Partial Discharges (PD) can occur in any insulation medium used in high voltage power equipment. Even though the initial micro level discharges may not affect the breakdown voltage immediately, the sites where PD occur may deteriorate with time and eventually a total failure of equipment may take place. The early detection of partial discharges can prevent major damage to the equipment.Partial Discharge (PD) measurement technique has been in practice for several years for the diagnosis of the insulation problems in High Voltage equipment. The on site PD measurement still remains as a problem to be solved due to the presence of background noise and electromagnetic interference. The on-line PD measurements when realised with acceptable accuracy and the meiiningful interpretation of the PD patterns, can lead to better condition monitoring and prediction of the service life of equipment.The developments in the PD measurement include nonintrusive sensors which enable on-line PD measurements and an efficient use of advanced digitizers. This technology can be further advanced by the application of Artificial Intelligence (AI) and Expert systems, specifically Artificial Neural Networks (ANN), which have already been successfully tried out on problems of diagnostic nature and pattern recognition. This paper explores the issues related to application of Artificial Neural Network technique to PD measurements. These include prospects of the ANNs, the choice of the ANN networks e.g, back propagation, Linear Vector Quantization and input output preparation and interpretation of PD patterns to assess the conditions of insulation. The paper will review some of the recent works in this areii carried out elsewhere along with a report on the experimental developments at Royal Melbourne Institute of Technology (RMIT).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.