Design and Implementation of an Intelligent Single Line to Ground Fault Locator for Distribution Feeders

Abdulaziz Aljohani, Turki Sheikhoon, Abdulaziz Fataa, Md Hossain, Mohamed Yousif

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

Power quality disturbances became a major issue in modern commercial distribution grids; hence, an innovative attempt to diagnose the faults is necessary for their optimal management. This paper presents a hybrid approach using Stockwell Transform (ST) and Machine Learning Techniques (MLT) to detect, classify and locate the single-line-to-ground (SLG) faults in a modeled distribution feeder to a laboratory scale. The three-phase current signals are processed with ST to extract useful features whereas the Machine Learning Techniques (MLT) including the Multilayer Perceptron Neural Network (MLP-NN) and Extreme Learning Machine (ELM) were employed to diagnose the SLG faults based on the extracted features. The presented results proven to be highly satisfactory for the proposed hybrid SLG fault diagnosis methodology under the presence of measurement noise and load demand uncertainties.

Original languageEnglish
Title of host publication2019 International Conference on Control, Automation and Diagnosis, ICCAD 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728122922
DOIs
StatePublished - Jul 2019
Externally publishedYes
Event3rd International Conference on Control, Automation and Diagnosis, ICCAD 2019 - Grenoble, France
Duration: 2 Jul 20194 Jul 2019

Publication series

Name2019 International Conference on Control, Automation and Diagnosis, ICCAD 2019 - Proceedings

Conference

Conference3rd International Conference on Control, Automation and Diagnosis, ICCAD 2019
Country/TerritoryFrance
CityGrenoble
Period2/07/194/07/19

Keywords

  • Distribution Networks
  • Extreme learning machine
  • Fault diagnosis
  • Machine learning
  • Multilayer perceptron neural network
  • Stockwell transform

Fingerprint

Dive into the research topics of 'Design and Implementation of an Intelligent Single Line to Ground Fault Locator for Distribution Feeders'. Together they form a unique fingerprint.

Cite this