of polymer electrolyte membrane fuel cells (PEMFCs), further optimization and downsizing of cell and stack design are necessary. To support cell and stack design, we have been developing a three-dimensional (3D) PEMFC simulator P-Stack [1]. P-Stack can estimate power generation performance in good agreement with experimental results, and can clarify relationships between the performance and internal states depending on cell designs, operation conditions, and material properties [1]. Although 3D PEMFC simulations are powerful means, preparing 3D model is very time consuming, thus it was difficult to explore various cell designs. In the present study, we have developed a new modeling technology for PEMFC flow fields. In this technology (Figure (a)), flow fields of anode and cathode gases are drawn by lines in 2D canvas. Properties of the lines such as width, depth and wall angle are given, then it automatically generate 3D model for anode/cathode flow fields and bipolar plates, coolant flow field, seal, gasket and MEA. Power generation performance of the 3D model can be estimated by P-Stack continuously. In addition, the line properties, vertex positions, and the number of repeat can be parametrized, enabling the parametric study of cell design. In this technology, pressure and flow rate distributions can be calculated by a network model based on Darcy's law [1], without generating 3D model. As shown in Figure (b), the result is in good agreement with computational fluid dynamics (CFD) calculation of 3D model. Although the CFD calculation takes several hours using 32 parallel processes, the calculation of the network model is finished within several seconds using single process. In similar manner, it can calculate flow rate distribution of cells in the stacking axis. Since the execution time per one case is very short, it can be combined with optimization by genetic algorithm (GA). In the presentation, we will talk about a result of the parametric study and optimization for more practical cell design. Influence of flow field design on power generation performance and internal states will be discussed. References [1] T. Tsukamoto, T. Aoki, H. Kanesaka, T. Taniguchi, T. Takayama, H. Motegi, R. Takayama, S. Tanaka, K. Komiyama, M. Yoneda, J. Power Sources 488 (2021) 229412 Figure 1