Synthetic Turbulence Based on the MTLM Approach as an Inlet Condition for LES

Abstract

To prescribe inlet conditions for the numerical simulation of turbulent flows using LES, it is not sufficient to specify second-order statistics, as in RANS, since part of the turbulent structures must be explicitly resolved. DNS or precursor simulations are usually used to obtain and impose realistic fields, which leads to a high computational cost. Synthetic turbulence generation, on the other hand, avoids the direct resolution of the Navier–Stokes equations, but its rapid dissipation, due to the lack of spatial correlations, as well as the absence of self-deformation mechanisms and interscale kinetic energy transfer, requires an extensive re-development region and a computational domain much larger than the region of interest. In this work, the MTLM approach, which generates synthetic fields with statistical, geometrical, and structural characteristics of homogeneous and isotropic turbulence, is extended to produce anisotropic and non-homogeneous fields, and these are imposed as inlet conditions in a channel flow modeled with LES. The results show a reduction in the length of the adaptation region compared to other methods, reducing the computational cost and maintaining the accuracy of the simulation.