| Resumo: | Computational Fluid Dynamics has been implemented for indoor environments ventila‐ tion studies. Despite of its maturity and usefulness, its use for engineering applications such as heating, ventilation, and air conditioning (HVAC) projects is still scarce in Por‐ tugal. With this aim, this dissertation presents some introductory studies to create a computational fluid dynamics (CFD) workflow for the use by HVAC engineers with min‐ imal knowledge of CFD practices. Thus, some aspects were investigated, such as mesh design, discretization schemes, solvers, boundary conditions and turbulence models. For the latter, from a large set of turbulence models, six Reynolds‐averaged Navier‐Stokes (RaNS) turbulence models were tested and validated for indoor ventilation, the standard k − ε, the renormalization group (RNG) k − ε, the realizable k − ε, the v ′2 − f, the k − ω and the k−ω shear stress transport (SST). The standard k−ε model continues to present the most satisfactory results of all the studied turbulence models. The RNG k − ε model also showed good agreement with the experimental data, however it required more time to achieve convergence. It was not possible to achieve convergence with the realizable k −ε model. The k −ω and k −ω SST models yield results significantly different from the experimental measurements, especially in the far wall region. These studies were com‐ pared with the benchmarks from the International Energy Agency Annex 20. This work also presents 2D and 3D cases, mesh convergence studies, effects of flow buoyancy (incompressible, isothermal and non‐isothermal cases), and steady‐state and tran‐sient flow simulations.
|
|---|