Prediction of PEMFC Performance via a One-Dimensional Model Considering the Effect of Mechanical Stress

A one-dimensional proton exchange membrane fuel cell (PEMFC) model was developed to consider the effect of mechanical stress on the transport parameters of each layer inside a cell. A genetic optimization algorithm was integrated to rapidly calibrate key parameters under various operating conditions. Sensitivity of the model parameters was systematically evaluated via Sobol global sensitivity analysis. The main conclusions are: the cell performance under different operating conditions can be obtained quickly and accurately by the developed model, and the maximum relative and absolute errors are controlled within 1.2% and 10 mV, respectively. The compression ratio of the membrane electrode assembly due to mechanical stress is a sensitive parameter. Determining the influence of mechanical stress can more accurately describe the transport phenomenon in the cell and improve the model’s prediction accuracy. By integrating genetic optimization algorithms, the optimal combination of model parameters can be efficiently determined. Moreover, the model exhibits certain blind calibration capability, maintaining absolute calibration errors within 25 mV and relative errors within 5%. This attribute suggests that the model can provide powerful support for design optimization of real cell systems, optimization of operation and management control strategies, and rapid assessment of the performance of mass-produced cell stacks.