An Intelligent Backup Scheme for Current Source Converter-High Voltage Direct Current Systems Based on Artificial Neural Networks

Ricardo C. Santos, Simon Le Blond, Denis V. Coury, Raj K. Aggarwal

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This paper presents an intelligent and integrated backup scheme based on artificial neural networks for current source converter-high voltage direct current (CSC-HVDC) systems. Taking advantage of the properties of ANNs to identify and classify patterns, the proposed scheme is able to detect and correctly locate a fault occurring either at the rectifier substation, the DC line or at the inverter substation. In this scheme, only local signals are used at the rectifier substation and no communication link is necessary, thus improving the system's protection reliability and reducing the required capital cost. Moreover, the proposed scheme could be promptly applicable in practical situations, as the hardware requirements and response times are not critical issues. All steps of the proposed scheme for the aforementioned purpose are accurately discussed in addition to all the information about the CSC-HVDC system under analysis. Finally, a detailed analysis of the influence of the main fault parameters on the algorithm's performance is also conducted.

Original languageEnglish
Pages (from-to)1892-1904
Number of pages13
JournalElectric Power Components and Systems
Volume45
Issue number17
DOIs
Publication statusPublished - 21 Oct 2017

Keywords

  • Artificial neural networks
  • backup protection
  • current source converter
  • decision support tool
  • digital relay
  • fault detection
  • HVDC transmission
  • intelligent system
  • pattern recognition
  • power system protection

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Mechanical Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'An Intelligent Backup Scheme for Current Source Converter-High Voltage Direct Current Systems Based on Artificial Neural Networks'. Together they form a unique fingerprint.

  • Cite this