Um modelo eficiente para otimização de torres eólicas de concreto armado via algoritmos genéticos

• Ano de defesa: 2019
• Autor(a) principal: Duarte, Thays Guerra Araújo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Otimização estrutural; Algoritmos genéticos; Ação do vento; Concreto armado; Torre eólica
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/59273

In recent years, there has been growing interest in wind energy and investments in new technologies to enhance turbine power and minimize energy costs. Turbines are commonly supported by steel towers. However, to capture stronger and more consistent winds, towers need to reach higher altitudes. Concrete towers have become more advantageous as it allows higher heights and presents itself as a more economical solution. Besides, concrete towers have other advantages such as improved structural performance, durability, safety, and fewer transportation problems. The tower is the most material consuming part in the entire structure and by reducing the costs of concrete and steel, wind energy can become more competitive than fossil fuels. The efficient design of the tower is important to ensure the economic feasibility of a wind farm. Better projects can be achieved by optimizing the tower design, for example reducing the cost of materials. The study of numerical optimization techniques applied onto concrete towers becomes very interesting and little has been done, especially when considering the dynamic nature of wind. The main objective of this paper is the formulation of a cost optimization model for reinforced concrete wind towers discussing the dynamic modeling of wind action. Finite Element Analysis is applied, and 1D elements are investigated. Simplified models for the treatment of physical nonlinearity and dynamic wind modeling are investigated for efficient optimization. In tower analysis, NBR 6118 and IEC 61400-1 loading conditions and recommendations are followed. Once the design variables, represented by the characteristic dimensions of the tubular cone tower segments, reinforcement and the concrete resistance, are treated as discrete, associated with the nonlinearities of the problem, the model is solved with genetic algorithms. Towers 100 m or more in height are studied. Preliminary optimized solutions were efficiently obtained for the wind towers during the dimensioning phase of these structures that have complex behavior.