The Research On The Suppression Method Of Disturbances For On-line Partial Discharge Monitoring Of Transformer

The power transformer is the important equipment in power system, its operating state effects the safety of the electric energy production and the transformer accidents will lead to power cut in the local or large regions which affects the normal daily life. As the accidents statistics shown, about 60%~80% of the accidents are caused by the insulation deterioration of the power transformer for the main reason of partial discharge (PD). The transformer has the complex internal structure, where there is the asymmetry electric field, due to the air and oil gap exist in the insulation material, the PD will occur when the electric field intension reach the certain degree, which will destroy the insulation and lead to the power cut accident, even the disaggregation of the transformer. Therefore, the online PD monitoring is very important for the safety and reliable operating of the power system.The research on partial discharge theory of the power system has been done, and on the basis of analysis research of the works completed by the domestic scholars or the foreigners, the online monitoring technology and anti-interference measure of the partial discharge were summarized.The producing reasons are analyzed according to the white-noise, periodical narrow-band noises and pulse disturbance in the partial discharge signal of the power transformer online monitoring. The wavelet analysis based method is proposed to suppress the white-noise and the digital trap filter is designed to suppress the periodical narrow-band noises based on the different characteristics of the three disturbances. Besides, the research on cluster analysis and insulator contamination which are proposed to suppress pulse disturbance has also been done in this paper. The simulation and experiment show that the methods could complete the anti-interference work well.