Leão, Fábio Bertequini [UNESP]Pereira, Rodrigo A. F. [UNESP]Mantovani, José R. S. [UNESP]2014-05-272014-05-272009-12-172009 IEEE Power and Energy Society General Meeting, PES '09.http://hdl.handle.net/11449/71484In this paper, a methodology based on Unconstrained Binary Programming (UBP) model and Genetic Algorithms (GAs) is proposed for estimating fault sections in automated distribution substations. The UBP model, established by using the parsimonious set covering theory, looks for the match between the relays' protective alarms informed by the SCADA system and their expected states. The GA is developed to minimize the UBP model and estimate the fault sections in a swift and reliable manner. The proposed methodology is tested by utilizing a real-life automated distribution substation. Control parameters of the GA are tuned to achieve maximum computational efficiency and reduction of processing time. Results show the potential and efficiency of the methodology for estimating fault section in real-time at Distribution Control Centers. ©2009 IEEE.engDistribution substationsFault section estimationGenetic algorithmsProtective relayingAutomated distributionBinary programmingControl parametersDistribution control centerFault sectionsProcessing TimeSet covering theoryAutomationControl system analysisEstimationMobile telecommunication systemsPotential energyPotential energy surfacesRelay protectionSCADA systemsComputational efficiencyTrabalho apresentado em evento10.1109/PES.2009.5275338WOS:000275524902105Acesso aberto2-s2.0-7184911110512489565932365150000-0002-8846-24230000-0002-8846-2423