基于神经网络的自行火炮变速箱轴承故障模式识别

Fault Identification of a rolling Element Bearing in Gearbox of selfpropelled gun Based on Neural Network

在分析振动信号的基础上，提取了反映自行火炮变速箱轴承故障的时域特征量和包络谱特征量．构造了结构合适的BP神经网络，实现了自行火炮变速箱轴承故障模式的计算机自动识别，有效地检测了自行火炮变速箱轴承的早期故障．

On the basis of analysing vibration signal,both time domain and envelope spectrum domain diagnostic parameters are gained.A reasonable BP neural network which is used to identify rolling element bearing fault pattern is composed.The validity of the BP neural network to detect initial fault of rolling elememt bearing is confirmd by experiments.