The Research And Simulation Of Detecting Harmonics Based On Wavelet Transform

The contradiction between the poor quality of electric power and the demand of high power quality has come to an important isuue nowdays. As the content of harmonic is an important parameter, study of the harmonic detection have great value in the theory and practice.This paper introduces the definition of harmonic, description of harmonic, hazard of harmonic, and deficiency of traditional harmonic measurement methods. Also it analyzes several common sources of harmonic, and consructs their models. Moreover it discusses some the types of wavelet transform and the methods of the wavelet analysis,such as continuous wavelet transform, discrete wavelet transform and wavelet packet transform etc. And the applications of the wavelet analysis and instantaneous reactive power theory in harmonic detection are researched by the experimental simulation in this paper.Due to the complexity of theory and model in harmonic detection, the process of formula is very lengthy. Also the workload of mathematical analysis is especially large. So in the study, the people often analyze and deal with some of the key signals with the MATLAB mathematical tool. In this paper SimuWave wavelet analysis toolbox was first introduced to China and creatively applied to simulation model of the harmonic detection. Comparing to the commonly used M language programming and GUI wavelet analysis method, the simulation studies of the harmonic detection are greatly simplified. Based on the toolbox, we developed the SlMULINK module, called "wavelet filter module", which has a function of wavelet analysis. This module can be inserted into any simulation link. At the same time it together with the harmonic source model and other modules can integrated algorithm model of the harmonic detection which makes the process of simulation more coherent and has intuitive and user-friendly features.Against the conditions that currently performance requirements of the harmonic detection become harsher, the paper bulids a virtual platform of harmonic detection with wavelet analysis, and study the two harmonic detection technology programs in the detail--the joint time-frequency domain analysis of direct wavelet analysis and time-domain analysis of the instantaneous reactive power theory. Using SimuWave toolbox, we constructs a SIMULINK model of harmonic detection based on the multiresolution analysis and wavelet packet analysis, discusses the meathod that harmonic signal can be directly processed by the wavelet analysis, put forwards the algorithm framework of harmonic detection in time-domain analysis and builds its simulation model.Moreover designs a "low delay, low-pass filter anology results, good dynamic" digital filter algorithm based on wavelet model. A novel meathod of harmonic detection is acuquired with a combination of the wavelet analysis and the instantaneous reactive power theory.The results show that the real-time and accuracy of direct wavelet analysis is not satisfactory. To further improve the performance of detection algorithm, we need to the meathod which combined instantaneous reactive power and wavelet analysis theory, and we can find that not only the delay time is reduce by more than half, but the accuracy is maintained at a very high level compared with traditional methods. Also it can detect rapidly the changing harmonic signal, because this method has expanded effectively dynamic range of the system.