Loading...
Publication
Planeamento de uma rede de distribuiĆ§Ć£o num contexto de Smart Grid
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 6.86 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
O setor elĆ©trico atual encontra-se numa transiĆ§Ć£o para um novo paradigma. O futuro que se
avizinha incita a novos desafios, cada vez mais exigentes, para todos os intervenientes desta
cadeia de valor. A constante evoluĆ§Ć£o tecnolĆ³gica, a introduĆ§Ć£o do conceito de Smart Grid,
a elevada penetraĆ§Ć£o da produĆ§Ć£o distribuĆda, a massificaĆ§Ć£o dos veĆculos elĆ©tricos, sĆ£o
alguns dos fatores que tornam o planeamento das redes elƩtricas mais complexo, mais
criterioso, mas fundamental. Apesar da complexidade introduzida, Ʃ necessƔrio continuar a
garantir a satisfaĆ§Ć£o dos requisitos tĆ©cnicos impostos para funcionamento contĆnuo do
sistema, obtendo sempre um retorno financeiro dos investimentos efetuados.
A exigĆŖncia, as incertezas associadas aos dados de entrada, a satisfaĆ§Ć£o de diversos objetivos
que, em muitos dos casos torna-se complexo encontrar um compromisso entre si, promove
a utilizaĆ§Ć£o de ferramentas tĆ©cnicas avanƧadas de otimizaĆ§Ć£o, atravĆ©s de complexos
algoritmos matemƔticos auxiliados por potentes meios informƔticos, com o intuito de
alcanƧar a melhor soluĆ§Ć£o para o problema. InĆŗmeros autores sugerem diferentes
metodologias e abordagens com vista Ć melhoria das soluƧƵes do problema de planeamento,
contudo nĆ£o integram o contexto de Smart Grid e os recursos associados. Assim sendo, os
principais contributos desta dissertaĆ§Ć£o assentam no desenvolvimento de uma metodologia
que para alĆ©m de incorporar uma elevada penetraĆ§Ć£o dos produtores distribuĆdos, Ć© feito
atravĆ©s de um mĆ©todo determinĆstico e em tempos de execuĆ§Ć£o aceitĆ”veis.
O problema Ć© formulado atravĆ©s de simplificaƧƵes das nĆ£o linearidades inerentes ao
tratamento matemĆ”tico e Ć© resolvido por recurso a tĆ©cnicas determinĆsticas. A ferramenta
informĆ”tica concebida ā SG Plan (Smart Grid Planning) ā considera uma minimizaĆ§Ć£o de
uma funĆ§Ć£o multiobjetivo com trĆŖs termos: custo da energia de perdas, custo da energia nĆ£o
fornecida esperada e custo de investimento em novas linhas. Assim sendo, a metodologia
apresentada foi aplicada em duas redes distintas de forma a testar a sua eficĆ”cia e eficiĆŖncia.
A anĆ”lise dos resultados, comprovou a adequaĆ§Ć£o, robustez e versatilidade da metodologia
mostrando ser Ćŗtil na tomada de decisĆ£o ao nĆvel do planeamento de redes de distribuiĆ§Ć£o de
energia elĆ©trica em MĆ©dia TensĆ£o (MT), no contexto de Smart Grid.
The current electricity sector is in a transition to a new paradigm. The future brings new and more complex challenges for all stakeholders in the value chain. The constant technology evolution, the introduction of the Smart Grid concept, the high penetration of distributed generators, the increasingly use of electric vehicles, are some of the factors that make electrical networks planning crucial. Despite this complexity, it is necessary to continue to guarantee the technical requirements imposed in order to maintain the continuous operation of the system, obtaining always a financial return of the investments. The complexity, the uncertainties associated to the input data, the fulfillment of several objectives which, in many cases becomes hard to find a compromise between them. Thus, it can be promoted, the use of advanced technical optimization tools, in order to achieve the best solution for the problem. Several authors suggest different methodologies and approaches to improve the solutions of the planning problem, however they do not integrate the Smart Grid context. In this way, the main contributions of this thesis are considering high penetration distributed energy resources (namely distributed generators, energy storage systems, EVs parking lots) in a medium voltage distribution network planning problem; development of a deterministic methodology with an acceptable computational execution time for medium voltage distribution network planning problem considering all available resources. The problem is formulated through simplifications of the nonlinearities inherent to the mathematical and its solved using deterministic techniques. The decision-support tool developed ā SG Plan (Smart Grid Planning) ā consider the minimization of a multi-objective function with three terms ā losses cost, expected energy not supplied cost and investment in new lines. The proposed methodology was applied in two different networks in order to test its efficiency. The analysis of the results, proved the adequacy, robustness and versatility of the proposed methodology, showing to be useful in the decision making at the planning level of distribution networks, in a Smart Grid.
The current electricity sector is in a transition to a new paradigm. The future brings new and more complex challenges for all stakeholders in the value chain. The constant technology evolution, the introduction of the Smart Grid concept, the high penetration of distributed generators, the increasingly use of electric vehicles, are some of the factors that make electrical networks planning crucial. Despite this complexity, it is necessary to continue to guarantee the technical requirements imposed in order to maintain the continuous operation of the system, obtaining always a financial return of the investments. The complexity, the uncertainties associated to the input data, the fulfillment of several objectives which, in many cases becomes hard to find a compromise between them. Thus, it can be promoted, the use of advanced technical optimization tools, in order to achieve the best solution for the problem. Several authors suggest different methodologies and approaches to improve the solutions of the planning problem, however they do not integrate the Smart Grid context. In this way, the main contributions of this thesis are considering high penetration distributed energy resources (namely distributed generators, energy storage systems, EVs parking lots) in a medium voltage distribution network planning problem; development of a deterministic methodology with an acceptable computational execution time for medium voltage distribution network planning problem considering all available resources. The problem is formulated through simplifications of the nonlinearities inherent to the mathematical and its solved using deterministic techniques. The decision-support tool developed ā SG Plan (Smart Grid Planning) ā consider the minimization of a multi-objective function with three terms ā losses cost, expected energy not supplied cost and investment in new lines. The proposed methodology was applied in two different networks in order to test its efficiency. The analysis of the results, proved the adequacy, robustness and versatility of the proposed methodology, showing to be useful in the decision making at the planning level of distribution networks, in a Smart Grid.
Description
Keywords
Fiabilidade OtimizaĆ§Ć£o multi-objetivo Planeamento ProduĆ§Ć£o DistribuĆda Rede de DistribuiĆ§Ć£o Smart Grid TrĆ¢nsito de PotĆŖncias Ćtimo Distribution Network Distributed Generation Multi-objective Optimization Optimal Power Flow Planning Reliability Smart Grid