Abstract:
Direct numerical simulations (DNS) of turbulent channel flow over hexagonally-packed arrays of hemispheres, were performed at friction Reynolds numbers 200-600. The inner-scaled roughness height k^+ = 20 was maintained for all Reynolds numbers. The spacing between hemispheres was varied between d/k = 2-4. The analysis of amplitude modulation (AM) was employed in all rough-wall flows and showed enhanced modulation effect on the near-wall small-scale fluctuations imparted by the outer-layer large-scale motions compared to the smooth-wall flows. Based on the framework of AM, an empirical inner-outer predictive model was adapted in this work and extended to all three velocity components by exploiting the principal component analysis (PCA) to take into account the anisotropy effect in wall turbulence. Moreover, wall shear stress in terms of the roughness-cell averaged statistics was further investigated and a universal cell-averaged wall shear stress was extracted from the AM-based model to show more fundamental statistical behaviors without the strong influence from the outer-layer large-scale structures. The statistics of the universal wall shear stress were also explored using various roughness-cell sizes to advance turbulence modeling from the practical perspective of an actual ``wall-modeled'' large-eddy simulation (LES).
Bio:
Sicong Wu is a Ph.D. candidate in Theoretical and Applied Mechanics at the University of Illinois at Urbana-Champaign. He received his Bachelor of Science in Mechanical Engineering from the University of Minnesota at Twin-Cities. He has been working with Prof. Pantano since 2014 and his research topics include performing direct numerical simulation of turbulent boundary layers with complex surface geometry and analyzing flow physics from the perspective of inner-outer boundary layer interactions. Based on the flow interactions, an improved mathematical model can be proposed to better characterize wall turbulence and drag behaviors over rough surfaces.
Host: Professor Leo Chamorro
