Noise reduction method for leakage current signal of metal oxide arrester

Abstract

The invention discloses a noise reduction method for a leakage current signal of a metal oxide arrester. The noise reduction method comprises the following steps: firstly, acquiring an original signal of the leakage current of the metal oxide arrester through monitoring equipment; secondly, a proper wavelet basis function is selected, the leakage current signal is converted to a wavelet domain to be subjected to multi-scale division, meanwhile, energy of instantaneous frequency is superposed one by one according to the center frequency of a harmonic signal wavelet frequency spectrum, the frequency spectrum energy is gathered on an instantaneous frequency curve, and a synchronous extrusion signal is obtained; finally, band-pass filtering is conducted on the synchronous extrusion signals, high-frequency noisy points are removed, and wavelet components are combined through a time sequence to obtain noise-reduced leakage current signals. According to the method, the collected original leakage current signal of the metal oxide arrester is subjected to synchronous extrusion wavelet transform noise reduction, the problems of aliasing effect, spectrum leakage, fence effect and the like existing in traditional noise reduction methods such as fast Fourier transform and continuous wavelet transform are avoided, a time-frequency curve is effectively sharpened, and waveform features are extracted.

Description

technical field [0001] The invention relates to a noise reduction method for a leakage current signal of a metal oxide arrester. Background technique [0002] Metal Oxide Arresters (MOA) is an overvoltage protection device on power transmission and transformation equipment in the power grid, and has now become an indispensable part of the safe operation of the power system. At present, the general criterion for evaluating the operating state of MOA is its leakage current, and the leakage current measured under working conditions is often accompanied by background noise, such as strong random noise without structure and disordered impulse noise. Therefore, the noise reduction and noise reduction of the leakage current plays a crucial role in the accurate evaluation of the operating state of the MOA. [0003] Usually the measured leakage current is a nonlinear and non-stationary time series signal. After noise reduction processing, it is often difficult to retain the characte...

AI Technical Summary

Problems solved by technology

[0003] Usually the measured leakage current is a nonlinear and non-stationary time-series signal. After noise reduction, it is often difficult to retain the characteristic information of the leakage current, which puts forward higher requirements for the noise reduction method of the leakage current.

The current mainstream leakage current noise reduction methods include: wavelet transform (Wavelet Transform, WT), fast Fourier transform (Fast Fourier Transform, FFT), ensemble empirical mode decomposition (Ensemble Empirical Mode Decomposition, EEMD), etc., and WT method will As a result, the signal reconstruction deviates from the original signal, the FFT method will omit the local characteristics of the signal, and the EEMD method is too dependent on the selection of the number of iterations

Images