关键词: 小波分析, 神经网络, 故障诊断

Abstract: This paper focuses on the system structure and algorithms of Wavelet Neural Network.For various features of gear vibrating signals in frequency domain,feature vectors are extracted as inputs of the Wavelet Neural Network which are capable of mapping the feature vectors to the corresponding fault modes.Based on this algorithm,a gear fault diagnosis model has been designed.Simulation results indicate this model,compared with the conventional BP neural network model,can remarkably reduce the training time.It is feasible for mechanical fault diagnosis.

Key words: wavelet analysis, neural network, fault diagnosis