Optimization of blocked channel design for a proton exchange membrane fuel cell by coupled genetic algorithm and three-dimensional CFD modeling

摘要

Installing blocks in cathode flow field can effectively enhance the transfer of oxygen from channel to the reaction sites of catalyst layer, thus boosting the performance of the fuel cell. In this work, an optimization methodology combined with genetic algorithm and three-dimensional fuel cell modeling is developed to optimize the design of partially blocked channel for a proton exchange membrane fuel cell (PEMFC) with parallel flow field. In the optimization, the heights of the blocks are assumed to be linearly increased and two parameters (i.e., height of the first block and the height increase between adjacent blocks) are considered. The impact of the optimized design of the blocked channel on cell performance is analyzed, and the effects of the optimized blocked channel designs with increasing-height and uniform-height block height distributions were also compared in detail. With this optimization methodology, the optimal height distribution of the blocks in the channel can be obtained for various block numbers. With varying the block numbers, the cell voltage and net cell power are firstly improved until the maximal values reached and then lowered. The maximal net cell power is reached for the block number of 16. As compared with the flow channel without adding blocks, the net power of the PEMFC can be enhanced by about 10.9%. For pressure drop behavior, with the optimized block height distribution, the total pressure drop in cathode flow field can be maintained in similar level with varying block numbers from 4 to 20. Considering both the net power and pressure drop, the optimized blocked channels with adding 8 to 16 blocks are recommended in this study. Besides, it is indicated that the performance of the optimized block design with increasing-height is higher than that of the optimized block design with uniform-height. (C) 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.