Study on Information Fusion Based on Wavelet Neural Network and Evidence Theory in Fault Diagnosis

Aimed at the low diagnosis accuracy of traditional fault diagnosis algorithm, information fusion algorithm based on wavelet neural network and evidence theory is proposed. In the initial diagnosis stage, wavelet neural network is used to improve the fault diagnosis ability of the local diagnosis networks. In the decision diagnosis stage, information fusion frame based on wavelet neural network and evidence theory is constructed to improve the accuracy of fault diagnosis by virtue of various redundant and complementary fault information. Simulation experiment results show the efficiency of the algorithm.