Índices de confiabilidade devido a vegetação e planejamento de podas de árvores em redes de distribuição

• Ano de defesa: 2013
• Autor(a) principal: Souza, Eduardo de [UNESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Poda; Arvores; Manutenção; Redes eletricas; Pruning
• Link de acesso: http://hdl.handle.net/11449/110525

Parts of interruptions in electricity supply are caused by contact with the edges of the vegetation system elements, impairing the reliability and quality of distribution systems. Increased costs with no electricity supply and corrective maintenance are the consequences of the absence of plans for carrying out the maintenance of vegetation. The need to maintain the supply of electricity and consequently the increased confidence of the distribution system is the aim of programming models focused on increasing the system reliability at the low cost. This work propose a mathematical model of dynamic multi-objective binary programming problem for preventive maintenance of vegetation under a system of power distribution. The optimization model contemplates to minimize three objectives such as cost of unserved energy due to failures related to the vegetation, maintenance costs such as pruning or removal of trees, and an indication of system unreliability due to continuity of service based on FEC (Equivalent Frequency for Consumer). The reliability of the feeders is determined using probabilistic models as a renewable process, considered in the mathematical model as an unreliability index. The solution methodology is proposed a multi-objective algorithm NSGA-II implemented in the programming language C++, obtaining a set of non-dominated solutions among themselves, which tends to converge to Pareto optimal front. This methodology is tested on a real distribution system of a medium sized city.