Partial Discharge Spectrum Feature Pattern Recognition Algorithm Based on Gaussian Parameter Fitting

Abstract

The invention discloses a characteristic pattern recognition algorithm for a partial discharge spectrum based on Gauss parameter fitting. The algorithm comprises a) the partial discharge spectrum in the partial discharge process to be measured is obtained; b) based on the obtained partial discharge spectrum, the distance between the to-be-identified characteristic quantity of partial discharge and the standard characteristic quantity of partial discharge in the partial discharge process to be measured is obtained via a Gaussian function; and c) based on the obtained distance, an identification result of the partial discharge mode is obtained and output. The characteristic pattern recognition algorithm can overcome the defects that in the prior art, the anti-interference capability is low, the identification difficulty is high and the reliability is low, and has the advantages of high anti-interference capability, low identification difficulty and high reliability.

Description

technical field [0001] The invention relates to the technical field of discharge spectrum processing, in particular to a partial discharge spectrum characteristic pattern recognition algorithm based on Gaussian parameter fitting. Background technique [0002] In order to meet the needs of power system maintenance, the partial discharge on-line detection technology of high-voltage electrical equipment, especially the key technology-partial discharge pattern recognition, has been greatly developed. A group of classification-related parameters used to describe the partial discharge fault pattern recognition system is the feature quantity of the recognition system. The basic task of feature quantity extraction is how to find out the most effective feature quantities from many features of the system and study how to compress the high-dimensional feature space into a low-dimensional feature space in order to effectively design classifiers. [0003] In the 1990s, the pattern recog...

AI Technical Summary

Problems solved by technology

However, on the one hand, many uncertain factors affect the collection of partial discharge signals; on the other hand, the connotation and law of the information contained in partial discharge signals are not completely clear, and there is no mature diagnostic theory or standard so far; therefore, partial discharge Discharge pattern recognition technology is still in the research stage

[0004] In the process of realizing the present invention, the inventors found that there are at least defects such as poor anti-interference ability, difficult identification and low reliability in the prior art

Images