Experimental study on dominant vortex structures in near-wall region of turbulent boundary layer based on tomographic particle image velocimetry. (10th September 2019)

• Record Type: Journal Article
• Authors: Wang, Chengyue; Gao, Qi; Wang, Jinjun; Wang, Biao; Pan, Chong
• Is Part Of: Journal of fluid mechanics. Volume 874(2019)
• Page Start: 426
• Page End: 454
• Publication Date: 2019-09-10
• Subjects: turbulent boundary layers; Fluid mechanics -- Periodicals; 532.005
• DOI: 10.1017/jfm.2019.412
• Languages: English
• ISSNs: 0022-1120

Abstract

Abstract : Vortex structures are very popular research objects in turbulent boundary layers (TBLs) because of their prime importance in turbulence modelling. This work performs a tomographic particle image velocimetry measurement on the near-wall region ( $y<0.1\unicode[STIX]{x1D6FF}$ ) of TBLs at three Reynolds numbers $Re_{\unicode[STIX]{x1D70F}}=1238$, 2286 and 3081. The main attention is paid to the wall-normal evolution of the vortex geometries and topologies. The vortex is identified with swirl strength ( $\unicode[STIX]{x1D706}_{ci}$ ), and its orientation is recognized by using the real eigenvector of the velocity gradient tensor. The vortex inclination angles in the streamwise–wall-normal plane and in the streamwise–spanwise plane as functions of wall-normal positions are investigated, which provide useful information to speculate on the three-dimensional shape of the vortex tubes in a TBL. The difference between the orientations of vorticity and swirl is discussed and their inherent relationship is revealed based on the governing equation of vorticity. Linear stochastic estimation (LSE) is further deployed to directly extract three-dimensional vortex models. The LSE velocity fields for ejection events happening at different wall-normal positions shed light on the evolution of the topologies for the vortices dominating ejection events. LSE based on a centred prograde spanwise vortex provides a typical packet model, which indicates that the population density of the packets in a TBL is large enough to leave footprints in conditionally averaged flow fields. This work should help to settle the severe debate on the existence of packet structures and also lays some foundation for the TBL model theory.