主催: 一般社団法人 日本機械学会
会議名: 2023年度 年次大会
開催日: 2023/09/03 - 2023/09/06
We developed an LES analysis method for viscoelastic fluid turbulence by extending the Smagorisky model for LES of Newtonian fluid turbulence. In the extended model, we perform the time-marching correcting the elapsed time at each local location in the flow field so that the time scale for repeating increases and decreases in the intensity of vortex structures in the viscoelastic fluid turbulence is the same as that in the Newtonian fluid turbulence. The amount of the correction at each location is estimated by using a subregion whose size is large enough to properly capture the characteristics of the vortex structure. The size of the subregion is determined from the length of time variation in the minimal flow unit, which is previously performed for each condition of viscoelastic fluid turbulence. In this study, we performed DNS and LES for different types of viscoelastic fluid turbulence with different parameters set in the Giesekus and FENE-P models. By comparing the results, it was shown that LES with the extended model reproduced the DNS results more accurately than the conventional Smagorinsky model. It was proven that the extended model can predict wall turbulence with drag reduction in viscoelastic fluids over a wide range of conditions.