Study On Detection And Analysis Of Power Quality Disturbance

In the dissertation, the recent increased interest in power quality is explained and the art of the state of study on the power quality are summarized. Several key issues, such as power quality disturbance detection, voltage sags analysis and characteristics extraction, transient disturbances analysis and characteristics extraction and power quality disturbance signal compression are comprehensively and thoroughly studied and analyzed in the thesis .The main contents and contributions are as follows:1. A new power quality disturbances detection approach based on the Dynamics is proposed. The concept of the Dynamics is first introduced to the field of power quality. Based on the flooding simulation, an efficient algorithm of Dynamics is presented. Through the analysis of the waveform of power quality disturbance signals, it is found that the Dynamics of distortion point is great different from the Dynamics of the peak or valley of the signal, so the distortion point can be detected easily. Then the new approach is tested by several simulation disturbance signals and measurements. The test results show that the new method is effective, which can accurately detect several kinds of disturbances, such as transients disturbances, voltage sags and harmonic distortion.2. The simulation program and STFT time frequency method are used to analyze and extract the characteristics of voltage sags caused by system fault, motor starting and transformer energizing. The analysis results illuminate that the voltage sags disturbance caused by the transformer energizing usually contains the 2nd to 5th harmonic components. By selecting the suitable window function and the length of the window, different frequency components can be separated and extracted with STFT. Then the features of three kinds of sags are summarized which are important for disturbances identification and the improvement of the power quality.3. The method based on the Prony algorithm is suggested to estimate the frequency and damping of the transient disturbances, mainly oscillatory transients. The capacitor energizing transients and voltage magnification phenomena are simulated with EMTP in the paper. Then the Prony algorithm is used to estimate the magnitude, phase, frequency and damping of each component of the transients disturbance. The efficiency of the Prony algorithm for transient disturbance is proved by the results.4. A new power quality disturbance compression method is proposed. Through analyzing the local time-frequency characteristics of power system disturbance signals, the LeastMaximum method is presented to overcome the threshold problem of the signal compression. Some factors such as decomposition level and frequency-band overlap, which have important effect on the compression results, are also studied in the paper. The test examples prove the validity of the new method, with which the local time-frequency characteristics can be kept and the compression ratio is rather large.