Programação linear inteira mista aplicada na proteção adaptativa de sobrecorrente direcional considerando contingência n-1

• Ano de defesa: 2022
• Autor(a) principal: Martin, Klaus Tesser
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Engenharia Elétrica; UFSM; Programa de Pós-Graduação em Engenharia Elétrica; Centro de Tecnologia
• Palavras-chave em Português: Proteção de sobrecorrente; Programação linear inteira mista; Proteção adaptativa; Coordenação de relés; Overcurrent protection; Mixed integer linear programming; Adaptive protection; Relay coordination; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: http://repositorio.ufsm.br/handle/1/29212

This work presents a methodology for adjusting and coordinating overcurrent protection in interconnected transmission systems considering a possible network contingency state. For the protection parameterization, the methodology uses the short-circuit data of the current network topology, however, the maximum load current data in the protection section of each relay is obtained considering the n-1 criterion, already foreseeing the disconnection of some network line. Thus, if a line gets disconnected, other protective devices will not improperly actuate by the redistribution of load currents in the network. The objective is to propose an adaptive protection scheme, so that the adjustment and coordination are redone for each topological change in the network. To this end, the research considers a smart grid environment, where there is a supervisory system and remote communication between a central and digital relays. As the focus of this research concentrates on interconnected transmission systems, the protection consists of directional overcurrent relays, ANSI-51 and ANSI-67 functions. The coordination problem, originally non-linear and non-convex, is linearized, allowing its formulation as a problem of Mixed Integer Linear Programming. For processing the model, the commercial solver CPLEX® is used. Results show that the coordination is obtained in a reduced computational processing time, showing the potential of the methodology for real time application.