Influence of the variation in the load and the size of the zone in the accuracy of a troubles locator for distribution circuits
Published 2007-05-21
Keywords
- Fault location,
- power distribution systems,
- classifiers,
- support vector machines
How to Cite
Abstract
A recent and interesting study topic to utilities engineers and customers has been the electric power quality. The product itself and the customer support are the main aspects considered. The product quality means satisfy requirements of wave quality and service continuity. This last aspect is the one considered when the fault location problem is considered.
This paper shows an analysis of a fault location method applied to power distribution systems, developed with a classifier based in support vectors machines. This fault location trained using fault data from nominal conditions and tested considering the influence of load variations. Also, its capability to maintain high performance indexes where the power system is sub-divided in an increasing number of zones in checked.
From an application example which uses a model taken from a real power system, it is shown how the proposed approach is highly effective to salve the problem, having mean precision scores above 90% to locate the faulted zone, in case of single phase faults to ground, which is the most difficult case proposed.
Downloads
References
Bernhard & Smola (2002). "Learning with Kernels Support Vector Machines, Regularization, Optimization and Beyond'. The MIT Press, Cambridge. 2002.
Bollen (2000). "Understanding Power Quality Problems: Voltage Sags and Interruptions". IEEE Press. 2000.
Boser, Guyon & Vapnik (1992). "A Iraining algorithm for optimal margin classifier". Proccedings of the 5th ACM Workshop on Computational Leaming Theory. Pag:144-152. 1992.
Burges (1998). "A tutorial on support vector machines for pattern recognition". Data Mining and Knoledge Discovery, 1998. Pag: 121-167.
Dagenhart (2000). "The 40- Ground-Fault Phenomenon" IEEE Transactions on Industry Applications, vol. 36, no. 1, pages 30-32, 2000.
Das (1998). "Determining the Locations of Faults in Distribution Systems". Doctoral Tesis, University of Saskatchewan. Saskatoon, Canada. 1998.
IEEE Std C37.114 (2004). "IEEE Guide for Determining Fault Location on AC Transmission and Distribution Lines" Power System Relaying Committee 2004.
Lin, Chang & Hsu (2004). "A Practical Guide to Support Vector Classification". National Taiwan University, 2004.
Mora (2003). "Voltage Sag Characterization and Classification for Diagnosis in Electric Power Quality Domain", Master dissertation. University of Girona, España. 2003.
Mora, Muñoz & Carrillo (2005). "Técnicas algorítmicas de localización de fallas como alternativa para reducir el efecto de las salidas en sistemas de potencia – Una revisión". III Internacional Symposium of Power Quality. Bogotá, Colombia, Nov. 2005.
Mora, Melendez & Bedoya (2006). "Extensive Events Database Development using ATP and Matlab for Fault Location in Power Distribution Systems". In CD proceedings of 2006 IEEE PES Transmission and Distribution Conference and Exposition: Latin America. IEEE Catalog Number 06EX1340C. ISBN 1-4244-0288-3.
Mora, Pérez & Carrillo (2006). "Fault location in power distribution systems using ANFIS nets and current patterns" In CD proceedings of 2006 IEEE PES Transmission and Disttibution Conference and Exposition: Latin America. IEEE Catalog Number 06EX1340C. ISBN 1-4244-0288-3.
Morales & Gómez (2005). "Estudio e implementación de una herramienta basada en máquinas de soporte vectorial aplicada a la localización de fallas en sistemas de distribución". Universidad Industrial de Santander. 2005.
Novosel, Hart & Myllyrnaki (1998). System for locating faults and estimating fault resistance in distribution networks with tapped loads". 1998. US Patent number 5,839,093.
Vapnik (2000). "The nature of Statistical Learning Theory" Second Edition, Springer
Verlag, 2000.