Fault Diagnosis of Rolling Bearing Based on Local Mean Decomposition and Transient Extracting Transform

To solve the problem of inconspicuous feature extraction when LMD method is used to extract rolling bearing fault characteristic signals, A fault feature extraction method based on Local Mean Decomposition (LMD) and Transient- Extracting Transform (TET) was proposed. Firstly, the rolling bearing fault signals were processed by LMD and the feature components with rich fault information were screened out by using the kurtosis values. Then, the secondary feature extraction and envelope analysis of the acquired components were carried out by using TET method. The experimental results showing that this method can extract the pulse characteristics of the impact signal of rolling bearings efficiently, and is suitable for the fault diagnosis of rolling bearing.