A Mechanical Fault Diagnosis Method Based on Probability Box Model Modification

A technology of mechanical faults and diagnostic methods, which is applied in the testing of mechanical components, testing of machine/structural components, instruments, etc., can solve problems affecting the correct rate of pattern recognition and overlapping between models, and solve the problems of difficult and reliable space-time registration. Strong transplant effect

Active Publication Date: 2020-07-31
KUNMING UNIV OF SCI & TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the probability box modeling still has the following problems: First, there is overlap between the models, which affects the accuracy of pattern recognition
Second, the compactness of the original probability box model is not optimal

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
  • A Mechanical Fault Diagnosis Method Based on Probability Box Model Modification
  • A Mechanical Fault Diagnosis Method Based on Probability Box Model Modification
  • A Mechanical Fault Diagnosis Method Based on Probability Box Model Modification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] Embodiment 1: Numerical simulation: take a probability box containing 10 focal elements as an example to perform simulation calculations. Assuming that the random variable x obeys the exponential distribution, and the interval of the exponential distribution parameter λ is [2, 3], the original DSS and the improved DSS' are respectively obtained by using the probability box modeling method and the model correction method. The comparison results of simulation experiment data are shown in Table 1:

[0062] Table 1 Simulation experiment data (n=10)

[0063]

[0064]

[0065] The improved focal element interval is smaller than the original focal element interval, namely Therefore, the improvement rate of focal element compactness must also be improved. Judging from the results of focal element compactness improvement rate, the compactness of 10 focal elements has been improved to varying degrees. The original probability obtained by adding the DSS before and after co...

Embodiment 2

[0066] Embodiment 2: In this embodiment, the rolling fault signal is used as the experimental verification object, and the vibration acceleration signals of the normal bearing state, inner ring fault, outer ring fault and rolling element fault are respectively collected, and the original probability box model and the corrected probability are obtained The box model is as figure 1 and figure 2 shown, from figure 1 and figure 2 It can be seen that the modified probability box is more compact than the original probability box; the overlapping phenomenon of the probability box of rolling element fault and inner ring fault has been improved after the model is revised, which will be beneficial to the pattern recognition based on the probability box;

[0067] Extract the corresponding 8 feature indicators from the obtained probability boxes to form a fault feature vector group. After converting the fault signal into the corresponding skewness probability box, 100 probability boxe...

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 mechanical fault diagnosis method based on probability box model correction. According to the method, fault data in an industrial process is collected, and an original probability box is acquired; an appropriate probability box model is selected; an original DSS is acquired; a comprehensive additional information quantity of industrial test data is defined; an optimized DSS is extracted; and a new probability box is obtained. The mechanical fault diagnosis method based on probability box model correction is proposed to solve the overlapping phenomenon among probability boxes in the industrial mechanical fault diagnosis process and improve the compactness of the probability boxes. Through the method, the probability box model of the industrial test data is obtainedthrough a probability box modeling method, a mean value of focal element intervals and a data fluctuation quantity between adjacent focal elements are used as the additional information quantity, a Bayesian method based on maximum entropy is utilized to correct the probability box model, the compactness of the corrected model is improved, the overlapping phenomenon among models is relieved, and more accurate information is provided for further increasing the correct recognition rate of mechanical fault diagnosis through the probability box model.

Description

technical field [0001] The invention relates to a mechanical fault diagnosis method based on probability box model correction, belonging to the technical field of industrial process mechanical fault diagnosis. Background technique [0002] The impact of mechanical failure is particularly prominent in actual industrial production. In rotating mechanical equipment, more than half of mechanical failures are caused by the inability to accurately detect and identify mechanical failures. The operating state of the machine directly affects the performance and efficiency of the rotating machine, so the research on its fault diagnosis is of great significance. Information fusion has changed the traditional research mode of fault diagnosis without intersection among various research points, and has become a new research hotspot. The use of probability box modeling can take into account many subjective and objective uncertain factors in fault diagnosis, which makes up for the defect t...

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): G06F30/20G01M13/00G01M13/045
Inventor 杜奕蒋慧英丁家满刘力强
Owner KUNMING UNIV OF SCI & TECH
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap