Fast learning algorithm for synchronous generator instability prediction

George G. Karady, Ahmed A. Daoud, Mansour A. Mohammed, Ibrahim A. Amin

Research output: Contribution to journalArticle

Abstract

This paper proposes a new method of power system instability prediction. The proposed method does not require any prior knowledge of the power system parameters. After fault clearing in the power system the generators' relative rotor angles are measured for about 0.3-0.5 second and the variation of each generator rotor angle is predicted for the next 0.5-2 second period using the developed algorithm and the measured data. If a generator rotor angle is larger than an experimentally determined threshold, the system is considered unstable and remedial actions are initiated. The WECC test system is used to prove the validity of the proposed method for generator relative rotor angle prediction. The study indicated that the prediction error is less than 10%.

Original languageEnglish (US)
Pages (from-to)67-72
Number of pages6
JournalEngineering Intelligent Systems
Volume12
Issue number2
StatePublished - Jun 2004

Fingerprint

Synchronous generators
Learning algorithms
Rotors

Keywords

  • Fast learning algorithm
  • Generator instability prediction
  • Power systems
  • WECC test system

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Karady, G. G., Daoud, A. A., Mohammed, M. A., & Amin, I. A. (2004). Fast learning algorithm for synchronous generator instability prediction. Engineering Intelligent Systems, 12(2), 67-72.

Fast learning algorithm for synchronous generator instability prediction. / Karady, George G.; Daoud, Ahmed A.; Mohammed, Mansour A.; Amin, Ibrahim A.

In: Engineering Intelligent Systems, Vol. 12, No. 2, 06.2004, p. 67-72.

Research output: Contribution to journalArticle

Karady, GG, Daoud, AA, Mohammed, MA & Amin, IA 2004, 'Fast learning algorithm for synchronous generator instability prediction', Engineering Intelligent Systems, vol. 12, no. 2, pp. 67-72.
Karady, George G. ; Daoud, Ahmed A. ; Mohammed, Mansour A. ; Amin, Ibrahim A. / Fast learning algorithm for synchronous generator instability prediction. In: Engineering Intelligent Systems. 2004 ; Vol. 12, No. 2. pp. 67-72.
@article{76de2569c9a94736b4abc04688a18ebf,
title = "Fast learning algorithm for synchronous generator instability prediction",
abstract = "This paper proposes a new method of power system instability prediction. The proposed method does not require any prior knowledge of the power system parameters. After fault clearing in the power system the generators' relative rotor angles are measured for about 0.3-0.5 second and the variation of each generator rotor angle is predicted for the next 0.5-2 second period using the developed algorithm and the measured data. If a generator rotor angle is larger than an experimentally determined threshold, the system is considered unstable and remedial actions are initiated. The WECC test system is used to prove the validity of the proposed method for generator relative rotor angle prediction. The study indicated that the prediction error is less than 10{\%}.",
keywords = "Fast learning algorithm, Generator instability prediction, Power systems, WECC test system",
author = "Karady, {George G.} and Daoud, {Ahmed A.} and Mohammed, {Mansour A.} and Amin, {Ibrahim A.}",
year = "2004",
month = "6",
language = "English (US)",
volume = "12",
pages = "67--72",
journal = "International Journal of Engineering Intelligent Systems for Electrical Engineering and Communications",
issn = "0969-1170",
publisher = "CRL Publishing",
number = "2",

}

TY - JOUR

T1 - Fast learning algorithm for synchronous generator instability prediction

AU - Karady, George G.

AU - Daoud, Ahmed A.

AU - Mohammed, Mansour A.

AU - Amin, Ibrahim A.

PY - 2004/6

Y1 - 2004/6

N2 - This paper proposes a new method of power system instability prediction. The proposed method does not require any prior knowledge of the power system parameters. After fault clearing in the power system the generators' relative rotor angles are measured for about 0.3-0.5 second and the variation of each generator rotor angle is predicted for the next 0.5-2 second period using the developed algorithm and the measured data. If a generator rotor angle is larger than an experimentally determined threshold, the system is considered unstable and remedial actions are initiated. The WECC test system is used to prove the validity of the proposed method for generator relative rotor angle prediction. The study indicated that the prediction error is less than 10%.

AB - This paper proposes a new method of power system instability prediction. The proposed method does not require any prior knowledge of the power system parameters. After fault clearing in the power system the generators' relative rotor angles are measured for about 0.3-0.5 second and the variation of each generator rotor angle is predicted for the next 0.5-2 second period using the developed algorithm and the measured data. If a generator rotor angle is larger than an experimentally determined threshold, the system is considered unstable and remedial actions are initiated. The WECC test system is used to prove the validity of the proposed method for generator relative rotor angle prediction. The study indicated that the prediction error is less than 10%.

KW - Fast learning algorithm

KW - Generator instability prediction

KW - Power systems

KW - WECC test system

UR - http://www.scopus.com/inward/record.url?scp=4344616319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=4344616319&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:4344616319

VL - 12

SP - 67

EP - 72

JO - International Journal of Engineering Intelligent Systems for Electrical Engineering and Communications

JF - International Journal of Engineering Intelligent Systems for Electrical Engineering and Communications

SN - 0969-1170

IS - 2

ER -