Research On Application Of Wavelet Transform In Power Quality Detection And Analysis

The power quality issues are more and more serious, while power system is developing quickly and a great deal of nonlinear, impact and unsymmetrical load is taken to the power system. The first step to improve the power quality is the detection and analysis of power quality. We should detect the power quality issues quickly and exactly, and take them into analysis effectively. So we can control and treat the issues, while we confirm the types and areas of the issues.The detection and analysis of power quality is detecting and analyzing the power system signal waveform. The technical parameters of power signal are taken by detecting and analyzing the power signal, and they provide evidence for improving the power quality. As an effective digital signal processing method, the wavelet transform applied and developed fast in the detection and analysis of power quality. It is significant in theory and practice to study on wavelet tranform theory and application in power quality.The research on application of wavelet transform in power quality detection and analysis is addressed in this thesis, which includes harmonic analysis of power system based on the wavelet transform, detection and analysis of power quality transient based on the wavelet transform, identification of power quality disturbance based on the wavelet transform and perceptron algorithm. The main research works and achievements are outlined as follow:1. After studying harmonic analysis method of power system and wavelet transform theory, a method for harmonic analysis of power system based on the wavelet transform was presented. It analyzes the harmonic of stationary power signal and non-stationary power signal, which uses the advantage of multiresolution of wavelet transform and the characteristic of power system harmonic. First, the appropriate wavelet function is chosen, then the optimal decomposition level is selected, and the frequency band of signal is well divided. At last, the electric signal of each frequency band is decomposed and reconstructed by wavelet transform, and the fundamental components and the harmonic components are obtained. To analyze the each of harmonic component, the method which used the Daubechies40 wavelet packet analyzing and recomposing can show all of the needed harmonic component. It solved the problem of the even harmonic analysis, because the frequency of even harmonic was always in the demarcation point in the traditional way.2. After studying the detection and analysis of power quality transient and the theory of wavelet to detect singular signal, a method for the detection and analysis of power quality transient based on the wavelet transform was presented. According to the non-stationary, random and transient characteristic of power quality transient and capability of wavelet to detect singular signal, the wavelet transform is applied to analyze the power quality transient. The signals of swell, sag and temporary interruption are detected and analyzed by the method of wavelet transform. The experiment show the method can locate the positions of the fault signal. It also can obtain the duration and amplitude of the transient signal. The experiment proved the advantage of this method.3. After studying the detection and identification of power quality disturbance signal and the theory of the wavelet transform and perceptron algorithm, a method for identification of power quality disturbance based on the wavelet transform and perceptron algorithm was presented. By the wavelet transform and energy analysis of power quality disturbance signal, two features were extracted from the wavelet transform results of the power quality disturbance signals and they are related to the power quality disturbance signals. And the two features can be used as the input vectors of the perceptron algorithm to identify the power quality disturbance signals effectively. The validity and accuracy of this method is verified by the experiment results.