Health state prognostic of fuel cell based on wavelet neural network and cuckoo search algorithm

This paper proposes a novel degradation prognosis of Proton Exchange Membrane Fuel Cell (PEMFC) based on Wavelet Neural Network (WNN) and Cuckoo Search Algorithm (CSA). The proposed method considering the main operating conditions PEMFC can be applied to health state prognostic under different conditions. First, data are reconstructed by locally weighted scatterplot smoothing filter noise. Then, WNN that analyze characteristics (global trend reversible phenomena) is adopted build model PEMFC. Finally, structure parameters optimized CSA improve accuracy for used predict remaining useful life validated 3 tests results show greatly WNN. CSA-WNN achieve higher precision than other traditional methods.