PEMFC aging modeling for prognostics and health assessment

摘要

When a system suffers from a too short lifetime, applying prognostics is a good solution to help taking actions extending its life duration. This solution is applied to Proton Exchange Membrane Fuel Cell (PEMFC) stacks in this paper. An important requirement for prognostics of a PEMFC stack is a well-defined framework as well as a great understanding of the degradation mechanisms and failures occurring within the stack. These requirements are addressed here and allow building an efficient model integrating the different levels (stack - cells - components) as well as the multiple causes leading to degradation. Such a model enables then health assessment and remaining useful life predictions. This work proposes a model built based on a selection of critical degradations and to validate it for both state of health estimations and prognostics. The results show that the stack's state of health during aging can be followed accurately with coefficients of correlation greater than 0.9. Also, the behavior of the system can be assessed with a coefficient of correlation greater than 0.9 showing the great predictive capabilities of the model.