Optimization design of trapezoidal flow field proton exchange membrane fuel cell combined with computational fluid dynamics, surrogate model, and multi-objective optimization algorithm

The flow field structure plays the key roles in the operating reliability and power output of proton exchange membrane fuel cell (PEMFC). This study investigates and compares two typical structural parameters of PEMFC with the trapezoidal flow channel (TFC) and the trapezoidal flow channel with block (TFCB). A three-dimensional (3-D) multiphase TFC model is first developed, and then the multi-objective optimization is performed by using the trained artificial neural network (ANN) surrogate model and non-dominated sorting genetic algorithm (NSGA-II). Finally, the technique for order preference by similarity to an ideal solution (TOPSIS) is used to investigate the optimized structural parameters of TFC. The results show the net power output and the oxygen uniformity index of the optimized TFC are increased by 19.77% and 21.92% compared with the straight flow channel (SFC). Furthermore, it is also found using block in the trapezoidal flow channel (TFCB) can increase the performance of PEMFC, and it exhibits a 25.10% improvement for the net power output and 27.88% for oxygen uniformity index, respectively.