Extension theory based partial discharge pattern recognition using statistical operators

Partial discharges are very small, local dielectric breakdowns in an insulation system. They cause damage to the insulation and leads to the failure of dielectric much before the expected lifetime. PD can be used as diagnostic tool for identifying the defects in power equipment. Two challenging issues are investigated in this work. The first issue is the feature extraction using statistical operator approach for obtaining representative attributes from the original PD measurement data. The second issue is to use the Extension Theory Pattern Recognition Algorithm for identifying various types of PD sources. The proposed method does not include any process learning or tuning of parameters, and it is easily implemented using MATLAB software in this paper.