GIS partial discharge optical diagnosis method based on multiple fractal and extreme learning machine

Abstract

The invention discloses a GIS optical partial discharge identification method based on multiple fractal and an extreme learning machine. According to the technical scheme, the method comprises the steps that various GIS typical insulation defect models are designed, a laboratory optical detection system is built, optical partial discharge signals are collected, and a graying optical partial discharge spectrum of the GIS under different defects is drawn; based on the multiple fractal theory, the multiple fractal characteristic quantities of difference box dimension and information dimension ofthe graying optical partial discharge spectrum are extracted; the extreme learning machine is constructed as a classifier, and a global minimum value is searched through a linear parameter mode; and training and test samples are input, and an identification result is test. According to the GIS optical partial discharge identification method, the multiple fractal characteristics can improve the identification accuracy of the GIS optical partial discharge spectrum, the extreme learning machine can improve the identification speed of the GIS optical partial discharge spectrum, and the combinationof the multiple fractal characteristics and the extreme learning machine can guarantee the optical diagnosis efficiency of GIS partial discharge.

Description

technical field [0001] The invention belongs to the field of GIS insulation defect detection, and relates to a GIS partial discharge optical diagnosis method based on multifractal and extreme learning machine, which can perform feature extraction and pattern recognition more efficiently and accurately. Background technique [0002] Gas Insulated Switchgear (GIS) is a kind of SF 6 Gas-insulated metal-enclosed switchgear with gas as the insulating medium. Compared with traditional open substations, GIS has the advantages of high insulation strength, stable operation, small footprint and small maintenance workload, so it is widely used in large and medium-sized city power grids. However, from the operation situation in the past ten years, many accidents have occurred in the use of GIS at home and abroad, among which insulation failure is the main one. Therefore, it is of great significance to accurately grasp the nature and characteristics of defects in GIS to carry out parti...

AI Technical Summary

Problems solved by technology

Establishing the mapping relationship between optical characteristic parameters and partial discharge defect types is a fu...

Images