| Abstract: | The ultrahigh frequency (UHF) method is used for detecting the three kinds of surface discharges started by metallic particles on the spacer insulators in gas‐insulated switchgear (GIS). Three different designs of the same surface defect are applied to ensure the consistency of the results. The present experiments are made on a 220‐kV GIS, which is close to the actual situations to study the surface discharge characteristic. A measuring system for this purpose consists of an oscilloscope connected to UHF sensors. Partial discharge (PD) data was captured by UHF detection using an oscilloscope and stored as phase‐resolved pulse sequence (PRPS) data. Since PRPS data contain the complete information of the PD, PRPS analysis (PRPSA) pattern and three‐dimensional ϕ − q − n (3D PRPD) pattern can be calculated from the recorded PRPS datasets. The PRPSA data also allows us to obtain the actual test voltage (rms) of every cycle, the number of measuring cycles, and phase information of all PD pulses, together with their peak values, so that a statistical analysis of these data can be discussed. Besides, this paper also focuses on the frequency analysis and time–frequency analysis of the single‐pulse time‐domain signal. It is shown that the characteristic parameters obtained from the PRPS data, frequency spectrum, and time–frequency spectrum have the potential to be used as PD characteristic interpretation and further research on PD pattern identification. © 2015 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. |
