Data-fusion Prognostics of Proton Exchange Membrane Fuel Cell Degradation

Proton exchange membrane fuel cell (PEMFC) degradation prediction is essential especially in transportation applications, since one of the major issues that hinder its worldwide commercialization is represented by its durability. However, due to the complex physical phenomena inside the fuel cell, which are usually strongly inter-coupled, the conventional semi-empirical model-based prognostics approach may fail to predict the aging phenomena under varies fuel cell operating conditions. In order to improve prognostics accuracy, this paper proposed a data-fusion approach to forecast the fuel cell performance based on long short-term memory (LSTM) recurrent neuron network (RNN) and auto-regressive integrated moving average (ARIMA) method. LSTM can efficiently make a prediction regarding a long-term physical degradation, whereas the fusion with ARIMA can effectively track the degradation tendency. In order to validate the performance of the proposed data-fusion approach, two different PEMFCs are tested for recording the aging experimental datasets. The forecasting results indicate that the proposed LSTM-ARIMA approach can accurately predict PEMFC degradation, which can be then used directly to optimize fuel cell performance implemented in transportation applications.