Estudo dos métodos de tomada de decisão aplicados a reconfiguraçãosazonal de redes de distribuição de energia elétrica

• Ano de defesa: 2012
• Autor(a) principal: Jésus Jonatan Souza Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Engenharia Elétrica; Engenharia elétrica
• Link de acesso: http://hdl.handle.net/1843/BUOS-APFRW6

The Distribution Networks for Electricity (PDN) are responsible for providing the majority of commercial customers, residential and industrial. In southeastern Brazil, for example, have around 90 % of establishments connected to the power grid. However, these networks are exposed to inclement weather, lightning, climate change andvariations in load and ambient temperature which often exhibit seasonal behavior. In this context, given the importance of providing PDN, realized the need of a seasonal approach to network reconfiguration. Importantly, the increases in temperature in a given region are associated with greater use of cooling loads, reflecting the increased loads. In contrast milder temperatures favor a reduction in power consumption compared to the warmer periods. In this study we used information from the climateof southeastern Brazil for the modeling of scenarios, from observing the behavior of the climate region, which generated three instances of a network of 100 nodes, each corresponding to a scenario of seasonality. Thus, it is necessary to use methods of decision making for the choice of solutions to be implemented according to each proposed scenario, since the preference of the decision maker is altered in relation to established criteria. In this study we used the method ELECTRE III of the French School, which is based on the concept of over-the method of Analytic Hierarchy Process developed by Thomas Saaty and application of fuzzy logic for decision making, in order to represent the vague judgments inherent in human decision maker. Since this work will not apply the classical methods based on utility theory due to limitations relatedto the modeling of preference and the treatment of non-convex portions of the Pareto-optimal as previously explored in the literature. The purpose of this study is to evaluate the methods of decision making (DM), applied to optimal reconfiguration of Distribution Networks for Electricity (PDN) within the seasonal scenarios. Where we have figures of merit such as the following objectives: monetary cost of the network, cost of failure (associated with network reliability) and number of maneuversto reset, and used a Discrete Differential Evolution Algorithm for multiobjective optimization. Will also be evaluated by decision makers criteria retrospectively obtained through Monte Carlo simulation, to evaluate the sensitivity to uncertainties in load and energy prices, within the horizon of seasonality. Since each picture is used for instance a network benchmark modified according to the specific case, are alsomodeled and the decision-making methods with preferences in accordance with each scenario. Finally, we evaluated the DM second aspect: the effort to interpret the parameters, the modeling effort to the preference and robustness of the decision method evaluated in relation to sensitivity of the methods the variation of parameters.