Partial discharge detection in cable systems
Partial discharge detection in cable systems
This thesis is concerned with the development of suitable methods for partial discharge (PD) detection, quantification and location within high voltage cable systems. Acoustic emission (AE) techniques are suitable for PD on-line monitoring. This approach is free from electrical interference. Application of acoustic emission techniques to PD detection in polymeric cable insulation has been investigated using a planar experimental model. The obtained AE signals were processed in the time and frequency domains, as well as using statistical patterns and operators. Factors influencing AE signal measurement and attenuation have been investigated. The relationship between the discharge acoustic emission and electric signals has been evaluated. AE measurements have been used to investigate PD behaviour during electrical tree growth with cable insulation. Artificial neural networks (ANN) were used to characterise AE signals. The frequency spectrum obtained by short duration Fourier transform of AE signal and the wavelet decomposition coefficients were used as ANN input parameters. Both the feed forward neural network using the back propagation algorithm and the Kohonen self organising map neural network using the learning vector quantization algorithm were applied. Results indicate that ANN can identify different PD sources effectively.
The development of capacitive coupler techniques for PD on-line detection has been investigated. High deletion sensitivity can be obtained and PD site location can be realised by evaluating the time of flight from more than one coupler. A cross-correlation algorithm was implemented to automatically estimate the time of flight. An alternative PD test method for long cable/joint system has been established. This method reduces the power supply requirement of the test system.
University of Southampton
Tian, Yuan
dfe3fbc6-fc51-4b82-87f5-d44715faf7c6
2001
Tian, Yuan
dfe3fbc6-fc51-4b82-87f5-d44715faf7c6
Tian, Yuan
(2001)
Partial discharge detection in cable systems.
University of Southampton, Doctoral Thesis.
Record type:
Thesis
(Doctoral)
Abstract
This thesis is concerned with the development of suitable methods for partial discharge (PD) detection, quantification and location within high voltage cable systems. Acoustic emission (AE) techniques are suitable for PD on-line monitoring. This approach is free from electrical interference. Application of acoustic emission techniques to PD detection in polymeric cable insulation has been investigated using a planar experimental model. The obtained AE signals were processed in the time and frequency domains, as well as using statistical patterns and operators. Factors influencing AE signal measurement and attenuation have been investigated. The relationship between the discharge acoustic emission and electric signals has been evaluated. AE measurements have been used to investigate PD behaviour during electrical tree growth with cable insulation. Artificial neural networks (ANN) were used to characterise AE signals. The frequency spectrum obtained by short duration Fourier transform of AE signal and the wavelet decomposition coefficients were used as ANN input parameters. Both the feed forward neural network using the back propagation algorithm and the Kohonen self organising map neural network using the learning vector quantization algorithm were applied. Results indicate that ANN can identify different PD sources effectively.
The development of capacitive coupler techniques for PD on-line detection has been investigated. High deletion sensitivity can be obtained and PD site location can be realised by evaluating the time of flight from more than one coupler. A cross-correlation algorithm was implemented to automatically estimate the time of flight. An alternative PD test method for long cable/joint system has been established. This method reduces the power supply requirement of the test system.
Text
786504.pdf
- Version of Record
More information
Published date: 2001
Identifiers
Local EPrints ID: 464341
URI: http://eprints.soton.ac.uk/id/eprint/464341
PURE UUID: 9ff19c36-3708-43bb-864a-2fdd49fa828d
Catalogue record
Date deposited: 04 Jul 2022 22:18
Last modified: 16 Mar 2024 19:26
Export record
Contributors
Author:
Yuan Tian
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics