Subgrid modelling for particle-LES by Spectrally Optimised Interpolation (SOI)

This paper focuses on Large-Eddy simulation of particle-laden flow. We present a novel model for the effect of unresolved scales on the particles. The model can be regarded as an extension of the approximate deconvolution method towards higher wavenumbers. The basis of the model is a specific interpolation method which is constructed such that the spectrum seen by the particles attains a model spectrum. Thus, the model is called Spectrally Optimised Interpolation (SOI). The model is developed and tested in the framework of homogeneous isotropic turbulence. A comparison of SOI against ADM [J.G.M. Kuerten, Subgrid modelling in particle-laden channel flow, Phys. Fluids 18 (2006) 025108] shows that in particular in coarse LES, SOI is far more accurate than ADM. The computational costs for SOI are comparable to fourth order interpolation. Possible extensions of the model for general flows are briefly sketched.

► The effect of unresolved scales on particles in turbulent flows is modelled. ► Particles are treated in a Lagrangian way within an Eulerian flow simulation. ► The spectrum of the flow seen by the particles is optimised by the interpolation. ► The model is tested in isotropic turbulence. ► The model is far more accurate than previously published comparable models.