Research On PD Signal Propagation Characteristics And Wavelet Packet Based Entropy Threshold Method For Interference Suppression Of Power Transformer

On-line PD monitoring of power transformer is presently the research hotspot in the area of high voltage insulation measurement and diagnosis. Due to the weakness of PD signals and bigness of interferences, the extraction of PD pulses faces many difficulties. The developed techniques, with some achievements, have not met the demand of engineering application yet.Aiming at the present problems, this thesis studies the PD signal extraction of Pulse Current Method (PCM) in two directions: signal propagation and interference suppression.Firstly, with software Saber, it sets up the transient models of transformer windings and bus-connected networks and studies the propagation of PD signals in the system. The influences of bus-connected networks, different monitoring points, low voltage windings and types of neutral point grounding on the frequency responses of PD signals are analyzed which give suggestions for the selection of monitoring band and monitoring points and PD localization. Combined with researches on propagation characteristics, the micro-current sensor for on-line PD monitoring is developed and the effects of coil number, stray capacitance and loading resistance on pass-band and sensitivity are also studied.Based on Wavelet Packet Decomposition (WPD), Entropy Threshold Method (ETM) has been put forward to reject the Discrete Spectrum Interference (DSI) in on-line PD monitoring. With the frequency division of WPD, ETM uses Shannon entropy as the criterion of determining whether or not DSI exist in certain WPD tree nodes and interference are suppressed successfully. Lots of simulation data, lab data and on-site data have indicated that ETM works with good efficiency, without pre-knowing of DSI information, extracts the phase of PD pulses accurately and can calibrate quantity of single type discharge. It also studies the influencing factors including sampling frequency, PD waveform, decline time, time intervals between PD pulses, mother wavelet, center frequency of interference, entropy threshold value and SNR.In addition, it applies Wavelet Packet Method (WPM) for non-stationary signal de-noising in on-line PD monitoring and analyzes the processing results under different PD waveform, sampling frequency and noise level. Firstly, the curve of PDpulse is presented on condition of ideal white noise. Signal distortion analysis has shown that the criterion of valuing de-noising capability should be waveform parameters of PD pulse rather than noise level. Conclusions drew in this thesis can be used to guide the design of on-line monitoring system, some of which, tested by on-site data, can be directly applied in on-line PD monitoring of transformer.