Detection and classification of defects in ceramic insulators using RF antenna

Abstract

Defects in ceramic insulators like broken, cracked and punctured discs give rise to the initiation of partial discharge (PD) activities within the samples which has a detrimental effect on the insulator life. Hence it is important for the utilities to identify such defective samples as early as possible so that appropriate replacement strategies can be devised. The work presented in this paper involves the investigation of a number of cases of insulator defects, with the goal of developing an online RF-based PD technique for monitoring ceramic disc insulators. The three classes examined are a cracked ceramic insulator disc; a disc with a hole through the cap, and a completely broken insulator disc. The defective discs are considered individually and are also incorporated into strings of 2, 3, and 4 insulators. The captured RF pulses are processed by extracting wavelet packet based features. Feature reduction and selection is carried out and classification results are obtained. To classify the discharges arising from different types of defects, an artificial neural network (ANN) algorithm is applied to the extracted features, and recognition rates of more than 95% were reported for each class. The results of preliminary field tests carried out on a 40 feet high test transmission tower are also reported and their analysis showed good discrimination between the different defect types.