Fuel cell fault forecasting system using grey and extension theories

This study proposes a fault forecasting system for proton exchange membrane fuel cells (PEMFCs), which uses a set of wireless sensors to accomplish fuel cell (FC) condition monitoring. The software interface of the FC fault forecast system uses LabVIEW software. Owing to the time delay condition of FC reactions, this study first forecasts all the operation features of the FC using a Grey prediction model. Then, an extension diagnosis method uses the forecasted values of the features to forecast the future operational conditions. Thus, the complex condition monitoring and fault forecasting problem of the FC can be implemented effectively. To demonstrate the effectiveness of the proposed method, comparative studies using a multilayer neural network and k-means algorithm are conducted on 400 sets of field-test patterns of 200 W PEMFC with rather encouraging results.