Based On The Wavelet Algorithm Of Power System Harmonic Analysis

With the development of power electronics, the number of rectifier, frequency converter arc furnace and all sorts of power electronics facility are continuously increasing in distribution network. The electricity characteristic of nonlinear, impact and unbalance has seriously polluted power-supply system. On the other hand, electricity equipments of modern industry, commerce and inhabitant user are more sensitive to power-supply quality and have higher demand on it. Nonharmonic has become the main symbol of green. Much attention is pay to harmonic analysis.Wavelet transform is a kind of time-frequency analysis, which has the characteristic of multiresolution and can symbolize the signal's characteristic in time-frequency field. The shape of time-frequency window is variable while the size is invariable, namely time window and frequency window both are variable. It has higher frequency resolution and lower time resolution in low frequency segment while higher time resolution and lower frequency resolution in high frequency segment. It's quite fit for detecting normal signal which carrys transient abnormal phenomena and showing its components.Firstly, the basic knowledge of power harmonic and wavelet is introduced in this paper. In chapter 3 the Morlet wavelet function is used to design FIR filter. By selecting parameter a, the ideal value of stopband attenuation and transition band can be obtained. The FIR filter has better characteristic and easy design. And then, By combining selfadaptive IMS arithmetic with wavelet transform, the selfadaptive wavelet LMS arithmetic, which has higher convergence speed, is constructed and used to analyse power harmonic. In chapter 4 the second generation wavelet, which can analyse signal's frequency characteristic by transforming only in time field, has better quality and is applied to the design of power active filter. In chapter 5 a multilayer forward wavelet neural network(MFWNN) is constructed and used to simulate band-pass filter and detector of parallel harmonic measure setting. Trained up, the MFWNN can measure harmonic accurately. The more sample types the MFWNN is trained, the more accurate the MFWNN , which can measure mutation signal, will be. The number of mutation signal types which can be measured will be more too. Finally.combined wavelet arithmetic with Prony analysis, with the help of nicer noice-cancel capability of wavelet arithmetic, computation precision on data with noice haved been improved.Simulation results are presented to verify the analysis and synthesis methods. Overall, excellent results are obtained.