UPFC supplementary damping control synthesis: A forward neural networks approximated energy function approach

Hung Chi Tsai, Chia Chi Chu

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

3 Scopus citations

Abstract

This paper aims to develop supplementary damping control actions of UPFCs by using forward neural networks (FNN) approximated energy function methods. The energy function based approach will first be considered for designing the supplementary damping control based on the network-preserving model. In order to provide more flexible solution, we proposed the Energy Function based Forward Neural Network Supplementary Damping (EFFNNSD) controller of UPFCs. The EFFNNSD can approximate the supplementary control action based on energy function methods. The EFFNNDS not only retain the idea of energy function methods, but also has powerful on-line learning ability for damping control adjustments. Numerical simulations on two benchmark systems have been performed to validate the proposed control for providing the extra damping and suppressing power swings even under severe operating conditions.

Original languageEnglish
Title of host publication2018 IEEE Industry Applications Society Annual Meeting, IAS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538645369
DOIs
StatePublished - 26 11 2018
Externally publishedYes
Event2018 IEEE Industry Applications Society Annual Meeting, IAS 2018 - Portland, United States
Duration: 23 09 201827 09 2018

Publication series

Name2018 IEEE Industry Applications Society Annual Meeting, IAS 2018

Conference

Conference2018 IEEE Industry Applications Society Annual Meeting, IAS 2018
Country/TerritoryUnited States
CityPortland
Period23/09/1827/09/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE

Keywords

  • Damping control
  • Energy functions
  • Forward neural networks
  • UPFC

Fingerprint

Dive into the research topics of 'UPFC supplementary damping control synthesis: A forward neural networks approximated energy function approach'. Together they form a unique fingerprint.

Cite this