Research And Improvement Of Voltage Fluctuation And Flicker Detection Method

In recent years, with the increase of the impact load and nonlinear load, the power system load becomes much more complicated, making the power quality problems turn out to be more serious. Being the most vital power quality issues, voltage fluctuations and flickers are now deeply concerned. Therefore, it is of considerable significance to take appropriate and effective measures to detect and investigate voltage fluctuations and flickers, which is a target for promoting the development of flicker detection device and alleviating the power quality problems.Firstly, the basic concepts and the relevant terms of the voltage fluctuations and flickers are illustrated, with the detection technology development status being summarized. In the aspect of voltage fluctuation and flicker signal envelope extraction, the Paper discusses several commonly-used detection methods, focusing on the comparative analysis of Hilbert transform method, Teager energy operator method, wavelet transform synchronization method and Square demodulation method. In the aspect of estimation of the parameters related to the modulation component of the extracted signal envelope, the paper compares the Pisarenko harmonic decomposition method with the extended Prony algorithm, both of which are frequently-used modulation spectrum estimation algorithm. The paper mainly analyzes the Hilbert transform and the extended Prony algorithm, combines them, and applied the combination to the detection of voltage fluctuation and flicker. The extended Prony algorithm is easily affected by noise, exerting an adverse effect on the analysis of non-stationary signal and noise signal. As a result, extended Prony analysis and mathematical morphology filter, in the paper, are integrated into the application of voltage flicker detection. First, the method adopts Mathematical morphology filters for signal filtering; it can filter the noises which cause interference. Then, using the Hilbert transform to extract the flicker envelope. Finally, using the extended Prony algorithm to carry out spectral analysis on the envelope and obtain the modulation amplitude, frequency and other parameters.Example analysis shows that the above fusion algorithm can effectively improve the accuracy of the Prony algorithm in terms of estimating the characteristics of component parameters of the flicker signal. Also, it illustrates that the method has great practical significance in the voltage fluctuations and flickers detection and analysis of the power system.