Parametric sensitivity analysis of a PEMFC physics-based model developed for prognostics

In order to face one of the bolt for Proton Exchange Membrane Fuel Cell, PEMFC's industrial breakthrough, the development of Prognostics and Health Management (PHM) is a promising option. Indeed, the prognostics part of the PHM whose goal is to estimate the Remaining Useful Life (RUL) of the system in order to furnish the decision part of PHM the information needed for select the good actions has not been really developed for PEMFC. It is then the next step that need development and that would allow to optimize the use of the fuel cell and then extend it's too short life duration. For that purpose, the development of a behavioral model is necessary as it would be able to predict the dynamic of the fuel cell. A model of this kind is developed here, it is composed of a static and a dynamic part. The ageing in this model can be included by the parameters in each parts. However it is necessary to analyze which parameters is critical for the development of the ageing, as the number of parameter is high face to the data available. The discussion of the parametric sensitivity of the model is presented here as well as the validation of the model.