A linear approach to determining an SVM-based fault locator's optimal parameters

The setting up process for support vector machines (SVM) is discussed in this paper. Such settings are normally obtained from exhaustive testing of SVM, settled on by using several configuration parameter values and evaluating performance by using techniques such as cross validation. The linear appr...

Full description

Saved in:
Bibliographic Details
Published in:Ingeniería e investigación 2009, Vol.29 (1), p.76-81
Main Authors: PEREZ HERNANDEZ, Lucas, MORA FLOREZ, Juan, BEDOYA CEBAYOS, Juan
Format: Article
Language:English
Subjects:
Applied sciences
Computer science
control theory
systems
Data processing. List processing. Character string processing
ENGINEERING, MULTIDISCIPLINARY
euclidean distance norm
Exact sciences and technology
fault location
linear programming
localización de fallas
Memory organisation. Data processing
máquinas de soporte vectorial
norma de la distancia euclidiana
programación lineal
Software
support vector machine
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The setting up process for support vector machines (SVM) is discussed in this paper. Such settings are normally obtained from exhaustive testing of SVM, settled on by using several configuration parameter values and evaluating performance by using techniques such as cross validation. The linear approach presented in this paper is based on redefining the classical SVM second-order objective function. Better setting parameters were obtained by using low computational cost methodology for resolving new linear optimisation. The proposed approach was applied to a typical classification problem regarding fault location in power distribution systems; the results so obtained were compared to those obtained using classical methodology. An 80% improvement was achieved in mean error when estimating fault location and 56% reduction in the computing time needed for obtaining the best results when using classical approaches.
ISSN:0120-5609
2248-8723
2248-8723
DOI:10.15446/ing.investig.v29n1.15146