Ferramenta de suporte para autorrecuperação de rede de distribuição de energia elétrica utilizando redes neurais artificiais

• Ano de defesa: 2017
• Autor(a) principal: Avelar, Fabio da Silva
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Sistemas de Energia; UTFPR
• Palavras-chave em Português: Energia elétrica - Distribuição; Redes neurais (Computação); Redes elétricas inteligentes; Algorítmos genéticos; Software - Desenvolvimento; Simulação (Computadores); Sistemas de energia elétrica; Electric power distribution; Neural networks (Computer science); Smart power grids; Genetic algorithms; Computer software - Development; Computer simulation; Electric power systems; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::TELECOMUNICACOES::SISTEMAS DE TELECOMUNICACOES; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/2974

A tool was developed to assist in the self-recovery of the electricity distribution network, with the help of software, to simulate a real system. The electrical system considered has intelligent keys capable of identifying a momentary fault in the line and finding the best reconfiguration for its reclosing, characterizing a Smart Grid. Using artificial intelligence, Artificial Neural Network (ANN), simulated fault situations in certain stretches of the electrical network and analyzed power flow through OpenDSS, observing the most appropriate switching within the shortest time interval, an implementation was also performed via ELIPSE in the IEEE electrical system in question for better visualization identifying the reclosing of this system. The algorithm developed through a fault chooses the best configuration to restore the energy to the largest number of consumers during it. With the results of the simulations, tests and analyzes were performed to verify their robustness and velocity when compared to the actions of the operators, in the hope that the developed model will be faster than an experienced Operator of a Distribution Operation Center in its task of analysis. This work presents an algorithm application for different distribution network configurations, reducing the time and quantity of affected consumers, allowing a better targeting of the electrician teams for the restoration, thus gaining time, minimizing the wear of professionals, components electricity distribution and operators.