Spatially distributed membrane degradation in PEM fuel cells — A systematic approach for long-term simulations

Abstract

Modeling local degradation phenomena in proton exchange membrane fuel cells is essential for understanding and monitoring cell aging. Assessing these local effects via simulation usually requires high-fidelity, spatially distributed models, which are computationally expensive. This paper presents a simulation framework allowing for efficient long-term simulations while keeping a high resolution of spatially resolved effects within the fuel cell, incorporating dynamic operation and local degradation. A nonlinear, physically motivated quasi-2D fuel cell model is coupled with a model for chemical ionomer degradation, allowing for assessing the effect of local degradation effects on overall cell health and performance. To achieve fast and reliable long-term simulations, a network of local linearized models is derived and examined. The proposed approach reduces simulation time significantly while preserving the key dynamic behavior of both operational states and degradation effects, despite differing time scales. The results demonstrate that the numerical demand of models describing distributed fuel cell degradation can be effectively reduced without compromising fidelity, offering a valuable tool for long-term degradation assessment, optimization of operating strategies, and cell monitoring over the full lifetime under varying and dynamic operating conditions.