An insulating oil raman spectrum feature extraction method, system, device and medium

By using the GAF-Laws method to extract features from the Raman spectra of insulating oil in electrical equipment, the problem of extracting effective information from complex spectra is solved, and efficient and accurate oil condition diagnosis is achieved.

CN116720064BActive Publication Date: 2026-06-09GUANGDONG POWER GRID CO LTD +1
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
GUANGDONG POWER GRID CO LTD
Filing Date
2023-05-26
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing technologies are insufficient to effectively extract the effective feature information from the Raman spectra of insulating oil in electrical equipment, resulting in insufficient diagnostic accuracy, especially when dealing with complex insulating oil spectra where baseline and peak changes are not obvious.

Method used

The GAF-Laws method is used to extract features from Raman spectra using the Gram angular field matrix and an improved Laws texture energy metric. This includes normalization, Gram angular field matrix calculation, and texture energy metric analysis to filter out invalid information and highlight the characteristics of the signal of the component to be measured.

Benefits of technology

This method enables simple and rapid extraction of Raman spectral features from insulating oil, reducing human error, ensuring the validity of spectral data and the accuracy of oil condition diagnosis, and improving diagnostic accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116720064B_ABST
    Figure CN116720064B_ABST
Patent Text Reader

Abstract

The application provides an insulating oil Raman spectrum feature extraction method, system, device and medium, the method comprises the following steps: acquiring the original Raman spectrum of an insulating oil sample, and performing normalization processing on the original Raman spectrum to obtain a normalized Raman spectrum; performing feature expansion on the normalized Raman spectrum to obtain a corresponding Gram angle field matrix, and obtaining a corresponding Gram angle field image according to the Gram angle field matrix; performing feature extraction on the Gram angle field image according to a Laws texture energy measure to obtain insulating oil Raman spectrum features. The application can not only effectively extract insulating oil Raman spectrum feature information, highlight the characteristic of the to-be-measured component signal, ensure the effectiveness of the spectrum data, but also is simple, fast, requires a small number of parameters, can avoid the introduction of human error in the calculation process, effectively ensures the accuracy of oil state diagnosis using Raman spectrum data, and has high application value.
Need to check novelty before this filing date? Find Prior Art