Classification of partial discharge sources in mica-epoxy-glass insulation sample using statistical analysis

Abstract

Partial Discharge (PD) in rotating machine insulation system is deleterious. Even though the main insulation of rotating machine is Mica, which is highly PD resistant, the discharges taking place in slot region and end windings are highly deteriorating to insulation system. In this paper, PD measurements are taken in Mica-Epoxy-Glass insulation sample using cylindrical-plane electrode system. The sample contains a defined void on the surface. Distinguished Phase Resolved Partial Discharge (PRPD) patterns are obtained for different location of void. The different PRPD patterns obtained corresponds to internal void in the bulk of insulation, void between high voltage electrode and insulation surface, void between ground electrode and insulation surface; and surface discharge. The PRPD patterns obtained from void between high voltage electrode and insulation surface, in the actual scenario, indicates PD originating near the conductor surface inside the insulation system of rotating machines. The PRPD patterns obtained from void between ground electrode and insulation surface indicates PD originating on the surface of insulation such as slot discharge, end winding discharge in rotating machine insulation system. The variation of PD parameters like Charge magnitude (q) with respect to applied voltage is shown for different location of void. Further statistical parameters like skewness and kurtosis are calculated for charge distribution (n-q) and phase distribution (n-Φ) for different PRPD patterns. It is inferred that statistical analysis of PD parameters is a good tool to classify of partial discharge sources.