Coordinated Design of Multiple Robust FACTS Damping Controllers: A BMI-Based Sequential Approach with Multi-Model Systems

Jingchao Deng, Can Li, Xiao Ping Zhang

Research output: Contribution to journalArticlepeer-review

43 Citations (Scopus)

Abstract

This paper presents the coordinated design of multiple FACTS supplementary damping controllers to improve system small-signal stability. A BMI-based multi-objective multi-model system approach is formulated to solve the robust damping control problem. Two-step method is proposed to determine controller variables. Regional pole placement and control effort optimization are set as the control objectives. An SVC and a TCSC damping controller are designed sequentially with minimized decoupling effort. Selection of the feedback signals for the FACTS damping controllers is realized by evaluating the modal residues of each feedback signal to the system input. To cover multiple inter-area oscillation modes, wide-area decentralized control methodology is proposed and remote feedback signals are selected. Real-time simulation is carried out on real-time digital simulator (RTDS) to test the performance of the proposed controller. The results show that the FACTS damping controllers are feasible and able to increase system damping to a satisfactory level under different system operating points.

Original languageEnglish
Article number7024187
Pages (from-to)3150-3159
Number of pages10
JournalIEEE Transactions on Power Systems
Volume30
Issue number6
DOIs
Publication statusPublished - 1 Nov 2015

Keywords

  • BMI
  • coordinated design
  • H<inf>2</inf>
  • H<inf>∞</inf>
  • inter-area oscillations
  • LMI
  • multi-model system
  • multi-objective
  • robust damping control
  • RTDS
  • sequential design
  • SVC
  • TCSC

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Energy Engineering and Power Technology

Fingerprint

Dive into the research topics of 'Coordinated Design of Multiple Robust FACTS Damping Controllers: A BMI-Based Sequential Approach with Multi-Model Systems'. Together they form a unique fingerprint.

Cite this