Modelagem de condições de contorno para escoamentos turbulentos utilizando simulações das grandes escalas

• Ano de defesa: 2012
• Autor(a) principal: Damasceno, Marcelo Maia Ribeiro
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Engenharia Química; Engenharias; UFU
• Palavras-chave em Português: Escoamento turbulento; Métodos de simulação; Condições de entrada; Modelos de turbulência; Simulação das grandes escalas; Turbulent inlet conditions; Turbulence models; Large-eddy simulation; CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/15221

Turbulence is a complex and chaotic phenomenon which is extremely sensitive to initial and boundary conditions. Due to the wide range of time and length scales generally present, it requires a large amount of computational resources to be completely captured. Approaches to turbulence models are used to simplify this situation. However, they may reduce the ow description quality. In this context, 2D and 3D numerical simulations of backward-facing step ows, in laminar and turbulent regimes, were experimented in the present work. These simulations were done with a computer code developed in FORTRAN, based on nite volumes with Centered Dierences Scheme and Backward Dierence Formula as spatial and temporal discretizations, respectively, using the Large- Eddy Simulation methodology. This method is based on the appliance of a spatial lter to the equations of motion. Scales larger than this lter are resolved, while the smallest ones are modeled. Three alternatives for subgrid modeling were tested (models proposed by Smagorinsky 1963, Germano et al. 1991 and no turbulence modeling) and the application of three dierent turbulent inlet conditions generators were analyzed. A better turbulent ow characterization was achieved when the turbulence dynamic model was used with turbulent inlet conditions generators, which presented time and space correlations.