Multi-Objective Optimization of a Tapered Cathode Flow Channel in a Proton Exchange Membrane Fuel Cell

This study explores the design of a tapered cathode flow channel in a proton exchange membrane fuel cell (PEMFC), leveraging artificial intelligence and multi-objective optimization techniques to attain an optimal configuration. First, the influence of the channel height ratio and mass flow rate on PEMFC performance was systematically examined. The results reveal that decreasing the height ratio and increasing the mass flow rate lead to reduction in the standard deviation of current density, accompanied by a monotonic rise in pressure drop. The average current density initially rises before exhibiting a slight decline. Subsequently, a surrogate model based on a Backpropagation (BP) neural network was constructed, with height ratio and mass flow rate as input variables, to accurately predict the average current density, its standard deviation, and the channel pressure drop. The findings demonstrate that the BP-based surrogate model can reliably predict current density, its standard deviation, and channel pressure drop. The Mean Relative Errors (MREs) for current density, standard deviation, and pressure drop are 0.84%, 1.44%, and 1.77%, respectively, with all coefficients of determination (R²) exceeding 0.999. Finally, Pareto optimal solutions for current density, standard deviation, and pressure drop of the tapered PEMFC were obtained through integration a multi-objective genetic algorithm. Results show that the optimized tapered PEMFC achieves the current density of 3141.41 A/m², the standard deviation of 53.58 A/m², and the channel pressure drop of 5.49 Pa. Compared with the conventional channel, the optimized PEMFC exhibits an 7.02% increase in current density and an 3.7% reduction in standard deviation, while maintaining the pressure drop within an acceptable range.