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

Semi-supervised bearing fault diagnosis method based on graph theory

A fault diagnosis and semi-supervised technology, applied in the direction of mechanical bearing testing, etc., can solve problems such as fault diagnosis interference, category imbalance, and poor performance of fault diagnosis algorithms

Active Publication Date: 2018-12-28
BEIJING JIAOTONG UNIV
View PDF8 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional fault diagnosis algorithms tend to perform poorly in this situation where there is only a small amount of labeled data and the categories are unbalanced.
In addition, bearings generally operate under uncertain variable conditions, which will also interfere with fault diagnosis

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
  • Semi-supervised bearing fault diagnosis method based on graph theory
  • Semi-supervised bearing fault diagnosis method based on graph theory
  • Semi-supervised bearing fault diagnosis method based on graph theory

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0077] The present invention will be further described in detail below in conjunction with accompanying drawings and examples.

[0078] The invention proposes a visual graph algorithm to convert the signal into a complex network, and extracts the visual graph features as the input of the semi-supervised learning algorithm, so as to realize the fault diagnosis method. like figure 1 As shown, the specific steps are as follows:

[0079] Step 1: Transform the bearing vibration signal into a complex network through a visual graph algorithm.

[0080] For a sample with N data points, first convert the signal into a visual graph by the following method. like figure 2 Schematic for conversion.

[0081] The math works like this:

[0082] Any two points of the signal (t a ,x a ), (t b ,x b ), corresponding to two nodes in the visual graph, the conditions for connecting two nodes are as follows:

[0083] Let any point between these two points be (t c ,x c ),Satisfy

[0084] ...

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 provides a semi-supervised bearing fault diagnosis method based on a graph theory. The method comprises the following steps: firstly converting a bearing original vibration accelerationsignal obtained by virtue of a sensor into a complex network by utilizing a visibility graph algorithm; then calculating structural parameters of the complex network, and extracting a mean value and astandard deviation of degree distribution as well as a network complexity index; and finally processing unlabeled samples by utilizing semi-supervised learning based on a graph, thus bearing fault diagnosis is realized. The method provided by the invention is based on a small quantity of labeled samples and unlabeled samples, realizes the bearing fault diagnosis under the condition that working conditions are variable and sample classes are unbalanced, is high in fault identification accuracy and has obvious use value.

Description

technical field [0001] The invention relates to a graph theory-based semi-supervised fault diagnosis method for bearings, which belongs to the technical field of fault diagnosis of mechanical parts. Background technique [0002] Rolling bearings are one of the most frequently used mechanical components. According to statistics, more than 40% of all mechanical failures are caused by bearings. In the industrial field, the collection of labeled samples is a difficult job, especially the scarce fault samples, but there are a large number of unlabeled samples. For this situation where there is only a small amount of labeled data and the categories are unbalanced, traditional fault diagnosis algorithms often perform poorly. In addition, bearings generally operate under uncertain variable conditions, which will also interfere with fault diagnosis. Therefore, it is of great significance to improve the accuracy of fault diagnosis by using unlabeled data. [0003] The visual graph...

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): G01M13/04
Inventor 王志鹏陈欣安贾利民张蛰秦勇王宁耿毅轩
Owner BEIJING JIAOTONG UNIV