Fault location in underground systems through optimum-path forest

• Autor(a) principal: Souza, André N. [UNESP]
• Data de Publicação: 2011
• Outros Autores: Da Costa Jr., Pedro [UNESP], Da Silva, Paulo S. [UNESP], Ramos, Caio C. O., Papa, João Paulo [UNESP]
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1109/ISAP.2011.6082204; http://hdl.handle.net/11449/73076
• topic: Fault Location; Optimum-Path Forest; Pattern Recognition; Underground Systems; Artificial Neural Network; Pattern recognition techniques; Reflectometry; Signal acquisitions; Time domain; Underground distribution system; Underground systems; Electric fault location; Forestry; Intelligent systems; Neural networks; Pattern recognition; Power transmission; Signal processing; Time domain analysis; Algorithms; Classification; Defects; Electric Power Distribution; Forests; Neural Networks

In this paper we propose an accurate method for fault location in underground distribution systems by means of an Optimum-Path Forest (OPF) classifier. We applied the Time Domains Reflectometry method for signal acquisition, which was further analyzed by OPF and several other well known pattern recognition techniques. The results indicated that OPF and Support Vector Machines outperformed Artificial Neural Networks classifier. However, OPF has been much more efficient than all classifiers for training, and the second one faster for classification. © 2011 IEEE.