CEEMD与广义形态差值滤波结合的故障诊断方法研究

为了提取滚动轴承早期微弱故障特征信息，提出一种互补总体平均经验模态分解(Complementary Ensemble Empirical Mode Decomposition，CEEMD)与广义形态差值滤波结合的故障诊断方法.该方法首先对振动信号进行CEEMD分解成若干不同尺度的本征模函数(Intrinsic Mode Function，IMF)分量，利用相关系数-峭度准则来选取故障信息丰富的IMF分量信号，并对其进行重构；然后采用广义形态差值滤波器对重构后的信号进行滤波，以滤除噪声干扰；最后利用Teager能量算子(Teager-Kaiser Energy Operator，TKEO)对去噪后的振动信号进行分析，提取振动信号的故障特征.滚动轴承振动信号分析试验结果证明了本文方法的有效性.

Research on fault diagnosis method combining CEEMD with generalized morphology difference filter

In order to extract the early fault feature of rolling bearing， a method based on the combination of Complementary Ensemble Empirical Mode Decomposition (CEEMD) and generalized morphological difference filter for fault diagnosis is proposed in this paper. Firstly， the vibration signals are decomposed by the CEEMD into different scales of IMF component signals， and the IMF component signals with rich fault information are reconstructed by the correlation coefficient and kurtosis criterion. Then the reconstructed signals are filtered by the generalized morphological difference filter to filter the noise. Finally， characteristics of signals are extracted from the vibration signal which filtered signals using Teager-Kaiser Energy Operator (TKEO). The experiment results have shown that the proposed method applied in the rolling bearings fault detection is effective.