Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.527122
Title: Condition monitoring of outdoor insulation using artificial intelligence techniques
Author: Crespo-Sandoval, J.
Awarding Body: Cardiff University
Current Institution: Cardiff University
Date of Award: 2005
Availability of Full Text:
Access through EThOS:
Abstract:
The work reported in this thesis is concerned with the application of artificial intelligence to monitoring of outdoor insulation. The work comprised a comprehensive literature survey, the development of computerised systems for capturing, storing and processing data recorded in laboratory and field tests. Extensive programmes of pollution tests on insulating material samples and complete outdoor insulators have been carried out, and the results were analysed using an artificial intelligence technique. In addition, existing long-term field data from a natural pollution testing station have been analysed and classified. The extensive literature survey reviewed the mechanisms causing degradation and failure of insulators, techniques for monitoring insulator degradation and the application of artificial intelligence techniques to their condition monitoring. The data acquisition systems were designed to interface with existing accelerated ageing unit and fog chamber facilities and to capture and store large quantities of leakage current data. Analysis of the laboratory test results on silicone rubber samples by means of the proposed artificial intelligence technique enabled certain types of leakage current waveshapes to be identified that were related to the extent of insulator degradation. Based on the results, a new technique was proposed for monitoring polymeric insulators and predicting imminent failure. Further analysis of the tests results has revealed that the rate of increase of accumulated energy can be used as an indicator of imminent insulator failure and this result is new and has not been published before to the author's knowledge. Clean fog tests were performed on polluted insulators and the results analysed using the artificial intelligence technique. The effects of increasing insulator degradation, pollution severity and applied voltage were investigated. By applying a normalisation procedure, it was possible to apply the monitoring technique developed on insulator samples, and it was demonstrated that the technique can distinguish good insulators from those that have been subjected to severe degradation levels. A new analysis technique was developed to convert existing field data into an easily accessible format, to perform a diagnostic analysis of the data in order to indicate imminent insulator failure and to act as a user-fiiendly interface for insulator monitoring. A computer programme was developed which incorporated the field data analysis and diagnostic procedure.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.527122  DOI: Not available
Share: