Outdoor Insulator Surface Condition Evaluation using Image Classification

D. Pernebayeva, D. Sadykova, A. P. James, M. Bagheri

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Outdoor high voltage insulators play a key role in power transmission systems. Due to adverse meteorological conditions, the surface of outdoor overhead insulators frequently covered with snow and ice and exposed to mechanical and electrical stress. This condition leads to insulation breakdowns and power transmission line outages. This paper studies the surface of glass insulator covered with snow, ice and wet condition using image processing techniques. Image filtering methods and state-of-the-art classification algorithms are utilized to discriminate the states of insulator surface using real images of high voltage glass insulator. The obtained results showed that four different insulator surface conditions classified correctly with the accuracy of 99%. An Unnamed Aerial Vehicle (UAV) application is proposed as a prospective data acquisition system.

Original languageEnglish
Title of host publicationICHVE 2018 - 2018 IEEE International Conference on High Voltage Engineering and Application
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538650868
ISBN (Print)9781538650868
DOIs
Publication statusPublished - Feb 13 2019
Event2018 IEEE International Conference on High Voltage Engineering and Application, ICHVE 2018 - Athens, Greece
Duration: Sep 10 2018Sep 13 2018

Publication series

NameICHVE 2018 - 2018 IEEE International Conference on High Voltage Engineering and Application

Conference

Conference2018 IEEE International Conference on High Voltage Engineering and Application, ICHVE 2018
CountryGreece
CityAthens
Period9/10/189/13/18

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Outdoor Insulator Surface Condition Evaluation using Image Classification'. Together they form a unique fingerprint.

Cite this