Fault Zone Classifier Performance Improvement of PMU-Enabled Distribution System Through Feature Engineering

Thumbnail Image

Paper number

1998

Working Group Number

Conference name

CIRED 2019

Conference date

3-6 June 2019

Conference location

Madrid, Spain

Peer-reviewed

Yes

Short title

Convener

Authors

Sta. Cruz, Anton Domini, University of the Philippines, Philippines
Pedrasa, Michael Angelo, University of the Philippines, Philippines
Dobbe, Roel, University of California Berkeley, USA

Abstract

Fault localization algorithms are prone to multiple estimates on most distribution networks due to their tree-shaped radial topology. To address this problem, localization may be supported by the identification of network section or zone where the fault occurred. This study aims to explore the development of fault zone classifiers that exploit the phase angle information given by phasor measurement units (PMUs) through feature engineering. Results show that the PMU angle information contributes to better clustering of different fault zones in the feature space which leads to better classification performance. Also, the study demonstrated the effectiveness of further improving the classifier performance through feature engineering. Analysis of a developed fault zone classifier was also performed to understand the characteristics of the misclassified fault instances.

Table of content

Keywords

Publisher

AIM

Date

2019-06-03

Permanent link to this record

https://cired-repository.org/handle/20.500.12455/681
http://dx.doi.org/10.34890/901

ISSN

2032-9644

ISBN

978-2-9602415-0-1