Simulation of the proton-exchange membrane (PEM) fuel cell life-cycle performance with data-driven parameter estimation

Degradation parameters are included in a computational model to simulate the fuel cell output voltage as a function of time, with these degradation parameters optimized using experimental data. The contributions of the loss of catalytic activity and membrane/electrode conductivity to the fuel cell output voltage are represented through the degradation parameters. This is achieved without having to include detailed electrochemical mechanisms or their rate data but, instead, using the data on global fuel cell voltage as a function of time. Using parameter estimation based on output data, complex degradation effects aresummed into a small number of parametric equations, so that the overall behavior can be reproduced. The simulation results are in excellent agreement with the data and also point to the main mechanisms of degradation. Also, it is shown that at least two parameters are needed, one for the loss of catalytic activity and another one for the ohmic losses, to faithfully simulate the loss of fuel cell performance across different current densities over time. This approach can be extended to predict the lifetime performance of fuel cells in general, if there is a minimum amount of data to apply the parameter estimation algorithm for the determination of a small number of degradation parameters.