Catalog Home Page

Investigation of propagation of partial discharges in power transformers and techniques for locating the discharge

Hettiwatte, S.N., Wang, Z.D. and Crossley, P.A. (2005) Investigation of propagation of partial discharges in power transformers and techniques for locating the discharge. IEE Proceedings - Science, Measurement and Technology, 152 (1). pp. 25-30.

[img]
Preview
PDF - Published Version
Download (704kB)
Link to Published Version: http://dx.doi.org/10.1049/ip-smt:20050944
*Subscription may be required

Abstract

The location of partial discharges in a power transformer can be determined based on the characteristics of the transfer function from the discharge source to the measuring terminal. Previous studies partially validated the technique using computer simulation and practical experiments based on a PD calibrator to represent a discharge signal. 'Real' discharges produced by insulation defect models were used to study how discharges propagate in an 11 kV plain-disc-type transformer winding: A corona model and a 'floating objects in oil' model produced discharge signals with different durations at various locations along the winding. Measurements were taken at the tap of the bushing capacitance through a conventional discharge detector. The signals were filtered, amplified and fed into a digital storage oscilloscope. The frequency spectra of the measured signals showed significant similarities irrespective of the type of discharge source. The characteristic of the transfer functions, i.e. the crests and troughs in the spectra, could be used for locating the source of the discharge. Energising the transformer increased the level of electric noise, which did affect the low frequency end of the spectra, but did not have any impact on the characteristics used for location.

Publication Type: Journal Article
Publisher: Institute of Electrical Engineers
Copyright: © IEE, 2005
URI: http://researchrepository.murdoch.edu.au/id/eprint/14157
Item Control Page Item Control Page

Downloads

Downloads per month over past year