A hybrid prognostic model applied to SOFC prognostics

Poor durability is one of the major hurdles for Solid Oxide Fuel Cell (SOFC) commercialization. Prognostic technology is an important approach to improve system durability. However, the current prognostic methods for fuel cells are mainly applied to proton exchange membrane fuel cell (PEMFC) systems, moreover, there are still shortages of recent prognostic algorithms. Therefore, an improved prognostic model is developed to predict the remaining useful life of the SOFC in this study, which combines a hidden semi-Mark model (HSMM) with an empirical model. To build the hybrid prognostic model, an HSMM and an empirical model are firstly built. The merit and demerit of the respective prognostic methodology are then analyzed. According to the analysis results, the hybrid prognostic model is proposed and applied to six sets of SOFC run-to-failure data. The results show the proposed prognostic model has a higher prediction accuracy and a faster forecasting speed compared with the existed approaches for SOFC prognostics.