Subgrid-scale fluid statistics along the inertial particle trajectory in isotropic turbulence

Fluctuations of sub-grid scale (SGS) velocity and its influence on the motion of particles are important issues for large eddy simulation (LES) of gas-particle turbulence. We obtain the SGS statistics from the direct numerical simulation (DNS) data of homogeneous isotropic turbulent flow, which is filtered by the top filter in physical space. The result shows that the SGS kinetic energy seen by particles reach a minimum value when the Stokes number St closes to 1. Moreover, the turning point shifts towards a larger St as the filter width increases. Different from that observed in DNS, the Lagrangian integral time scale is larger than the Eulerian integral time scale in SGS. For small particles, the Lagrangian timescale of SGS fluid seen by particle is very close to that of the fluid itself. For particles which are large enough, it approaches the Eulerian timescale of SGS fluid. For the intermediate particles, the predicted curve of SGS fluid timescale seen by particle varies with St and its variation is non-monotonic. (authors)