Sistema de localização de faltas em redes de distribuição de energia elétrica baseado em sequência de eventos

Abstract

This paper presents the initial results obtained while developing a failures identification program related to nets and distribution lines, denominated LOCFALT. The program was initially created to help Distribution Centers Operation (COD) concerning the interruptions with load loss. A lot of consumers phone calls complaining about the electric power loss are made to the Call Center and automatically transferred to the Distribution Center Operation (COD). Based on the first consumers information, in case of electrical aerial nets, the maintenance technician team workers are sent to inspect, visually, the distribution lines, and then the necessary providences are taken. It is desirable that the Distribution Center Operation (COD), in cases of permanent load loss, be provided by a tool that makes possible the determination of an area that contains the probable load loss place, with precision, optimizing the maintenance technician teams geographic position, avoiding them traveling great extensions, consequently increasing the time to repair the electrical system. The LOCFALT program was developed objectifying agility and efficiency when guiding the maintenance technician teams to the load loss permanents occurrences. It provides a precise indication of an area or more areas that would be included to the possible load loss place. The load loss place identification is accomplished by data's integration, compilation and manipulation form the Geographic Information System (GIS). The (GIS) has detailed physical and electric information data's of the distribution lines topology, of the duration curve tables average versus protection relay current and of the protection operation duration obtained by the System Control and Data Acquisition (SCADA).The LOCFALT performance was only evaluated in real situations, for this reason, more than 60 cases of accidental occurrences were evaluated in several Cemig system distribution lines, under many load loss conditions, usingthe obtained data from the registrars of Scada-Xomni and of Gemini (GIS). The results obtained in these analyses showed the advantages and the efficiency of this technique. The developed routine is now in test phase in CEMIG - Companhia Energética de Minas Gerais and it will be applied in the future in its practical system of localization of load loss in distribution lines, presenting appropriate results and contributing to the reduction of electrical reestablishment duration, therefore an increase of customers' satisfaction indexes.