Multi-feature fusion method

A technology of multi-feature fusion and calculation method, applied in the field of multi-feature fusion, it can solve the problems of large influence of subjective experience, lack of universality, and inaccurate selection of features.

Inactive Publication Date: 2021-08-06
SHENYANG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Different from the signal feature extraction in fault diagnosis, for the performance degradation characteristics of rolling bearings, continuous feature analysis of the signals is required, and the signals will show strong randomness with the degradation of rolling bearings. Therefore, for the trend of rolling bearing signals However, due to the complexity of rolling bearing vibration signals, this method often leads to inaccurate selection of features and is greatly affected by subjective experience. suitability

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
  • Multi-feature fusion method
  • Multi-feature fusion method
  • Multi-feature fusion method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0093] The experimental data of the IEEE PHM 2012 Data Challenge is used to verify the effect of the proposed method. Since the horizontal and vertical signals of rolling bearings have similar trends in the time domain, the vibration signals of the rolling bearing 1-1 horizontal direction in the IEEE PHM 2012 data set were selected for experiments. The time domain waveforms are shown in image 3 shown. Firstly, the multi-dimensional features of the time domain, frequency domain and frequency domain are extracted, as shown in Table 1-Table 3, and 200 samples in the complete cycle are uniformly selected for analysis in the experiment.

[0094] Table 1 Time domain characteristics

[0095]

[0096] Table 2 Frequency domain characteristics

[0097]

[0098] Table 3 Time-Frequency Domain Characteristics

[0099]

[0100] After extracting the 40-dimensional features of the vibration signal of the rolling bearing, use the similarity analysis based on the Hausdorff distance...

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 relates to the field of rotating machinery fault prediction and health management, in particular to a multi-feature fusion method used in rolling bearing performance degradation evaluation. According to the method, in order to solve the problem of redundancy possibly occurring in multiple features extracted from a mixed domain in rolling bearing performance degradation evaluation, a method for carrying out feature selection by combining similarity analysis based on a Hausdorff distance and a comprehensive evaluation criterion is provided. In the similarity analysis of the features, a Hausdorff distance is adopted, and compared with an Euclidean distance, the Hausdorff distance considers a bidirectional distance in two groups of data, so that the method is more suitable for the similarity analysis of the features, and in addition, a comprehensive evaluation criterion can consider the monotonicity and predictability of the degradation features. Moreover, the selected multiple features are fused, a feature fusion method based on class sensitivity-overlapping tendency is provided, and the fusion process considers the characteristics of the rolling bearing degradation features in different degradation stages and also considers the stage tendency of the features. In order to avoid interference of other factors, a feature overlapping tendency calculation method is provided, and the situation that local optimum is caused when the tendency of a long sequence with the large fluctuation amplitude is calculated is avoided.

Description

technical field [0001] The invention relates to the field of fault prediction and health management of rotating machinery, in particular to a multi-feature fusion method used in performance degradation evaluation of rolling bearings. Background technique [0002] With the rapid development of science and technology, modern mechanical equipment has begun to develop in a more intelligent direction, and rotating machinery is widely used in industry. The core components of many manufacturing equipment are rotating machinery. As one of the most widely used parts in rotating machinery, rolling bearings often operate in very complex environments. Once a failure occurs, it will cause huge property losses and may even threaten life safety. When the rolling bearing fails, the existing treatment method is generally replacement and refurbishment, which will increase the maintenance cost of the equipment and cause unnecessary waste of resources. The service life is one of the important ...

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): G06K9/62G01M13/045
CPCG01M13/045G06F18/22G06F18/24323G06F18/253
Inventor 齐晓轩董海刘英英洪振麒
Owner SHENYANG 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
Try Eureka
PatSnap group products