Abstract:
Given that SF
6/N
2 gas-insulated switchgear (GIS) has been gradually applied on a large scale, and most of the current maintenance experience comes from pure SF
6 GIS, how to differentiate the maintenance of SF
6/N
2 GIS has become an upcoming realistic issue. Therefore, a comparative analysis of the statistical characteristics of partial discharges in typical defects of SF
6/N
2 and pure SF
6 gases is carried out in this work. Firstly, the discharge signals of the two gases have been measured during the whole process of discharge under the typical defects of triple bonding point, suspension potential and tip bulge. Then the differences in the statistical behavior of the two gases are compared and the feasibility of training migration for diagnostic models have been analyzed. The results show that the discharge onset voltage of SF
6/N
2 under typical defects creepage dicharge, suspension discharge, point discharge is reduced to 50%, 48% and 82% of that in SF
6, respectively; and the pulse amplitudes, pulse repetition rates and discharge intensities of the two gas insulation systems differ significantly. Moreover, it is necessary to establish a specialized database for SF
6/N
2 typical defective discharge diagnosis, given that the original SF
6 diagnostic algorithms cannot achieve effective migration to mixed gases. Ultimately, a tracking and warning method applicable to typical defective discharges of SF
6/N
2 gas mixtures is established in this work, providing a basis for diagnosis and overhaul decision-making of internal discharges in environmentally friendly gas-insulated switchgear.
下载:
