Support vector machines for on-line security analysis of power systems

Author

Cortés-Carmona, M. ; Jiménez-Estévez, G. ; Guevara-Cedeno, J.

Author_Institution

Dept. of Electr. Eng., Univ. of Antofagasta, Antofagasta

fYear

2008

fDate

13-15 Aug. 2008

Firstpage

1

Lastpage

6

Abstract

The pattern recognition approach for security analysis (SA) of power systems has been presented as a promising tool for on-line applications. This paper applies a learning-based nonlinear classifier, which is a support vector machine (SVM) for SA. Three single SVM are trained to classify the state of the system: secure, alert and emergency. The final classification is obtained combining the output of each classifier with a Bayesian rule. The effectiveness of the proposed approach has been demonstrated on two IEEE test systems.

Keywords

Bayes methods; pattern recognition; power engineering computing; power system security; support vector machines; Bayesian rule; IEEE test system; learning-based nonlinear classifier; on-line security analysis; pattern recognition approach; power systems; support vector machine; Decision support systems; Pattern recognition; Power system analysis computing; Power system dynamics; Power system security; Power system stability; Power system transients; Support vector machine classification; Support vector machines; Voltage; Neural networks; security assessment; support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Transmission and Distribution Conference and Exposition: Latin America, 2008 IEEE/PES

Conference_Location

Bogota

Print_ISBN

978-1-4244-2217-3

Electronic_ISBN

978-1-4244-2218-0

Type

conf

DOI

10.1109/TDC-LA.2008.4641770

Filename

4641770