基于SVD和MED的滚动轴承特征提取

Feature Extraction of Rolling Bearing Based on SVD and MED

针对滚动轴承振动信号易受噪声影响,难以提取故障特征信息的问题,提出一种奇异值分解(singular value decomposition,SVD)重构结合最小熵反卷积(minimum entropy deconvolution,MED)增强的滚动轴承故障特征提取方法。首先,对振动信号进行SVD分解,并计算奇异分量(singular component,SC)对应线性峭度(L-kurtosis)值;其次,根据线性峭度值结合设定阈值筛选SC,叠加得到重构信号;随后,对重构信号利用MED进行增强,凸出信号中周期冲击成分;最后,结合包络解调提取故障特征频率。仿真信号及实测信号分析结果表明,该方法可以降低噪声对振动信号的影响且凸显故障的特征信息,实现故障诊断。

For the problem that the vibration signal of rolling bearing is easily affected by noise and it is difficult to extract fault feature information,a bearing fault diagnosis method based on singular value decomposition(SVD)and reconstruction combined with minimum entropy deconvolution(MED)enhancement is proposed.Firstly,the original signal is decomposed by SVD,and the linear kurtosis corresponding to the singular component is calculated.Secondly,the singular component(SC)are selected according to the linear kurtosis with the set threshold,superimposing to get the reconstructed signal.Thirdly,the reconstructed signal is enhanced by MED to protrude the periodic shock components in the signal.Finally,the fault characteristic frequency is extracted by envelope demodulation.The results of both simulated signal and measured signal show that this method can reduce the influence of noise on vibration signal,highlight fault characteristic information,and realize fault diagnosis.