Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Feature extraction method based on single-channel signal blind-separation rolling bearing

A rolling bearing and feature extraction technology, which is applied in the field of feature extraction based on single-channel signal blind separation of rolling bearings, can solve problems such as poor immediacy, very different wavelet-based separation effects, and long calculation time.

Inactive Publication Date: 2015-11-04
CHANGAN UNIV
View PDF6 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] For underdetermined blind separation, that is, the number of sensors is less than the number of sources, the main method is the median-based clustering algorithm blind separation method [1] , Underdetermined blind separation method based on potential function [2] , these existing problems are based on the sparseness of the source signal, and the separation effect is not good for signals with poor sparsity; there is also blind signal separation based on wavelet decomposition [3] , the problem it has is that it is highly dependent on the selection of wavelet bases, and the separation effect of different wavelet bases is very different; in addition, there are blind separation methods based on intrinsic mode function (EMD) and overall intrinsic mode function (EEMD) [4-5] , the problem of the EMD method is that there is a phenomenon of modal aliasing, and the problem of the EEMD method is that the calculation amount is large and the calculation time is long, so the immediacy is not strong in practical engineering applications

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Feature extraction method based on single-channel signal blind-separation rolling bearing
  • Feature extraction method based on single-channel signal blind-separation rolling bearing
  • Feature extraction method based on single-channel signal blind-separation rolling bearing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] The feature extraction method of the rolling bearing based on single-channel signal blind separation in this embodiment specifically includes the following steps:

[0070] Step 1: Given two original vibration signals s 1 and s 2 , respectively:

[0071] the s 1 =cos(2πf 1 t+π / 3)

[0072] the s 2 =cos(2πf b t)[1+βcos(2πf r t)]

[0073] Among them, f 1 = 25Hz, f r = 25Hz, f b =115Hz, β=2, the number of sampling points is 1024, the sampling frequency fs is 1000Hz, and the mixed signal model s=as 1 (t)+bs 2 (t)+n(t), wherein, a=1, b=1, n(t) is a random white noise signal. Simulation signal s 1 , s 2 and the time-domain and spectrogram of s, such as figure 2 shown.

[0074] Select the frequency slicing function, and then perform frequency slicing wavelet transform on the above mixed signal s to obtain its time-frequency diagram in the 0-fs / 2 frequency band, then inverse transform the signal after frequency slicing wavelet transform to obtain the reconstruct...

Embodiment 2

[0092] The method of the invention is used to perform blind separation on a mechanical bearing fault signal, and extract the fault characteristic frequency. The bearing fault is damage to a bearing rolling body. A vibration sensor is installed at the driving end of the bearing, and the sampling frequency fs is 12K Hz. The equipment is loaded with a 1HP load, and its rotational speed is 1777r / min, that is, its fundamental frequency is 29.6Hz. The characteristic frequency of the rolling element fault is calculated according to the characteristic frequency coefficient of its components is 118.1Hz. When the bearing is running, the bearing balls interact with the inner ring and the outer ring, and the source signal has a certain correlation.

[0093] Step 1: Measure the single-channel observation signal through the sensor, and its time-domain waveform diagram is as follows Figure 8 shown.

[0094] Select the frequency slicing function, and then perform frequency slicing wavelet ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a feature extraction method based on a single-channel signal blind-separation rolling bearing. Frequency slice wavelet transformation is performed on an original vibration signal, an energy spectrum of the signal is obtained, selection of a wavelet basis and selection of a slice are not relied on, and the problem that wavelet transformation frequency band selection is limited is also overcome. The number of signal sources is determined through adoption of a principal component analysis method, and the problem that a clustering conclusion is obtained difficultly when a sample is large is solved. A dimension-reduction vector projection matrix is obtained according to principal component analyses in a projection manner, an influence of signal sparseness is prevented, and then an underdetermined problem is converted into a properly posed problem. Furthermore, fault feature extraction of a source signal having some correlation also can be effectively achieved.

Description

technical field [0001] The invention relates to the technical field of mechanical fault diagnosis, and relates to a feature extraction method for blindly separating rolling bearings based on single-channel signals. Background technique [0002] Blind signal separation is a hot research topic in the field of signal processing in recent years. The so-called blind signal separation refers to the method of recovering the original signal that cannot be directly observed from the mixed signal of multiple signals observed by several sensors. Blind signal separation generally requires more sensors than signal sources, while single-channel blind separation means that there is only one sensor for observing mixed signals. This is a difficult point in blind signal separation, but it is closer to the actual conditions in the engineering application of mechanical fault diagnosis. [0003] For underdetermined blind separation, that is, the number of sensors is less than the number of sou...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G01M13/04
Inventor 段晨东薛周舟徐先峰宋苏臣袁野李婷刘晨祁霞耿博望马晓玉
Owner CHANGAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com