Classification of GIS Based on Adaptive Neural-Fuzzy Reasoning System

Partial discharge (PD) measurement is one of the most important diagnostic methods for the insulation system of high voltage equipment, which is convenient to evaluate the insulation state. Partial discharge activity may originate from various defects and exhibit different behaviors accordingly. Here, three PD patterns generated by different laboratory models representing GIS defects are recorded and analyzed. The purpose of this study was to conduct PD test with three GIS devices including prefabrication defects. statistical features are extracted from PD pattern data and reduced by linear discriminant analysis (LDA). the adaptive neural fuzzy inference system (ANFIS) was used to train the fuzzy inference system (fis). the proportion of trained fis used for ANFIS classification was as high as 95.83%.