A 3D fault location program for use in high-voltage substations using multilateration, geolocation and machine learning

Townsend, William E. and Reid, Alistair J. ORCID: https://orcid.org/0000-0003-3058-9007 2022. A 3D fault location program for use in high-voltage substations using multilateration, geolocation and machine learning. Presented at: 17th International Conference on Probabilistic Methods Applied to Power Systems (PMAPS), Manchester, UK, 12-15 June 2022. Proceeding of International Conference on Probabilistic Methods Applied to Power Systems (PMAPS). IEEE, 10.1109/PMAPS53380.2022.9810631

Full text not available from this repository.

Abstract

This paper proposes a program which calculates the location of a partial discharge fault based on time difference of arrival analysis of collected signal data from a sensor array. An intersecting spherical surface method uses the data and creates a 3D graph, making it easy to see the location. The geolocation feature can be used in areas such as substations to determine the fault location on site, ensuring quick maintenance. The machine learning feature uses k-means clustering to classify signal points to determine a present fault partial discharge signal and the start time. A four-sensor array was setup and used for testing. Results showed a high degree of accuracy.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Engineering
Publisher:	IEEE
ISBN:	9781665412124
Last Modified:	04 Mar 2023 02:21
URI:	https://orca.cardiff.ac.uk/id/eprint/151966

Actions (repository staff only)

Edit Item