A priori analysis for high-fidelity large-eddy simulation of wall-bounded transcritical turbulent flows
Transcritical turbulent flows are governed by the compressible Navier–Stokes equations along with a real-gas equation of state. Their computation is strongly susceptible to numerical instabilities and requires kinetic-energy- and pressure-equilibrium-preserving schemes to yield stable and non-dissip...
| Autores: | , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/407117 |
| Acceso en línea: | https://hdl.handle.net/2117/407117 https://dx.doi.org/10.1016/j.supflu.2024.106191 |
| Access Level: | acceso abierto |
| Palabra clave: | Turbulence Large-eddy simulation High-pressure Supercritical fluids Transcritical wall-bounded flows Turbulència Àrees temàtiques de la UPC::Enginyeria mecànica::Mecànica de fluids |
| Sumario: | Transcritical turbulent flows are governed by the compressible Navier–Stokes equations along with a real-gas equation of state. Their computation is strongly susceptible to numerical instabilities and requires kinetic-energy- and pressure-equilibrium-preserving schemes to yield stable and non-dissipative scale-resolving simulations. Building upon a recently developed kinetic-energy- and pressure-equilibrium-preserving discretization framework based on transporting a pressure equation, the objectives of this paper are to (i) derive a filtered set of equations suitable for large-eddy simulation, and (ii) characterize the properties of the resulting subfilter-scale terms by performing a priori analyses of transcritical wall-bounded turbulence direct numerical simulation data. The filtering operation leads to three unconventional subfilter-scale terms that emerge from the pressure equation and require dedicated modeling. The subfilter-scale stress tensor is dissected in terms of magnitude, shape and orientation based on an eigendecomposition analysis, and compared with existing subfilter-scale models. A priori analyses confirm that models of eddy-viscosity type are favorable for this framework, although the tensor shape is not fully captured. Closure expressions are finally proposed and tested for the novel subfilter terms, showing acceptable performances. |
|---|