基于改进HVD和包络谱的轴承故障诊断方法

Bearing Fault Diagnosis Method Based on Improved HVD and Envelope Spectrum

轴承是机械设备的重要零部件之一。希尔伯特振动分解与经验模态分解同样存在着端点效应,针对于端点效应提出了镜像延拓的改进方法,该方法通过在信号左右两端分别延拓一定的数据长度,信号分解后再截去左右延拓的数据。较之传统的希尔伯特振动分解方法,该方法能有效的抑制分离出的分量两端发生发散的现象,将改进的HVD与包络谱结合能够有效的应用于轴承故障诊断,能够有效地提取出轴承故障特征频率。

Bearings are one of the important components of mechanical equipment.Hilbert's vibration decomposition and empirical mode decomposition also have end-point effects.An improved method of image extension is proposed for the end effect.This method extends the data length at the left and right ends of the signal,and the signal is decomposed.Cut off the data of the left and right extensions.Compared with the traditional Hilbert vibration decomposition method,this method can effectively suppress the divergence of the separated components at both ends,and the improved HVD and envelope spectrum can be effectively applied to bearing fault diagnosis,which can effectively the bearing fault characteristic frequency is extracted.