My Manuscript

Status Accepted
journal_subject Part A
Article Type Regular Paper (More than 4 pages)
Article Filed Maintenance Engineering
Article Title Research on Bearing Fault Diagnosis Method Based on Singular Spectrum Decomposition
Keywords singular spectrum decomposition; instantaneous frequency extraction; experience am-fm decomposition;
Abstract Aiming at the problem that the fault characteristics of rolling bearings are difficult to detect under the influence of strong background noise and interference signal, a new adaptive signal processing method based on singular spectrum decomposition (SSD) is proposed. This method can divide non-stationary signal into several single-component signals from high frequency to low frequency by constructing a trajectory matrix, and adaptively select embedding dimension length. The singular spectral components with obvious impact characteristics and maximum kurtosis are selected according to the kurtosis criterion. The singular spectral components are separated into envelope components and pure frequency components with the empirical AM-FM decomposition method, and the instantaneous frequencies are calculated by using energy operators. To solve the problem of endpoint effect in the instantaneous frequency, a data extension method using support vector regression is proposed. The experimental results show that the instantaneous frequency extraction can be better applied to the vibration signal of rotating machinery in the background of strong noise, and the instantaneous characteristics can be extracted.
First Name Middle Name Last Name E-Mail Corresponding
Xiong Meng No
Su Lei No
Li Ke Yes
Upload Files Please Login to View Upload Files
Copyright ® 2011 ijce. All rights reserved