Application of self organizing map approach for partial discharge pattern recognition of insulators

This paper proposes a self organizing map (SOM) based recognition method to identify the insulation defects of electrical apparatus arising from partial discharge (PD). PD patterns are detected by a PD detecting system set up in the laboratory. The significant features of PD patterns are extracted by using the nonlinear principal component analysis (NLPCA) method. Based on the feature extracted, the SOM network is further employed for the PD pattern recognition. To verify the proposed approach, experiments are conducted to demonstrate the field-test PD pattern recognition of insulators by using 250 feature vectors of field-test PD patterns. Five types of models with artificial defects are purposely created to produce five common PD activities of insulators for the experiments. The practical results show that the proposed method is promising as a solution to the PD pattern recognition.