When in a large-eddy simulation of particle-laden flow the filtered fluid velocity is used in the equation of motion for the particles,an extra error is introduced with respect to direct numerical simulations. In this paper this error, which can be separated in asubgrid error and an interpolation error, is assessed by means of a priori analysis of particle-laden horizontal channel flow forvarious values of Stokes number, settling velocity and filter width. The subgrid error increases in cases where the turbulence hasa larger influence on particle behavior, for example for smaller particle relaxation time. In such cases partial defiltering of thefiltered fluid velocity improves the results. The interpolation error is negligible if fourth-order interpolation is used for the fluidvelocity at the particle position.
|Title of host publication||Two-phase flow modelling and experimentation 2004 : proceedings of the 3rd international symposium on two-phase flow modelling and experimentation, Pisa, Italy, 22-25 September 2004|
|Place of Publication||Pisa, Italy|
|Publication status||Published - 2004|
Kuerten, J. G. M. (2004). A priori testing of large-eddy simulation of particle-laden channel flow. In G. P. Celata (Ed.), Two-phase flow modelling and experimentation 2004 : proceedings of the 3rd international symposium on two-phase flow modelling and experimentation, Pisa, Italy, 22-25 September 2004 (pp. 10-). Edizioni ETS.