Classification of partial discharge sources using statistical approach

In high-voltage (HV) power equipment, degradation of insulation has been main concern for protection of equipment. This is due to occurrence of partial discharges (PD) activity within HV insulating systems which can be initiated from different types of local internal or external defects. Thus, partial discharge (PD) identification and classification are important for diagnostic insulation systems problems in order to ensure maintenance process can be carried out effectively and hence improve reliability and durable operation of HV equipment. In this work, the relation of the observable statistical characteristics from PD data with the characteristic of the defect is an important factor to determine the defect inside insulation system. Ultimately, the statistical parameters obtained from PD data can be used to classify different PD sources occur inside HV insulation system. Thus, the objective of this paper is to produce a unique pattern according to discharge source using statistical method. Several statistical parameters such as mean, variance, standard deviation, skewness and kurtosis have been used and analysed.