Research On Detection Of The Electrical Harmonic Based On Wavelet-Transform

In the recent years, the number of rectifier, frequency converter, are furnace and all sorts of power electronics facility are continuously increasing in distribution network with the development and far-ranging of power electronics. The electricity characteristic of nonlinear impact and unbalance has seriously polluted power-supply system. It has become a curse to power system, which deteriorates the quality of electric energy. Harmonic detection is the important part of harmonic issue and harmonic detection method is the key of harmonic measurement. This paper primarily studies the harmonic detection methods.This paper introduces the concept of harmonic, harmonic mechanism of production, hazard of harmonic, harmonic source and deficiency of traditional harmonic detection method. Then it discusses several transform methods, such as continuous wavelet transform, discrete wavelet transform and wavelet packet transform of which characteristics are analyzed. Against the exiting question about which is hard to real-time analyze and measure speedy and exactly measurement of changing harmonic, this paper studies and realizes a harmonic analysis and measurement method based on reconstruction of wavelet packet decomposed coefficients. This paper puts forward a new concept of harmonic virtual value. What's more, this paper studies, puts forwards and builds a detection method of harmonic virtual value and singularity. According to this method, power system harmonic track can be realized. This method provides theory basis for taking an effective method for harmonic real-time compensate. Using wavelet transform can make one voltage process decompose to a frequency band. According to sample frequency and decompose scale, frequency band can be confirmed. Harmonic content analysis is realized. All integral and non-integral harmonics are obtained. This method overcomes information missing resulting from spectrum leak, fence effect in traditional harmonic measurement. Reliability and accuracy of harmonic are improved. The self-adaptive power active filter is designed by combining the arithmetic of reconstruction of wavelet packet decomposed coefficients based on this theory.