奇异值分解与LMD结合的滚动轴承故障诊断研究

Research on Fault Diagnosis of Rolling Bearing Based on Singular Value Decomposition and LMD

针对滚动轴承故障信号具有的非线性和非平稳性,其故障特征难以提取的问题,提出一种奇异值分解(SVD)和局部均值分解(LMD)相结合的滚动轴承故障特征提取和诊断方法。首先,将轴承故障信号进行LMD分解得到若干PF分量;然后选取和原始信号相关度较大的PF分量,利用奇异值序列来构造其故障特征向量;最后,将得到的故障特征向量作为学习样本输入到支持向量机(SVM)中,对故障类型进行分类和识别。实验结果表明,LMD和SVD结合的故障特征提取方法,能有效提取滚动轴承不同状态下的故障特征,对不同故障状态做出准确分类。

Due to nonlinear and non-stationary of rolling bearing fault signals,the fault feature is difficult to extract. A novel method of fault feature extraction and diagno sis of rolling bearing based on the combination of singular value decomposition（SVD）and local mean decomposition（LMD）is proposed. First,decompose the bearing fault signal by LMD to obtain several PF components;Then the PF component with larger correlation degree of the original signal is selected,using the singular value sequence to construct the fault feature vector;Finally,the fault feature vectors are used as input to the support vector machine（SVM）to classify and identify the fault types. Experimental results show that the fault feature extraction method based on the combination of SVD and LMD,can effectively extract fault features of rolling bearings under different conditions,to make accurate classification of different fault status.