Stochastic analysis of partial discharges signals for diagnosis of insulation systems of AC rotating machine stator windings

This paper deals with fundamentals of diagnosis of the insulation system of AC rotating machine stator windings by means of the stochastic analysis of partial discharge (PD) signals. It is shown that the stochastic features of random pulses, generated by PD phenomena, can considerably help in PD inference. A new measuring system for the digital acquisition of PD-pulse signals, which operates at a sampling rate high enough to avoid the frequency aliasing but that provides an amount of PD pulses which enables PD stochastic analysis, has been designed and the prototype performances are presented. Results of PD measurements performed on different types of complete stator bars, with and without defects, are reported. The purpose of these investigations is to aid user decision as regards leaving or removing the machine from the service at a convenient time before failure and, eventually, to guide the repair