Comparative study of partial discharge localisation techniques within transformer windings

Partial discharge (PD) activity may occur inside power transformers and the existence of this activity produces a negative effect on the transformer until eventually there is catastrophic failure. Therefore, partial discharge detection and measurement within a transformer winding have become an important research area to provide asset health information enabling the maintenance and replacement processes to be carried out effectively. A modern condition monitoring system for PD activity inside HV transformers may use any observable phenomena but here we concentrate on an electrical detection method. The fundamental localization technique uses wavelet multi-resolution analysis for calculating the energy at different frequency ranges and Principal Component Analysis as dimensional reduction tool that produces the output that ultimately can be used as an indicator of PD location inside a transformer winding. However, an alternative method has been developed to using corresponding 3 dimensional filters to enable the localisation process. Both approaches are compared and obtained results are presented in this paper.