Fault Diagnosis of Rotating Machinery Based on Multi-Sensor Signals and Median Filter Second-Order Blind Identification (MF-SOBI)

Abstract

Feature extraction plays a crucial role in the diagnosis of rotating machinery faults. However, the vibration signals measured are inherently complex and non-stationary and the features of faulty signals are often submerged by noise. The principle and method of blind source separation are introduced, and we point out that the blind source separation algorithm is invalid in an environment of strong impulse noise. In order to solve the problem of fast separation of multi-sensor signals in an environment of strong impulse noise, first, the window width of the median filter (MF) is calculated according to the sampling frequency, so that the impulse noise and part of the white noise can be effectively filtered out. Next, the filtered signals are separated by the improved second-order blind identification (SOBI) algorithm. At the same time, the method is tested on the strong pulse background noise and rub impact dataset. The results show that this method has higher efficiency and accuracy than the direct separation method. It is possible to apply the method to real-time signal analysis due to its speed and efficiency.

Cyclically Shifted Extreme-point Symmetric Mode Decomposition (CS-ESMD)-based Progressive Denoising Approach for Ground-based Synthetic Aperture Radar Bridge Health Monitoring Signals

“…2, can use a small amount of prior information data to process the observed mixed signal by using the joint approximate diagonalization of the source signal time correlation and the covariance matrix to estimate the source signals. (23,24) It can estimate the number of source signals and separate Gaussian white noise. This algorithm does not have the problem of modal aliasing and is a robust algorithm.…”

A new fault feature extraction method of rotating machinery based on finite sample function

“…. Without considering noise, the mathematical model of the observation signal can be expressed as 6,8…”

“…4,5 BSS 6,7 refers to the method of recovering the original signal from the observed signal only depending on some basic statistical characteristics of the source signal when the characteristics of the source signal and the transmission channel are unknown. There have been many effective BSS algorithms with different characteristics, including the fast fixed-point algorithm, 8,9 natural gradient algorithm, Equivariant Adaptive Separation via Independence (EASI) algorithm 10,11 equivariant adaptive separation by algorithm, Joint Approximation Diagonalization (JAD) algorithm 12 and algorithm. 13 These have been widely used for fault diagnosis, 14 image processing, 15 speech recognition, 16 earthquake prediction, 17 and other fields.…”

Bridge Damage Identification by Ground-based Synthetic Aperture Radar Using Blind Source Separation and Noise Reduction Technology

“…Next, the covariance matrix of the albino signal with delay p is calculated, and the joint approximate diagonalization algorithm is used to diagonalize the covariance matrix of different time delays. (20) Then, an orthogonal matrix is obtained. Finally, the best estimations of the mixed matrix and source signal are calculated accordingly from the orthogonal and albino matrices.…”