Generator is an important core component in wind power system,in order to improve the stable and efficient operation of wind turbine,the fault prediction of wind turbine generator is necessary.Focusing on the problem of generator machine-side bearing temperature overrun fault prediction in wind power system,this paper takes into account that the collected fault characteristic signal is characterized by large noise,introduces CEEMDAN joint adaptive wavelet threshold denoising method to realize effective denoising of the signal,and at the same time establishes a fault prediction model by combining GA-BP neural network.By comparing the prediction indexes,error indexes and prediction effect graphs with BP neural network and GA-BP neural network,it is verified that the proposed algorithm can obtain better prediction effect.The error index and prediction effect are improved,and the accuracy of the prediction of generator failure of wind power system 15 days in advance reaches 92.98%.
wind energy systemgenerator failurefault predictionCEEMDANGA-BP neural network
国家科技期刊平台
NSTL
万方数据
