Comparison of Degradation Prediction Methods for Proton Exchange Membrane Fuel Cell

Proton exchange membrane fuel cell (PEMFC) is a kind of green energy converter, which has a very good future. As one of the constraints for commercialization of PEMFC, the durability problem is a hot research topic in the field of fuel cell. Prognostics and health management (PHM) is a powerful way to prolong the life of PEMFC, and degradation trend prediction is the key of this technology. Based on the working mechanism of fuel cells, the short-term prediction methods (LSTM and open-loop NARX model) and long-term prediction methods (closed-loop NARX model, PF and BPNN) of PEMFC degradation tendency are discussed respectively. By applying different methods to constant condition, dynamic condition and combined condition respectively, and comparing the effect of degradation trend prediction, the application conditions, advantages and disadvantages of short-term prediction method and long-term prediction method are analyzed.