Long-Term Forecasting of a Degradation Indicator for Proton Exchange Membrane Fuel Cells

This paper presents a methodology to predict a long-term degradation indicator for Proton Exchange Membrane Fuel Cells (PEMFC). The proposed approach uses a hybrid-based fuel cell degradation model to identify the degradation indicator. The proposed work employs Echo State Networks (ESNs) to predict the degradation indicator. The first stage of the prediction process focuses on tuning the parameters of the ESN model, ensuring good performance. Subsequently, a long-term prediction is performed to predict the previously calculated degradation indicator.