(316ah) Controlling Wall Turbulence by Modifying the Channel Wall

Direct numerical simulations (DNS) of turbulent channel flow were used to investigate the structure of the velocity field close to a channel wall when the wall is modified in such a way that it either induces slip or induces a body force on the fluid next to the wall. There is a strong motivation to study this type of flows because of recent findings that demonstrate the possibility of turbulence control by wall modification.

Given recent reports that ultra-hydrophobic surfaces generate slip at the wall in laminar flows, the question arises whether turbulence drag reduction can be achieved over such surfaces. If so, how is the near-wall turbulence structure affected? This work examines the effects that a specified wall velocity (or wall slip) can have on the turbulence field. Direct numerical simulations of a Poiseuille-Couette flow channel were used. Computations were completed for Re=5200 (based on centerline mean velocity and channel width). The runs included the cases where one wall moved with 0, 1, 2 and 4 plus velocity units in the flow direction relative to the opposite channel wall. The mean velocity maximum shifted towards the moving wall as the wall velocity increased, as well as the point at which the Reynolds stress crossed zero. The turbulence intensity was lower close to the moving wall side. The velocity correlation coefficients showed that near-wall structures became longer.

Another way of controlling turbulence close to the wall is to apply a body force on the fluid, which can be a function of the distance from the wall. Lee and Kim (Phys. Fluids, 14(7), 2523-2527, 2002) have observed drag reduction in the case when a body force dampens the velocity fluctuations in the spanwise direction. In the present study, we applied a body force on the fluid close to the wall on all directions and we compared the results on the turbulence structure to those for the case of imposing a slip at the wall (as described in the previous paragraph). We see that a very small body force, one that diminished to a negligible value beyond 1.5 wall length units from the wall, can have similar impact as a slip of 4 plus velocity units. The location of the mean velocity maximum shifted towards the wall that exhibited the body forces, and the correlation coefficients showed elongated near-wall structures. The Reynolds stress was also smaller close to the wall that exhibits these forces.

The paper will compare the two cases of wall-induced modification on the turbulence velocity field, and will discuss the implications on drag reduction.