Locating High-Impedance Faults in DC Microgrid Clusters Using Support Vector Machines

Navid Bayati*, Ebrahim Balouji, Hamid Reza Baghaee, Amin Hajizadeh, Mohsen Soltani, Zhengyu Lin, Mehdi Savaghebi

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

With the increasing number of DC microgrids, DC microgrid clusters are emerging as a cost-effective solution. Therefore, due to the possible long distances between DC microgrids, once a fault occurs and is cleared, it should be located. Especially, locating high impedance faults (HIFs) is challenging. With communication-free fault locating methods, implementation costs can be reduced, and noise and delay of communication can be eliminated. In this paper, a novel localized fault location method using support vector machines (SVMs) is proposed for DC microgrid clusters. The purpose of this study is to facilitate the post fault conditions by locating the accurate place of the faults, even the challenging HIFs, by using the local measurements at one end of each line. The proposed scheme applies the faults, and fault features generated experimentally to the SVM, which is trained in Python for determining the fault location. The experimental test results prove that the proposed scheme is immune against disturbances, such as noise and bad calibration, and can efficiently and reliably estimate the location and resistance of faults with high accuracy.
Original languageEnglish
Article number118338
JournalApplied Energy
Volume308
Number of pages11
ISSN0306-2619
DOIs
Publication statusPublished - 15. Feb 2022

Keywords

  • Clusters
  • DC Microgrid
  • Fault
  • SVM

Fingerprint

Dive into the research topics of 'Locating High-Impedance Faults in DC Microgrid Clusters Using Support Vector Machines'. Together they form a unique fingerprint.

Cite this