Desarrollo de un sistema para detección y localización de cadenas de aisladores dañadas a partir de señales registradas en subestaciones

• Autor(a) principal: Capelini, Renato Massoni
• Data de Publicação: 2015
• Outros Autores: Adami, José Feliciano [UNESP], Martinez, Manuel Luis Barreira, Da Costa Parentoni, Marcel Fernando
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.15446/ing.investig.v35n1.41832; http://hdl.handle.net/11449/177348
• topic: Current transformer; Insulator strings; Neural network; Transmission line

This paper presents a novel technique for remotely identifying and locating of damaged insulator strings based on an analysis of noise signals recorded on the ground connection of current transformers installed on transmission line substations. Damaged insulators usually present electrical surface activity. Surface and corona discharges may result in the generation of high frequency signals that can travel through the transmission line conductors, which are recorded by a data acquisition system. Each signal generated by the surface or corona discharge has a unique signature. This signature changes according to the position of the source, i.e., the damaged string along the line, which enables the association between the signal signature and the location of the string along the line. A method proposed to achieve this association is discussed in this paper as well as the tools and tests carried out during its development. An innovative way for capacitive coupling, which reduce detection system implementation costs is presented.