Résumé

Internal states knowledge of a proton exchange membrane fuel cell (PEMFC) is essential to enhance its performance level and further extend its lifespan. Real-time membrane water content estimation is crucial to prevent PEMFC from degrading conditions by avoiding drying or flooding. Estimating PEMFC humidity state can also contribute to establishing fault-tolerant control strategies. In this paper, a steady-state membrane water content estimation tool and a lumped cell voltage model have been developed. The cell voltage model has been validated using automotive fuel cell experimental data and compared with the outputs of a commercial simulation tool. The cell voltage value built on membrane water content estimation is sensitive to combined humidity and temperature variations, as validated by experimental data. This work paves the way for developing an online membrane water content observer for fault-tolerant control applications.