Development of a new LES/RANS formulation

Abstract

A new large-eddy simulation / Reynolds-averaged Navier-Stokes (LES/RANS) turbulence model is described in this work. In common with other such models developed at NCSU, the RANS-to-LES transition is facilitated by a flow-dependent blending function based on estimates of outer-and inner turbulence length scales. In contrast to prior work, the outer scale is a function of an eddy viscosity specifically used for this purpose. A transport equation for this eddy viscosity is solved, with excessive growth of eddy viscosity due to fluctuating strain rates controlled by a destruction term based on the von Kármán length scale. The new model is completely local and is capable of adjusting to the changing state of a boundary layer. The model is tested for incompressible and compressible flat-plate boundary layers at low and moderate Reynolds numbers and for flow over an airfoil with trailing-edge separation. The model is shown to be capable of predicting the composite structure of a boundary layer with minimal log-law mismatch.The model also possesses additional degrees of freedom that may enable improved predictions for strongly nonequilibrium turbulent flows.

A rapid and low noise switch from RANS to WMLES on curvilinear grids with compressible flow solvers

“…As classically observed in the framework of WMLES simulations on coarse grids ( [96,20]), a slight Log-Layer Mismatch (LLM) located near the RANS/LES interface can be depicted. The detailed discussion of the LLM issue is beyond the scope of this article but let us mention that no general, definitive and theory-motivated hybrid RANS/LES solution for curvilinear geometries has been published yet (see Shur et al [96] and Shen and Edwards [94]). It is observed in Ref.…”

On the Convection Velocity of Wall-Bounded Turbulence Resolved by ZDES Mode III at $$Re_\theta = 13 000$$ R e θ = 13000

“…Alternatively, the passive interface treatment may be improved by ad hoc functions and length scale definitions, such as in IDDES ( [30]). However, this mostly empirical formulation still deviates from the logarithmic law and highly depends on the mesh resolution [29]. Instead, the simple initial DES formulation may be retained for WMLES if the RANS/LES interface height is properly defined by the user to minimise the reported difficulties.…”