Diagnostics of insulating systems using statistical tools

Computer-aided analysis of partial discharges (PD) using conventional discharge detection was studied. It was found that the shape of the phase-position distributions contains much information on the origin and the type of discharge. These shapes are therefore analyzed, using statistical operators like skewness, kurtosis, cross-correlation factor, etc. Tests of discharges in artificial models in gases and solid materials have shown that several statistical operators are characteristic for actual discharge sources. The combination of 15 statistical operators can form the fingerprint of a discharge. To use these fingerprints for the recognition of unknown defects, an algorithm called recognition rate was developed. This technique was used to reveal the origin of discharges in actual objects such as a current transformer, cable, and gas-insulated switchgear. The promising results for the recognition rate algorithm show that the origin of discharges in technical objects can be traced by using computer-aided discharge analysis