Gerenciamento da oferta e da demanda em micro-sistemas de energia via otimização por enxames de partículas

• Ano de defesa: 2011
• Autor(a) principal: Thaís de Fátima Araújo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Despacho Econômico; Participação da Demanda; Otimização por Enxames de Partículas; Redes Inteligentes de Distribuição; Engenharia elétrica
• Link de acesso: http://hdl.handle.net/1843/BUOS-8WHJDA

Intelligent micro-grids, which integrate distributed energy resources with loads, are among the current major tendencies for electric power distribution systems. Distributed resources promote important changes in the operation and management of the distribution system, since they require that the control and the intelligence of the system must bedistributed on the network. In order to address the management of small electric power systems, modeled in the context of the distribution networks of the future, this work deals with the formulation andimplementation of grid management functions which, in addition to promoting the economic dispatch of generation, also manage consumer participation. The demand participation is modeled assuming that some consumers have a partially flexible consumption: they can shift part of their demand along a certain time horizon in order to reduce generation costs. The management model is implemented via an economic dispatch that considers not only the constraints of the linearized power flow, but also an energy constraint that imposes an amount of energy to be consumed during the time horizon. Therefore, due to this energyconstraint, the dispatch problem must be solved considering the whole time horizon. The economic dispatch with demand participation models proposed in this work are implemented using a particle swarm optimization algorithm and can be used in operations planning of micro-grids. The optimization algorithm is modified by including techniques to handle equality constraints. Although this is an important aspect of the problem, since restrictions represent the system operation, there are few studies in the literature on the subject. Two strategies are implemented: the first is an analytical methodology, while thesecond is based on the use of penalty functions to handle equality constraints. The results obtained suggest that the particle swarm optimization algorithm, modified by techniques to handle the constraints, is a suitable tool for solving the economic dispatch with consumer participation.