一种基于小波和神经网络的故障诊断系统

Fault Diagnosis System Based on Wavelet and Neural Network

基于小波变换的时频域局部化特征及神经网络的非线性映射特征,以滚动轴承为例,将小波变换和神经网络的优点结合起来。运用小波变换提取滚动轴承振动信号各频率成分的能量作为故障特征参数,将其作为神经网络的输入进行训练和故障识别,利用BP网络实现了对滚动轴承的故障诊断,取得了较好的效果。

Based on features of time frequency localization of wavelet transformation and nonlinear mapping of neural network, taking rolling bearing as an example, advantages of wavelet transformation are combined with advantages of neural network. Taking each frequency energy of rolling bearing vibration signal as fault characteristics parameters by means of wavelet transformation, and the parameters are taken as input data of neural network to carry out train and fault identification. By use of BP network, fault diagnosis for rolling bearing is performed and good result is acquired.