Experimentação numérica de bolhas em ascensão

• Ano de defesa: 2016
• Autor(a) principal: Barbi, Franco
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Mecânica
• Palavras-chave em Português: Engenharia mecânica; Escoamentos - Simulação por computador; Bolhas (Física); VOF; SAMR; Dinâmica de bolhas; Coluna de Bolhas; Bubble dynamics; Bubble collumn; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/18155; https://doi.org/10.14393/ufu.te.2016.107

In bubble dynamics studies computational tools plays an important role as it allows, through virtual experimentations, the increase of knowledge about forces that act in this kind of flow. Considering results reliability, it is common to observe constitutive force models improvement, as in drag, lift and virtual mass correlations. The AMR3D code, developed in the Fluid Mechanics Laboratory - MFlab, uses locally block-structured meshes that dynamically adapts itself to characteristics of a given flow. A projection method is used for the pressure-velocity coupling, and the Volume of Fluid - VOF method for the phases representation, including interface dynamics. The equations of linear momentum balance are discretized with a second order semi-implicit scheme (IMEX). The present thesis contributes to the code with the implementation of a new refinement criteria that captures the main structures of the flow, and an identification algorithm that allows users to perform operations on a single bubble between a swarm of bubbles. The present code is verified and validated against well known literature problems. Results of isolated ascending bubbles in a quiescent media are presented and compared with Grace’s diagram, showing good agreement. Fluid-dynamic forces are computed and analyzed, as well as fluid-dynamics forces coefficients for drag, lift and virtual mass, showing good agreement with literature. Simulations of a pair of bubbles interacting were presented and comparisons were realized against results of literature, showing good proximity. In one of the cases, the bubble identification algorithm was tested as it calculates bubbles known volumes and centroid positions consistently. Bubble swarm simulations were presented, and the identification algorithm were also applied to track a bubble as it ascend though the domain to expose the fluid-dynamics forces acting on it. With the presented results, one can conclude that the AMR3D code has a potential for investigative approaches in bubbly flows.