Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Fault diagnosis method of variable-speed rotation machine based on time-frequency spectrum segmentation

A technology for fault diagnosis and rotating machinery, applied in the testing of mechanical components, testing of machine/structural components, measuring devices, etc., can solve problems such as low signal-to-noise ratio, difficult impact signal extraction, and difficult training of neural network models , to achieve the effect of effective fault diagnosis

Active Publication Date: 2018-06-01
WUHAN UNIV OF SCI & TECH +1
View PDF6 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, there are two major defects in the above-mentioned patents: ① For impact fault signals with low SNR, the effect of wavelet decomposition is directly related to the selected wavelet base and the number of decomposition layers. ②For variable-speed rotating machinery, due to the uncertainty of the signal, it is difficult to train an effective neural network model through limited samples, so it is impossible to diagnose the fault of variable-speed rotating machinery

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
  • Fault diagnosis method of variable-speed rotation machine based on time-frequency spectrum segmentation
  • Fault diagnosis method of variable-speed rotation machine based on time-frequency spectrum segmentation
  • Fault diagnosis method of variable-speed rotation machine based on time-frequency spectrum segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0114] The impact component of the collected fault signal is non-periodic, sparse and heterogeneous, affected by noise, the impact component is submerged in the noise, and it is difficult to directly identify it. The classic Hilbert envelope spectrum analysis is used, such as figure 2 and image 3 As shown, it is still difficult to find the multiplier information, it is difficult to extract the characteristic frequency information of non-periodic impact faults, and it is impossible to carry out fault diagnosis. figure 2 and image 3 The simulated signal and its Hilbert envelope spectrum are shown.

[0115] Such as figure 1 As shown, applying this patented technology for sparse feature extraction, the steps are as follows:

[0116] (1) Obtain the normalized time-spectrum of the signal through multi-resolution generalized S transform, and generate multi-resolution binarized time-spectrum after binarization processing.

[0117] First, determine the value range of the scale ...

Embodiment 2

[0132] The outer ring crack fault vibration signal accelerated linearly from 150rpm to 600rpm within 3 seconds was collected from the rotating machinery fault simulation platform. The basic information is as follows: the outer ring of the bearing is fixed, the contact angle of the bearing is a=0, the pitch diameter D=39.5mm, the diameter of rolling elements d=7.5mm, the number of rolling elements Z=12, the sampling frequency is 10kHz, and the number of sampling points is 4096 points. Figure 14 Shown is the acquisition of vibration signals, Figure 15 is its Hilbert envelope spectrum.

[0133] Figure 14 It can be seen that in the process of accelerating rotation, the overall amplitude of the signal increases accordingly, and the intensity of the impact signal also gradually increases, accompanied by some noise components with large amplitudes. Occurs periodically with shorter and shorter intervals. and Figure 15 It is difficult to extract obvious periodic components fro...

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

PropertyMeasurementUnit
Diameteraaaaaaaaaa
Login to View More

Abstract

The invention provides a fault diagnosis method of a variable-speed rotation machine based on time-frequency spectrum segmentation. The method includes following steps: step 1, obtaining normalized time-frequency spectrums of signals through multi-resolution generalized S transform, and generating multi-resolution binarized time frequency spectrums; step 2, combining the binarized time-frequency spectrums with all resolutions, and obtaining an optimal binarized time-frequency spectrum; step 3, segmenting the optimal binarized time-frequency spectrum into a plurality of connected domains, and performing information annotation on each connected domain; step 4, extracting an optimal expression atom of each connected domain, forming an optimal atom set, and calculating the expression of a fault signal in the optimal atom set; and step 5, calculating the appearance time and amplitude of an impact theory, and realizing fault diagnosis of variable-sped mechanical equipment through informationcomparison. According to the method, most strong background noises can be filtered, and fault diagnosis of impact type faults including cracking, pitting corrosion or spalling etc. of the variable-speed rotation machine can be realized.

Description

technical field [0001] The invention relates to the field of fault diagnosis of rotating machinery, in particular to a fault diagnosis method for variable-speed rotating machinery based on time-spectrum division. Background technique [0002] The fault diagnosis of key rotating parts of rotating machinery plays a vital role in ensuring the safe and stable operation of the equipment and predicting the remaining life of the equipment. When impact faults occur on rotating parts, such as pitting, cracks, peeling, etc., damping vibration will occur. The vibration signal collected by the acceleration sensor is a pulse attenuation signal. The identification and extraction of such pulse attenuation signals is the key to fault diagnosis. premise. For mechanical equipment with stable speed, the characteristics of its pulse attenuation signal are periodicity, sparseness and weakness, and finally the fault type can be determined through the fault characteristic frequency; for variable ...

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
IPC IPC(8): G01M13/00
CPCG01M13/00
Inventor 严保康周凤星李维刚宁博文徐波
Owner WUHAN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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