Detection And Analysis Of Transient Power Quality Based On The S Transform

With VVVF drives, programmable controllers, computer information systems increasingly used in the modern enterprise, transient power quality has become a major impact on power quality problems. Exact detection and location as well as classification of the transient power quality disturbances are necessary precondition for its monitoring and controlling.This thesis firstly aims at five common transient power quality, including voltage sag, voltage swell, voltage interruption, surge and impulse which are described carefully. Secondly, it is the research on the detection and analysis of transient power quality.This thesis reviews the importance of the measure of transient power quality and introduces the research of transient power quality at home and abroad. And the comparison between Wavelet transform, Fourier transform and the S-transform analysis laid the foundation for the application of the S-transform.Then the thesis reviews the S-transform principle, its implementation and the differences between other time-frequency analysis methods, and the detection method based on the S-transform is proposed, which makes full use of the characteristic curves of the S-transform to detect transient power quality. The simulation of the method shows that the S-transform is powerful in the time-frequency analysis. In order to denoising, a novel morphological filter with multiple structuring elements based on the principle of mathematical morphology is designed. This novel filer is good at denoising the noise of transient power quality signals. The results of the simulation show that the novel filter is effective.Finally, the different characteristic curves of the S-transform could be used to detect various transient power quality disturbances, this method could show the occurred time, duration, amplitude and the frequency accurately, it also could recognize different disturbances types easily and directly. All the results and comparative analysis of simulation show that the detection method is effective.