Simulações de grandes escalas e simulação numérica direta sub-resolvida de escoamento turbulento de fluidos Herschel-Bulkley

• Ano de defesa: 2022
• Autor(a) principal: Basso, Felipe Oliveira
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Engenharia Mecânica e de Materiais; UTFPR
• Palavras-chave em Português: Fluidos não-newtonianos; Métodos de simulação; Viscosidade; Fluxo viscoso; Turbulência; Dinâmica dos fluidos; Fluidodinâmica computacional; Non-Newtonian fluids; Simulation methods; Viscosity; Viscous flow; Turbulence; Fluid dynamics; Computational fluid dynamics; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA::ENGENHARIA TERMICA; Engenharia Mecânica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/30322

The Herschel-Bulkley model’s turbulent flows of non-Newtonian fluids are encountered in several engineering applications, e.g. drilling fluids, mining slurries, and polymer solutions. However, studies regarding Large Eddy Simulations (LES) involving this type of fluid are scarce. The present work aims to contribute to this issue. The first part of the thesis focuses on evaluating the LES approach in predicting the flow features of Herschel-Bulkley fluids. The numerical methodology is compared against literature reference data. Two subgrid-scale (SGS) models are considered: Dynamic Smagorinsky and Wall Adapting Local Eddy-viscosity (WALE) models. Additionally, under-resolved direct numerical simulations (UDNS) are also performed. Following that, the solution method is used to analyze the turbulent flow of Herschel-Bulkley fluids through a concentric annular section with inner cylinder rotation at ReG ≈ 9, 000. The influence of the flow behavior index (n = 0.65, 0.70 and 0.75), the Bingham number (Bn = 0.10, 0.25 and 0.40), and the rotation rate (N = 0.15 and 0.30) on the flow characteristics are explored. The instantaneous flow quantities, including contours of the axial velocity and viscosity, vortex structures, the first- and second-order turbulence statistics, and mean flow features, such as mean viscosity profiles, mean turbulence kinetic energy, pressure gradient, and friction factor are investigated. The results show that weaker turbulent structures are generated as the n value is reduced and the Bingham number increases. Raising the rotation rate enhances the magnitudes of turbulent statistics and makes the levels of velocity fluctuations more asymmetrical. It is also demonstrated that predictions obtained with the SGS models and the UDNS do not present significant differences.