Automatic high voltage data analysis tools using mathematical techniques and AI

Williams, Iwan and Albano, Maurizio ORCID: https://orcid.org/0000-0002-5486-4299 2023. Automatic high voltage data analysis tools using mathematical techniques and AI. Presented at: 58th International Universities Power Engineering Conference (UPEC), Dublin, Ireland, 30 August - 1 September 2023. Proceedings of 58th International Universities Power Engineering Conference. IEEE, pp. 1-6. 10.1109/UPEC57427.2023.10294509

Full text not available from this repository.

Abstract

This paper investigates the methods used in analysing performance of high voltage insulators in HV laboratory tests. The performance of high voltage insulators requires analysis for the purposes of design improvement and comparison. The analysis can be supported using visual and infra-red cameras to record the insulator under stress in addition to the electrical parameters such as leakage current and voltage signals. A new procedure based on mathematical and AI techniques was developed to create a new engineering tool as an alternative to the current ones in use. Visual and infra-red data from high voltage insulator tests done at HV laboratory at Cardiff University were obtained and the developed MATLAB code was tested on these data. The identification of an insulator within an image were investigated and a new method based on normalised 2-dimensional cross-correlation was developed. The developed visual analysis methods investigate the colours present in the video along with artificial long exposure technique. The infrared investigation led to an alternative method of calculating the temperature profile of an insulator and investigates the use of deep learning in dry band detection. The developed methods have been successfully applied on previously recorded data and they offer useful tools to investigate the performance of insulators. The discussion of deep learning opens further opportunities for the use of deep learning in dry band detection and classification.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Engineering
Publisher:	IEEE
ISBN:	9798350316841
Last Modified:	18 Dec 2023 11:15
URI:	https://orca.cardiff.ac.uk/id/eprint/164700

Actions (repository staff only)

Edit Item