Power cable partial discharge signal denoising method based on wavelet analysis

Abstract

The invention provides a partial discharge signal denoising method based on wavelet analysis, which belongs the technical field of signal processing, and is mainly applied to power cable partial discharge signal denoising. The method for performing intelligent denoising processing on power cable partial discharge signals specifically includes: selecting optimal wavelet basis by using minimum description length (MDL) standard of the information theory, selecting a denoising method, performing wavelet decomposition according to a certain decomposition level, processing wavelet coefficients according to different threshold functions, performing wavelet reconstruction according to the processed wavelet coefficients, and achieving the denoising. The partial discharge signal denoising method solves the problem of manual screening of conventional partial discharge signal denoising, thereby achieving the purpose of the intelligent denoising of the partial discharge signals.

Description

technical field [0001] The invention relates to a method for denoising power cable partial discharge signals, in particular to a method for applying wavelet theory to power cable partial discharge signal denoising. Background technique [0002] Power cables are one of the most important assets of a utility company. With the continuous improvement of power supply reliability requirements, the current accident-based maintenance of power infrastructure is far from meeting the requirements, and the condition-based maintenance, namely "condition-based maintenance" has been paid more and more attention. "Condition maintenance" is to grasp the operation status of the equipment in advance, eliminate hidden dangers of equipment in time, and avoid accidents. Some domestic power companies attach great importance to the condition monitoring and condition maintenance of power equipment, and have successively introduced various advanced detection technologies and equipment. Good results ...

AI Technical Summary

Problems solved by technology

Long-term practice has proved that partial discharge is the main cause of damage to power cables. Therefore, detection and diagnosis of partial discharge is a major measure to protect power cables. However, noise and interference are often mixed in the detection process of partial discharge signals of power cables. , making it very difficult to measure

[0003] At present, the Fourier analysis method is mainly used to deal with the partial discharge signal noise of power cables, which has the ability to describe the signal globally, but it cannot give the law of how the signal spectrum changes with time, and for the non-stationary signal spectrum of power cable partial discharge In the case of large changes over time, the analysis method is required to accurately reflect the local time-varying spectral characteristics of the signal, and it is not enough to only understand the global characteristics of the signal in the time domain or frequency domain

Images