Abstract
Fuel Cell (FC) systems are promising power-generation sources that are more and more presented as a good alternative to current energy converters such as internal combustion engines. They suffer however from insufficient durability for stationary and transport applications, and lifetime may be improved. A greater understanding of underlying wearing processes is needed in order to improve this technology. However, FCs are in essence multi-physics and multi-scales systems (from the cells to the whole power system), which makes a modeling step of behaviors and degradation very difficult, even impossible. Thereby, data-driven Prognostic and Health Management (PHM) principles (as defined in condition-based- maintenance scheme CBM) appear to be of great interest to face with the problems of health assessment and life prediction of FCs. According to all this, the aim of this paper is to present the current state of the art on PHM for FCs. Developments emphasize on PHM of the Proton-Exchange Membrane Fuel Cells (PEMFC) stack. The paper is organized so that important aspects like "behavior and losses FCs", "observation techniques", and "advanced PHM techniques" are addressed. Also, a taxonomy of existing works on PHM of PEMFC is given accordingly to the processing layers of CBM. The whole enables PHM practitioners as well as FCs experts to get a better understanding of remaining challenging issues.
