Seminars

In this talk, I will speak about some recent direct numerical simulation results on two related problems in wall bounded flows.

The first is on the transition to turbulence in boundary layers. In this work done with Profs. David Goldstein and Garry Brown, we first examine the different stages of transition caused by a discrete roughness element using immersed boundary DNS. We demonstrate that after the initial receptivity, spatial amplification of the steady distortion, and the subsequent modal instability of the lifted up structure, transition proceeds via a mutual, local, amplification of the fluctuating streamwise and spanwise vorticity. We show that this process is responsible for the amplification of streamwise vorticity near the wall, leading to the formation of familiar hairpin vortices. The subsequent breakdown to chaos and counter gradient transport of mean vorticity to the wall is discussed. We study whether these mechanisms are common to other routes to transition, such as free stream driven bypass transition, thus remark on their potential ‘universality’.

The second part of the talk will about the work done with Profs. Goldstein and Robert Handler on the possibility of flow control by disrupting individual drag-producing coherent vortical structures. Polymer molecules have been known to disrupt such vortical structures and hence lead to drag reduction. Majority of the previous studies have considered the polymer to be uniformly present in the flow, or involved bleeding the polymer into the entire near-wall turbulent boundary layer. While such a method is moderately useful for internal flows, the large quantities of polymer required for external flows impose serious cost and pollution constraints. We examine the effect of the hypothetical ability to place the polymer only within or immediately adjacent to specific drag producing vortical structures. Direct numerical simulations of a hairpin vortex in a channel flow are performed and a scalar field is introduced in its neighborhood. On entering the vortical structure, the scalar field is observed to remain within the structure for the duration of the simulation. While this is a consequence of Kelvin’s theorem, we demonstrate its applicability to the near wall viscous regime. Our simulations also show that if the scalar field is associated with a FENE-P fluid concentration, the hairpin is disrupted, suggesting the viability of this targeted control.

Short bio:

Dr. Saikishan Suryanarayanan is a lecturer in the Department of Aerospace Engineering and Engineering Mechanics, and a postdoctoral fellow working with Prof. David Goldstein at of the University of Texas at Austin. He received his PhD in Engineering Mechanics from JNCASR Bangalore, working on turbulent free shear layers with Prof. Roddam Narasimha. He has Masters in Mechanical Engineering from Texas A&M University, College Station TX. He is interested in transitional and turbulent flows, flow control, computational fluid mechanics and turbo machinery.