A Mathematical Model toward Real-Time Monitoring of Automotive PEM Fuel Cells

A computationally efficient model toward real-time monitoring of automotive polymer electrolyte membrane (PEM) fuel cell stacks is developed. Computational efficiency is achieved by spatio-temporal decoupling of the problem, developing a new reduced-order model for water balance across the membrane electrode assembly (MEA), and defining a new variable for cathode catalyst utilization that captures the trade-off between proton and mass transport limitations without additional computational cost. Together, these considerations result in the model calculations to be carried out more than an order of magnitude faster than real time. Moreover, a new iterative scheme allows for simulation of counter-flow operation and makes the model flexible for different flow configurations. The proposed model is validated with a wide range of experimental performance measurements from two different fuel cells. Finally, simulation case studies are presented to demonstrate the prediction capabilities of the model.