Scientific Research and Essays

  • Abbreviation: Sci. Res. Essays
  • Language: English
  • ISSN: 1992-2248
  • DOI: 10.5897/SRE
  • Start Year: 2006
  • Published Articles: 2767

Full Length Research Paper

Support vector regression and rule based classifier comparison for power quality diagnosis

Azah Mohamed*, Mohamed Fuad Faisal and Hussain Shareef
Department of Electrical, Electronic and Systems, University Kebangsaan Malaysia.
Email: [email protected]

  •  Accepted: 18 October 2011
  •  Published: 16 January 2012

Abstract

This paper presents a comparative study for performing automated power quality diagnosis using rule base classifier (RBC) and support vector regression (SVR) to identify the causes of short duration voltage disturbances such as voltage sag and swell. In the proposed power quality diagnosis method, a time frequency analysis technique called the S-transform was used to analyse and extract features of voltage disturbances recorded from the power quality monitoring system. The RBC and SVR which are intelligent techniques were then used to identify whether the voltage disturbances were caused by permanent, non-permanent transient or incipient faults. Test results proved that the RBC performed better than the SVR in diagnosing the causes of short duration voltage disturbances.

 

Key words: Power quality diagnosis, support vector regression, s-transform.