New hybrid RANS/LES methods are proposed where explicit algebraic Reynolds stress modelling is introduced both in the RANS and LES regions. Both in the form of an IDDES method allowing for wall-modelled LES simulations and in the form of an DDES methods where attached boundary layers are shielded within RANS. The methods are extensions of the SST-(I)DDES metods by Gritskevich, et al., Flow, turbulence and combustion 88 (3), 431?449, 2012. In channel flow the log region is well predicted for different Reynolds numbers similar as with the baseline SST-IDDES, but the EARSM-IDDES give a more rapid transition to turbulence closer to the wall. The model gives reasonable results on very coarse meshes, basically similar with the baseline SST-IDDES, for periodic hill flow at Re = 10.000 and 37.000 Also for the 3D Stanford diffuser, the EARSM-IDDES gives good results, here superior to the SST-IDDES, which has problems in fully resolving the turbulence in the inlet channel. The model in DDES mode is tested on the developing shear layer. The so-called grey area problem is substantially mitigated compared with the SST-DDES resulting in a more physically correct and fully developed resolved turbulence.

Concerning wall resolved large-eddy simulation (LES), a considerable reduction of computational resources is achievable by employing the Explicit Algebraic subgrid scale model (EAM) (\cite{marstorp2009explicit}).LES of periodic hill is carried out using OpenFOAM with the EAM and a low-diffusive implementation that has been previously tested on a turbulent channel flow. The aim of the present study is to evaluate in a broad sense the influence of  the Reynolds number on the flow quantities.

In the present work a new model for hybrid RANS-LES computations based on the improved-delayed-detached eddy simulation (IDDES) approach is proposed.The model combines the Explicit Algebraic Reynolds Stress model for RANS and the Explicit Algebraic sub-grid scale stress model.Several tests using the present model have been conducted for a turbulent plane channel flow and periodic hill flow adopting the general-purpose finite-volume code OpenFOAM. For each geometry two different Reynolds numbers are investigated. For both the Reynolds numbers considered, hybrid computations using the EARSM-IDDES model reasonably predict the mean flow quantities and the Reynolds stresses, compared to the established $k-\omega$ SST-IDDES.The ability of the EARSM to capture flow anisotropy is advantageous especially close to the lower wall of the periodic hill flow. The proposed model induces a steeper RANS to LES transition, with a resulting reduction of the extent of the transitional region, but with a less smooth solution through the transition.

There is a rapidly growing interest in using general-purpose CFD codes based on second-order finite volume methods for Large-Eddy Simulation (LES) in a wide range of applications, and in many cases involving wall-bounded flows. However, such codes are strongly affected by numerical dissipation and the accuracy obtained for typical LES resolutions is often poor. In the present study, we approach the problem of improving the LES capability of such codes by reduction of the numerical dissipation and use of an anisotropy-capturing subgrid-scale (SGS) stress model. The latter is of special importance for wall-resolved LES with resolutions where the SGS anisotropy can be substantial. Here we use the Explicit Algebraic (EA) SGS model [Marstorp L, Brethouwer G, Grundestam O, et al. Explicit algebraic subgrid stress models with application to rotating channel flow. J Fluid Mech. 2009;639:403-432], and comparisons are made for channel flow at friction Reynolds numbers up to 934 with the dynamic Smagorinsky model. The numerical dissipation is reduced by using an OpenFOAM based custom-built flow solver that modifies the Rhie and Chow interpolation and allows to control and minimise its effects without causing numerical instability (in viscous, fully turbulent flows). Different resolutions were used and large improvements of the LES accuracy were demonstrated for skin friction, mean velocity and other flow statistics by use of the new solver in combination with the EA SGS model. By reducing the numerical dissipation and using the EA SGS model the resolution requirements for wall-resolved LES can be significantly reduced.

Properly resolved large-eddy simulations of wall-bounded high Reynolds number flows using standard subgrid-scale (SGS) models requires high spatial and temporal resolution. We have shown that a more elaborate SGS model taking into account the SGS Reynolds stress anisotropies can relax the requirement for the number of grid points by at least an order of magnitude for the same accuracy. This was shown by applying the recently developed explicit algebraic subgrid-scale model (EAM) to fully developed high Reynolds number channel flows with friction Reynolds numbers of 550, 2000, and 5200. The near-wall region is fully resolved, i.e., no explicit wall modeling or wall functions are applied. A dynamic procedure adjusts the model at the wall for both low and high Reynolds numbers. The resolution is reduced, from the typically recommended 50 and 15 wall units in the stream-and spanwise directions respectively, by up to a factor of 5 in each direction. It was shown by comparison with direct numerical simulations that the EAM is much less sensitive to reduced resolution than the dynamic Smagorinsky model. Skin friction coefficients, mean flow profiles, and Reynolds stresses are better predicted by the EAM for a given resolution. Even the notorious overprediction of the streamwise fluctuation intensity typically seen in poorly resolved LES is significantly reduced whenEAMis used on coarse grids. The improved prediction is due to the capability of the EAM to capture the SGS anisotropy, which becomes significant close to the wall.

Large-eddy simulation (LES) of turbulent channel flow are performed with a new subgrid-scale (SGS) stress model. The simulations show that with this model we can well predict turbulent wall flows at coarse resolutions and moderately high Reynolds numbers. The commonly used dynamic Smagorinsky model fails at coarser resolutions.