Alocação ótima de indicadores de falta em redes de distribuição de energia elétrica utilizando modelagem matemática
Carregando...
Arquivos
Data
2018-08-24
Autores
Vilela, Leticia Biagi
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Neste trabalho propõe-se um modelo matemático de Programação Não Linear Binária (PNLB) para resolver o problema de alocação ótima de dispositivos indicadores de falta (OPFI – Optimal Placement of Fault Indicator Devices) em redes radiais de distribuição de energia elétrica. O problema é modelado por meio de uma função objetivo que considera dois custos conflitantes: o Custo da Energia Não Suprida (CENS) e o Custo de Investimento em Sensores (CINV). O modelo matemático proposto pode ser resolvido por métodos exatos e para todos os testes simulados foram utilizados os softwares de programação matemática AMPL (A Mathematical Programming Language) e KNITRO versão 9.0.0. Para validação do método proposto, são realizados testes em um sistema de distribuição de 19 barras, e em outros dois sistemas de 34 e 70 barras, sendo que nesses dois sistemas os indicadores de falta foram alocados apenas no tronco principal desses alimentadores. Os resultados obtidos mostram que o modelo matemático é adequado para resolver o problema formulado. Assim, o método proposto traz contribuição à literatura, visto que o problema da alocação ótima desses dispositivos tem sido resolvido principalmente por meio de técnicas meta-heurísticas.
This paper proposes a Binary Nonlinear Programming model to solve the problem of Optimal Placement of Fault Indicator Devices (OPFI) in radial distribution networks. The problem is modeled by an objective function that considers two conflicting costs: Cost of Energy Not Supplied and Cost of Investing in Sensors. The proposed mathematical model can be solved by exact methods and for all the simulated tests the mathematical programming software AMPL (A Mathematical Programming Language) and KNITRO version 9.0.0 were used. For validation of the proposed method, tests are performed in a 19-bar distribution system, and in two other systems of 34 and 70 bars, where in these two systems fault indicators were only allocated to the main trunk of these feeders. The results obtained show that the mathematical model is adequate to solve the problem formulated. Thus, the proposed method brings contribution to the literature, since the problem of optimal allocation of these devices has been solved mainly by means of metaheuristic techniques.
This paper proposes a Binary Nonlinear Programming model to solve the problem of Optimal Placement of Fault Indicator Devices (OPFI) in radial distribution networks. The problem is modeled by an objective function that considers two conflicting costs: Cost of Energy Not Supplied and Cost of Investing in Sensors. The proposed mathematical model can be solved by exact methods and for all the simulated tests the mathematical programming software AMPL (A Mathematical Programming Language) and KNITRO version 9.0.0 were used. For validation of the proposed method, tests are performed in a 19-bar distribution system, and in two other systems of 34 and 70 bars, where in these two systems fault indicators were only allocated to the main trunk of these feeders. The results obtained show that the mathematical model is adequate to solve the problem formulated. Thus, the proposed method brings contribution to the literature, since the problem of optimal allocation of these devices has been solved mainly by means of metaheuristic techniques.
Descrição
Palavras-chave
Programação matemática, Programação não linear binária, Alocação de sensores indicadores de falta, Allocation of fault indicator devices, Mathematical programming, Binary nonlinear programming