A hybrid RANS-LES approach with delayed-DES and wall-modelled LES capabilities

Shur, M. L.; Spalart, Philippe R.; Strelets, M. Kh.; Travin, Andrey

doi:10.1016/j.ijheatfluidflow.2008.07.001

Cited by 1,682 publications

(760 citation statements)

References 24 publications

Supporting

Mentioning

752

Contrasting

Unclassified

Order By: Relevance

“…Hybrid RANS-LES models Spalart Allmaras (SA) [5] SST-DDES [6] Realizable k − ε (RKE) [7] SST-IDDES [8] k − ω SST (SST) [9] SA-DDES [6] Elliptic Blending k − ε v 2 − f (B-EVM) [10] SA-IDDES [11] Elliptic Blending Reynolds Stress Model (EB-RSM) [12] The principle behind hybrid RANS-LES methods is that a traditional wall-resolved LES is too expensive for complex high-Reynolds number flows that are influenced by near-wall effects. By applying a RANS model in the boundary layer and using LES outside of this region, the total cell count can by greatly reduced as RANS models do not require the same level of grid resolution in the boundary layer.…”

Section: Rans Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Assessment of RANS and DES methods for realistic automotive models

et al. 2016

View full text Add to dashboard Cite

show abstract

Section: Rans Modelsmentioning

confidence: 99%

“…In this study we investigate both standard DDES [6] as well as the variant, IDDES [11], which aims to improve some of the shortcomings of DDES and add a wall-modelled LES capability. Readers are advised to read the doctoral thesis of Mockett [14] for a comprehensive explanation of DES models and its variants.…”

Section: Rans Modelsmentioning

confidence: 99%

Assessment of RANS and DES methods for realistic automotive models

et al. 2016

View full text Add to dashboard Cite

show abstract

“…where h max is the maximum edge length of the cell, d w is the distance to the nearest wall and h wn is the grid step in the normal direction to the surface [11]. Additionally,f d is an empirical delay function that depends on the grid (i.e., cell aspect ratio and distance to the wall) and on the solution (i.e., local and instantaneous values of ν t , strain rate and vorticity tensor).…”

Section: Turbulence Model Descriptionmentioning

confidence: 99%

“…This test case was carried with the same flow and mesh parameters as Shur et al [11], using an almost identical hybrid model, but different discretization schemes. Since OpenFOAM R has intrinsically a second order spatial discretization, the schemes used in our model implementation are all second order; while the original reference case used fourth order central schemes.…”

Section: Numerical Frameworkmentioning

confidence: 99%

Hybrid turbulence models for atmospheric flow: A proper comparison with RANS models

Bautista

Dufresne

Masson

2015

E3S Web of Conferences

View full text Add to dashboard Cite

Abstract.A compromise between the required accuracy and the need for affordable simulations for the wind industry might be achieved with the use of hybrid turbulence models. Detached-Eddy Simulation (DES) [1] is a hybrid technique that yields accurate results only if it is used according to its original formulation [2]. Due to its particular characteristics (i.e., the type of mesh required), the modeling of the atmospheric flow might always fall outside the original scope of DES. An enhanced version of DES called Simplify Improved Delayed Detached-Eddy Simulation (SIDDES) [3] can overcome this and other disadvantages of DES. In this work the neutrally stratified atmospheric flow over a flat terrain with homogeneous roughness will be analyzed using a Reynolds-Averaged Navier-Stokes (RANS) model called k − ω SST (shear stress transport) [4], and the hybrids k − ω SST-DES and k − ω SST-SIDDES models. An obvious test is to validate these hybrid approaches and asses their advantages and disadvantages over the pure RANS model. However, for several reasons the technique to drive the atmospheric flow is generally different for RANS and LES or hybrid models. The flow in a RANS simulation is usually driven by a constant shear stress imposed at the top boundary [5], therefore modeling only the atmospheric surface layer. On the contrary the LES and hybrid simulations are usually driven by a constant pressure gradient, thus a whole atmospheric boundary layer is simulated. Rigorously, this represents two different simulated cases making the model comparison not trivial. Nevertheless, both atmospheric flow cases are studied with the mentioned models. The results prove that a simple comparison of the time average turbulent quantities obtained by RANS and hybrid simulations is not easily achieved. The RANS simulations yield consistent results for the atmospheric surface layer case, while the hybrid model results are not correct. As for the atmospheric boundary layer case, no meaningful conclusion could be established for RANS, and the DES results are not satisfactory. However the SIDDES model is capable of reproducing accurately the atmospheric boundary layer over flat terrain.

show abstract

“…Improved Delayed Detached Eddy Simulation (IDDES) [Shur et al, 2008] is a hybrid RANS-LES (Large Eddy Simulation) model based on the original Detached Eddy Simulation (DES) formulation. A particularly appealing aspect of this model in the present application is the fact that, when stimulated by fluctuating turbulent flow in the flow field external to a boundary layer, within the boundary layer IDDES behaves as a Wall-Modeled LES (WMLES) model, as opposed to previous versions of DES, which show RANS behavior under similar circumstances.…”

Section: Introductionmentioning

confidence: 99%