Ding Dengwei

Time-delay estimation algorithm of partial discharge ultra high frequency signals based on fractional Hilbert transformation

Time delay estimation is the basis of the localization algorithm for partial discharge (PD) based on time difference in GIS, and it is the key to the accuracy of localization. In this paper, the theory of fractional Hilbert transform for time delay estimation is described. Based on fractional Hilbert transform, a method for estimating the time delay of GIS ultra high frequency partial discharge is proposed, and the numerical implementation steps of the algorithm are given. According to the abnormal partial discharge signals found in the switch gate during the inspection of certain 220kV GIS equipment, high frequency electromagnetic wave is collected by the high speed oscilloscope. According to the time delay of the signal, the space position of the partial discharge source is located from the the disk type insulator pouring hole 1.25 meters. After the disintegration of the switch air chamber, the distance between the discharge source and the pouring hole is 1.3 meters. The error of the partial discharge source localization based on fractional Hilbert algorithm is significantly lower than that of the bi-spectrum and the frequency domain time delay estimation algorithm.

Abnormal vibration analysis of GIS based on time frequency detection

The stable operation of high voltage GIS equipment is very important to the safety of power grid. The abnormal vibration of GIS is a serious threat to the safety of equipment. In this paper, the abnormal vibration of 220 kV GIS equipment is detected by the wideband vibration measure system. By means of spectrum and continuous wavelet analysis of vibration signals, the time domain and frequency domain characteristics of the vibration signal are studied. The analysis results reveal the spectrum difference of the ultrasonic signal induced by partial discharge and the equipment vibration, and the propagation attenuation characteristics of the vibration signal on the GIS tube. In addition, the time delay of vibration signal is obtained by the wavelet entropy, therefore the vibration source is accurately located. It carried out on-site confirmation finally. It indicated that the time-frequency analysis can accurately locate the vibration source. It is conducive to the analysis of the causes of the abnormal vibration of GIS and the evaluation of the equipment operating conditions.

Research on the winding and iron core operation state of transformer based on the vibration acoustic fingerprint

Extensive studies show that the winding and iron core is the component of power transformer which is most prone to failure. The vibration detection of transformer is an effective method to evaluate the operation state and diagnose the condition of the winding and iron core. In this paper, the vibration system is applied to detection the signal of tow 220kV transformers. The spectrum distribution of the transformer vibration signal is analyzed by means of the wavelet transform, through which the vibration acoustic fingerprint and two typical characteristic parameters, i.e., the energy ratio of odd and even harmonics, and the waveform distortion ratio, are extracted respectively. The analysis results show there is a certain degree of deformation in the winding of the 1# transformer. It is discovered that the equivalent capacitance of the low voltage C phase winding of the 1# transformer is increased by 7% by off-line test. It further confirms the conclusion of the vibration detection and indicates the validity and practicability of the transformer vibration acoustic fingerprint and the two typical characteristic parameters.