Unlock instant, AI-driven research and patent intelligence for your innovation.

Rolling bearing fault detection method and system based on ewt, spectral effective value and knn

A rolling bearing and fault detection technology, applied in the testing of mechanical components, testing of machine/structural components, measuring devices, etc., can solve the problem that the wavelet transform method cannot adaptively realize frequency band division, end effect, and modal aliasing. Complete and other problems, to achieve the effect of highlighting the fault frequency band and eliminating useless signals

Active Publication Date: 2020-11-10
HANGZHOU ANMAISHENG INTELLIGENT TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among the commonly used signal processing methods, the empirical mode decomposition method has defects such as endpoint effects, mode aliasing and incomplete theoretical basis, and the wavelet transform method cannot adaptively realize the problem of frequency band division.

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
  • Rolling bearing fault detection method and system based on ewt, spectral effective value and knn
  • Rolling bearing fault detection method and system based on ewt, spectral effective value and knn
  • Rolling bearing fault detection method and system based on ewt, spectral effective value and knn

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] A rolling bearing fault detection method based on EWT, spectral effective value and KNN, such as figure 1 shown, including the following steps:

[0059] S100. Perform empirical wavelet transform processing on the collected bearing vibration signals to obtain several component signals;

[0060] S200. Process the component signals to determine the number of component signals that need to be reconstructed;

[0061] S300. Reconstruct the component signals that need to be reconstructed to obtain signals within the fault frequency range, and extract characteristic parameters for the signals within the fault frequency range;

[0062]S400. Divide the extracted feature parameters into a training set and a test set, determine bearing fault categories of samples in the training set in combination with the state of the bearing, and determine the bearing fault category to which the test set belongs by using the KNN method.

[0063] In the prior art, there are still certain difficu...

specific Embodiment

[0080] The bearing fault detection method of the present invention will be further explained below in conjunction with an embodiment, taking the acceleration data of the drive end collected on the Case Western Chu bearing simulation test bench at a rotational speed of 1797 r / min as an example.

[0081] Step 1: Perform empirical wavelet transform on the collected vibration signals, and obtain 21 component signals in total, among which, image 3 and Figure 4 is the time-domain diagram and frequency-domain diagram of the bearing vibration signal, Figure 5 It is the frequency band division diagram after the empirical wavelet transform of the bearing vibration signal. The empirical wavelet transform adopts the scale space method to adaptively divide the initial frequency band dividing points, and constructs an orthogonal bandpass filter bank according to the dividing points to obtain the component signals.

[0082] Step 2: According to each component signal, calculate the spectr...

Embodiment 2

[0090] A rolling bearing fault detection system based on EWT, spectral effective value and KNN, such as figure 2 As shown, it includes a preprocessing module 100, a processing module 200, a reconstruction extraction module 300 and a fault type determination module 400;

[0091] The preprocessing module 100 is configured to perform empirical wavelet transform processing on the collected bearing vibration signals to obtain several component signals;

[0092] The processing module 200 is configured to process the component signals and determine the number of component signals that need to be reconstructed;

[0093] The reconstruction extraction module 300 is used to reconstruct the component signals that need to be reconstructed to obtain signals within the fault frequency range, and extract characteristic parameters from the signals within the fault frequency range;

[0094] The fault type determination module 400 is used to divide the extracted characteristic parameters into ...

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 rolling bearing fault detection method based on EWT (Empirical Wavelet Transform), a spectrum effective value and KNN (K-Nearest Neighbor). The method comprises the followingsteps that: carrying out EWT processing on a collected bearing vibration signal to obtain a plurality of component signals; processing the component signals, and determining the number of component signals which need to be reconstructed; reconstructing the component signals which need to be reconstructed to obtain signals in a fault frequency range, and extracting characteristic parameters from the signals in the fault frequency range; and dividing the extracted characteristic parameters into a training set and a test set, combining with the state of a bearing to determine the bearing fault category of a sample in the training set, and adopting a KNN method to determine a bearing fault category to which the test set belongs. By use of the bearing detection method disclosed by the invention, a vibration signal is subjected to the EWT, adaptive frequency band division can be realized, and the problems of frequency band aliasing, end effects and the like can not be generated. The methoduses the KNN method, and the category classification situation of known data can be tested.

Description

technical field [0001] The invention relates to the technical field of bearing fault detection, in particular to a rolling bearing fault detection method and system based on EWT, spectral effective value and KNN. Background technique [0002] Rolling bearings are one of the key parts of rotating machinery and equipment. They support the rotating body and ensure the rotation accuracy. They are widely used in electric power, aerospace, rail transit, machinery and other components. In the working state, bearings are prone to failures, which will affect the performance of the entire mechanical equipment. Therefore, it is necessary to monitor the state of rolling bearings, diagnose the faults in the bearings in time, and avoid accidents. [0003] When the state of the rolling bearing is abnormal, there will be obvious changes in the time domain characteristic parameters of the vibration signal, and a fault frequency band will be generated in the frequency domain. At present, the...

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 Patents(China)
IPC IPC(8): G01M13/045
Inventor 李倩徐剑楼阳冰柳树林吴芳基
Owner HANGZHOU ANMAISHENG INTELLIGENT TECH CO LTD