Diagnosis of Cable Insulation Defects Based on ANFIS–PCA Method

Abstract

Partial discharge (PD) detection is an effective method to detect the condition of power cable insulation. Since different modes of cable defect have different discharge characteristics, the PD detection method is required to be able to collect PD pulse and identify the type of the defect. In this paper, the statistical parameters of partial discharge are extracted and an ANFIS-PCA-based method for diagnosing cable insulation defects is proposed. The ANFIS system is applied to pattern classification of partial discharge. The experimental results show that the method proposed in this paper can classify and identify the various discharge patterns effectively.