A method for approximate fault diagnosis based on symbol aggregation

By combining adaptive segmentation and fuzzy proximity, the problem of insufficient effectiveness and accuracy of existing symbolic aggregation approximation methods in fault diagnosis is solved, and a more efficient fault diagnosis effect is achieved.

CN116361627BActive Publication Date: 2026-07-10CHINA NAT PETROLEUM CORP +1
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
CHINA NAT PETROLEUM CORP
Filing Date
2021-12-24
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing symbolic aggregation approximation methods suffer from insufficient effectiveness and accuracy in fault diagnosis, especially in the selection of the number of segments, which ignores signal abrupt change information and fails to effectively preserve data feature information.

Method used

A variance-based symbolic aggregation approximation method (VA_SAX) is adopted. The data is adaptively segmented, and the number of interval segments is determined by the variance ratio. The fuzzy proximity method is combined with the method to identify fault modes, thereby improving the accuracy of feature extraction.

Benefits of technology

It achieves adaptive segmentation of data, retains more effective information, improves the accuracy and efficiency of fault diagnosis, and significantly enhances the effect of fault diagnosis by identifying fault modes through fuzzy proximity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116361627B_ABST
    Figure CN116361627B_ABST
Patent Text Reader

Abstract

The application discloses a kind of based on symbol aggregation approximate fault diagnosis method, comprising: according to the trend of vibration signal waveform change and the variance proportion of each interval to different interval is segmented, the number of string representing model change is counted as the feature vector of data, vibration signal is converted into string and carries out feature extraction, uses fuzzy closeness to the mode recognition of extracted feature, with the size of closeness indicates the probability that equipment is in certain state, to judge the running state of equipment.The application effectively improves the fault diagnosis effect, realizes the adaptive segmentation of data, retains more effective information of data itself;Fault mode is identified by fuzzy closeness method, and the fault diagnosis efficiency is effectively improved.
Need to check novelty before this filing date? Find Prior Art