Support vector regression and rule based classifier comparison for automated classification of power quality disturbances

Azah Mohamed, Mohamed Fuad Faisal

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The ability to detect and classify all the types of power quality disturbances encrypted in a voltage signal is vital in the analyses of identifying the causes of such disturbances. In this paper, a comparative study is conducted using support vector regression (SVR) and rule based classifier (RBC) for automatically classifying power quality disturbances in a three-phase industrial power system. The S-transform is used to extract features of the power quality disturbances which are mostly non-stationary and comprise of multiple disturbances that coexist together for only a short duration in time. The performances of the SVR and the RBC were compared for their abilities in making prediction for the classes of power quality disturbances. The results for the analyses on 744 numbers of single and multiple voltage disturbances gave classification accuracies of 92.5% (SVR) and 100% (RBC), respectively.

Original languageEnglish
Pages (from-to)2754-2761
Number of pages8
JournalInternational Review of Electrical Engineering
Volume6
Issue number6
Publication statusPublished - 2011

Fingerprint

Power quality
Classifiers
Electric potential
Mathematical transformations

Keywords

  • Disturbance Classification
  • Power Quality
  • Rule Based Classifier
  • S-transform
  • Support Vector Regression

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Support vector regression and rule based classifier comparison for automated classification of power quality disturbances. / Mohamed, Azah; Faisal, Mohamed Fuad.

In: International Review of Electrical Engineering, Vol. 6, No. 6, 2011, p. 2754-2761.

Research output: Contribution to journalArticle

@article{826fc2bd7637430ab546b515db5bedc8,
title = "Support vector regression and rule based classifier comparison for automated classification of power quality disturbances",
abstract = "The ability to detect and classify all the types of power quality disturbances encrypted in a voltage signal is vital in the analyses of identifying the causes of such disturbances. In this paper, a comparative study is conducted using support vector regression (SVR) and rule based classifier (RBC) for automatically classifying power quality disturbances in a three-phase industrial power system. The S-transform is used to extract features of the power quality disturbances which are mostly non-stationary and comprise of multiple disturbances that coexist together for only a short duration in time. The performances of the SVR and the RBC were compared for their abilities in making prediction for the classes of power quality disturbances. The results for the analyses on 744 numbers of single and multiple voltage disturbances gave classification accuracies of 92.5{\%} (SVR) and 100{\%} (RBC), respectively.",
keywords = "Disturbance Classification, Power Quality, Rule Based Classifier, S-transform, Support Vector Regression",
author = "Azah Mohamed and Faisal, {Mohamed Fuad}",
year = "2011",
language = "English",
volume = "6",
pages = "2754--2761",
journal = "International Review of Electrical Engineering",
issn = "1827-6660",
publisher = "Praise Worthy Prize",
number = "6",

}

TY - JOUR

T1 - Support vector regression and rule based classifier comparison for automated classification of power quality disturbances

AU - Mohamed, Azah

AU - Faisal, Mohamed Fuad

PY - 2011

Y1 - 2011

N2 - The ability to detect and classify all the types of power quality disturbances encrypted in a voltage signal is vital in the analyses of identifying the causes of such disturbances. In this paper, a comparative study is conducted using support vector regression (SVR) and rule based classifier (RBC) for automatically classifying power quality disturbances in a three-phase industrial power system. The S-transform is used to extract features of the power quality disturbances which are mostly non-stationary and comprise of multiple disturbances that coexist together for only a short duration in time. The performances of the SVR and the RBC were compared for their abilities in making prediction for the classes of power quality disturbances. The results for the analyses on 744 numbers of single and multiple voltage disturbances gave classification accuracies of 92.5% (SVR) and 100% (RBC), respectively.

AB - The ability to detect and classify all the types of power quality disturbances encrypted in a voltage signal is vital in the analyses of identifying the causes of such disturbances. In this paper, a comparative study is conducted using support vector regression (SVR) and rule based classifier (RBC) for automatically classifying power quality disturbances in a three-phase industrial power system. The S-transform is used to extract features of the power quality disturbances which are mostly non-stationary and comprise of multiple disturbances that coexist together for only a short duration in time. The performances of the SVR and the RBC were compared for their abilities in making prediction for the classes of power quality disturbances. The results for the analyses on 744 numbers of single and multiple voltage disturbances gave classification accuracies of 92.5% (SVR) and 100% (RBC), respectively.

KW - Disturbance Classification

KW - Power Quality

KW - Rule Based Classifier

KW - S-transform

KW - Support Vector Regression

UR - http://www.scopus.com/inward/record.url?scp=84855735660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84855735660&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84855735660

VL - 6

SP - 2754

EP - 2761

JO - International Review of Electrical Engineering

JF - International Review of Electrical Engineering

SN - 1827-6660

IS - 6

ER -