摘要
针对滚动轴承的故障振动信号的非平稳特性,提出了一种基于局部均值分解(Local mean decomposition,简称LMD)和神经网络的滚动轴承诊断方法。该方法首先对信号进行局部均值分解,将其分解为若干个PF分量(Product function,简称PF)之和,再选取包含主要故障信息的PF分量进行进一步分析,从这些分量中提取时域统计量和能量等特征参数作为神经网络的输入参数来识别滚动轴承的故障类别。通过对滚动轴承正常状态,内圈故障和外圈故障的分析,表明了基于LMD与神经网络的诊断方法比基于小波包分析与神经网络的诊断方法有更高的故障识别率,同时也证明了该方法可以准确、有效地对滚动轴承的工作状态和故障类型进行分类。
A roller bearing fault diagnosis method based on local mean decomposition (LMD) and neural network was proposed to deal with the non-stationary vibration signal from fault roller bearings.First of all,LMD method was applied to decompose the original signals into a finite number of product functions (PFs),then several PFs containing main fault information were selected for further analysis;subsequently,energy and time domain statistic feature parameters extracted from PFs were served as input parameters of neural network to identify fault patterns of roller bearing.The analysis results from roller bearing signals with inner race and outer race faults show that the diagnosis method based on LMD and neural network can identify roller bearing fault patterns accurately and effectively and is superior to that based on wavelet packet analysis and neural network.
出处
《振动与冲击》
EI
CSCD
北大核心
2010年第8期141-144,共4页
Journal of Vibration and Shock
基金
国家自然科学基金资助项目(50775068)
湖南省博士后科学基金(2008RS4004)
关键词
滚动轴承
LMD
神经网络
故障诊断
特征参数
roller bearing
LMD
neural network
fault diagnosis
feature parameter
