Phase Resolved Partial Discharge Identification using a Support Vector Machine

Partial discharge (PD) has a significant effect on the insulation performance of power apparatus in both transmission and distribution networks of power systems. Insulation performance and properties can be influenced by different types of PD activity. Therefore, PD source identification and diagnosis is of interest to both power equipment Manufacturers and utilities. With developments in measurement techniques, sensors and signal processing techniques, interpretation of the measured PD data and PD source identification are gaining more interest. Over the last two decades, research into computer-aided automatic PD source discrimination has attracted great attention. A number of papers have been published based on the use of artificial intelligence algorithms such as artificial neural networks, genetic algorithms and fuzzy logic. This paper investigates the application of a machine learning technique, namely the support vector machine (SVM) on PD source identification using phase resolved discharge distribution information (' – average q). PD data obtained from a conventional PD detector and a non-conventional radio frequency current transducer were used to assess the performance of the use of a phase resolved parameter for identification