Unique localization of faults in distribution systems by means of zones with SVM
This paper presents a new methodology for localizing faults in distribution systems by means of an artificial intelligence technique –Support Vector Machine– (SVM). This methodology divides the electrical system into different zones order to pinpoint the region where the fault exists with accuracy....
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universidad de Antioquia
2013-12-01
|
| Series: | Revista Facultad de Ingeniería Universidad de Antioquia |
| Subjects: | |
| Online Access: | https://revistas.udea.edu.co/index.php/ingenieria/article/view/17765 |
| Summary: | This paper presents a new methodology for localizing faults in distribution systems by means of an artificial intelligence technique –Support Vector Machine– (SVM). This methodology divides the electrical system into different zones order to pinpoint the region where the fault exists with accuracy. The advantage over classical distance methods is the unique estimation of the fault’s locus in branches systems. An example using a real system model shows that the proposed methodology is highly effective finding the fault localization. In such example load changes of ±40 % from nominal load are considered.
|
|---|---|
| ISSN: | 0120-6230 2422-2844 |