Comparative Study of Phase Resolved Partial Discharge Patterns obtained with Different Instruments on Individual Stator Bars and Coils

Abstract

Partial Discharge (PD) measurements are used both for on-line condition monitoring of stator windings and quality control on newly manufactured stator bars and coils. The use of Phase Resolved Partial Discharge (PRPD) pattern recognition has proven to be an effective tool to identify active PD sources within the insulation system. However, quantification of PD activity based on PD parameters extracted from the PRPD pattern is not trivial and requires a good understanding of the measuring techniques as well as of the physics of the PD phenomenon. This is particularly true for quality control tests on new bars and coils, where the reproducibility of PD results is essential. Since different PD instruments can be used, variability arising from different instruments must be evaluated. In this context, to validate the consistency of PD measurement results, a comparative study with four different commercially available PD instruments was performed in the laboratory on two individual stator bars and two coils. Quantification of PD results has been based on the combination of PRPD pattern analysis correlated with the maximum PD amplitude values. Qualitative analysis of PRPD pattern was principally used to correlate that the value of the maximum PD amplitude given by the PD instrument corresponds well to the PD activity measured on each bar and coil. The results presented in this paper reveal that the PRPD patterns obtained from all the instruments are generally consistent and that variability in the maximum PD amplitude from one instrument to another can be lowered by specifying experimental conditions limits. In this study, PRPD pattern analysis was used to correlate the maximum PD amplitude value with the contribution of various factors such as the sudden manifestation of new PD source the effect of the conditioning period and the capacitive coupler value.