Remaining Useful Life Prediction Method of PEM Fuel Cells Based on a Hybrid Model

To predict the remaining useful life (RUL) of proton exchange membrane fuel cell (PEMFC) in advance, a prediction method based on voltage recovery model and Bayesian optimization multi-kernel relevance vector machine (MK-RVM) is proposed this paper. First, empirical mode decomposition (EMD) was used to preprocess data, then MK-RVM train model. Next, algorithm optimize weight coefficient kernel function complete parameter update model, added realize rapid accurate RUL PEMFC. Finally, paper applied open data set PEMFC provided by Fuel Cell Laboratory (FCLAB), accuracy for obtained 95.35%, indicating that had good generalization ability verified when predicting The realization long-term projections not only improves life, reliability, safety but also reduces operating costs downtime.