Statistical approach to identify the discharge source in MV cables and accessories

Partial discharge (PD) analysis is a reliable tool to assess the integrity of electrical insulation. Representation and interpretation of the data, obtained from e.g. online PD monitoring, are key issues to reveal the discharge source, i.e. defect type, as well as the physical phenomena behind the occurrence. Analyses of various PD patterns such as discharge height distribution presented in this work provide useful statistical parameters to identify the discharge source. Research shows that the 2-parameter Weibull distribution is a reliable model to quantify the characteristics of the patterns of the defect. The model fits well to the charge-height distribution. In addition, trends in the discharge density pattern that occur over long times, can be used as complementary information to discover the discharge nature. It alerts for a possible failure and therefore assists in taking corrective measures to prevent failure. This paper presents the application of such statistical modeling to the area of on-line power cable diagnostics. Data obtained from laboratory experiments as well as field data have been studied.