论文部分内容阅读
提出了局部均值分解(Local mean decomposition,简称LMD)方法和径向基函数神经网络(Radial Basis Function Neural Network,简称RBF)相结合的滚动轴承故障诊断方法。LMD方法是一种新的自适应时频分析方法,能够有效地提取故障特征。该方法首先采用LMD对滚动轴承振动信号进行分解,计算分解得到的PF分量能量比,作为特征向量输入到RBF神经网络中,进行故障分类和识别。通过真实滚动轴承数据的故障诊断实验,验证了该方法的有效性。
A rolling bearing fault diagnosis method based on local average decomposition (LMD) method and radial basis function neural network (RBF) is proposed. LMD method is a new adaptive time-frequency analysis method, which can effectively extract fault features. In this method, the vibration signal of rolling bearing is decomposed by using LMD. The energy ratio of PF component obtained by decomposition is calculated and input into the RBF neural network as the feature vector for fault classification and identification. The fault diagnosis experiment of real rolling bearing data verifies the effectiveness of the method.