|
KTH /
Engineering Sciences /
Mechanics /
Research /
Project list /
Project information
Project description
Explicit algebraic sub-grid scale modelling for large-eddy simulations
| Research area: |
Fluid mechanics |
| Project members: |
|
| Sponsor: |
VR
|
Project description
Large eddy simulation (LES) is becoming a tractable tool for simulation of industrial flows. However, accuracy of LES predictions is still limited by the subgrid-scale model (SGS). If standard SGS models like the Smagorinsky model are to be used for complex flow cases, a large number of grid points have to be used to get satisfactory predictions. An SGS model that is capable of modeling the anisotropy of the unresolved scales in a realistic way has recently been developed in our group. It uses an explicit algebraic formulation of the transport equation of the anisotropy of the SGS stresses, (see Marstorp et al. JFM 2009, 639, pp 403-432). In the present project we have studied extensively the dependence of isotropic SGS models on resolution using a wide range of resolutions (Rasam et al. 2010, paper accepted for J. of Turb.) and have shown the considerable improvements that one can obtain using the explicit algebraic SGS model, which makes it a good candidate for industrial LES. Figure 1 shows that a great improvement in prediction of the wall friction can be obtained with the explicit algebraic model. Another topic in this project is LES of passive scalars (such as temperature, pollutants in air etc.), and modeling of the SGS scalar fluxes. In correspondence with the explicit algebraic model for the SGS stresses we have developed an explicit algebraic scalar flux model, which can realistically predict SGS scalar fluxes and its predictions are not generally aligned with the mean scalar gradient.Related articles published in journals or conference proceedings
Internal reports
|
|
|
