Control of boundary-layer separation using surface roughness
Boundary-layer separation is the detachment of a viscous boundary layer from the body surface. It is considered detrimental in most engineering applications, as it actively promotes transition to turbulence, and increases drag. Hence, the ability to control this phenomenon is considered crucial in many scientific applications. As flow separation occurs at the solid-fluid interface, it became natural to investigate the behaviour of fluids near a solid surface. In most research studies, a no-slip condition is imposed: the fluid sticks to the surface. However, it emerges that replacing the usual no-slip boundary condition with a certain surface roughness can be beneficial for flow control. Furthermore, the surface texture of real materials, which may appear smooth at visible scale, is, in fact, characterised by roughness at microscales. Macroscopically, surface roughness can be modelled in terms of slip boundary conditions. Thus, it emerges that a slip surface, not only depicts a more realistic scenario, but it can also provide beneficial effects on the control of boundary-layer separation.
This thesis aims to numerically explore surface roughness, modelled in terms of slip boundary conditions, as a method of controlling flow separation. Most of the work presented in this thesis exploits the capabilities of Nektar++, a spectral/hp element framework, to perform Direct Numerical Simulations (DNS) on a variety of flow configurations, where slip boundary conditions are applied.
A linear stability study on slip channel flows is undertaken. The surfaces are modelled exploiting a linear Navier slip condition imposed on the channel walls. The linearised stability of the flow is investigated via an Orr-Sommerfeld normal-mode approach, the numerical solution of which allows us to determine the level of slip at which the flow is stabilised, for all flow Reynolds numbers.
Two numerical studies on the effect of surface slip on the two-dimensional flow in a constricted, and dilated, channel are presented. The constriction is provided by bump, and dilatation by a smooth depression (gap), on the lower wall, upon which a Robin-type slip boundary condition is imposed. The effect of slip on the separation region formed behind the bump, and inside the gap concavity, is investigated. It is shown that surface slip attenuates the intensity of separation, delays the onset of separation (in the bump case), reduces the dimensions of the separation region, and reduces skin-friction and total drag. Ultimately, slip inhibits flow separation.
A numerical study on the effect of slip applied along two-dimensional small-scale Gaussian-shaped surface deformations on a at plate is shown. The deformations consist of isolated bumps and gaps of different dimensions and locations, and multiple-bump configurations. Slip surface is explored in these flow scenarios, and its control benefits are described.
Finally, the results obtained in this thesis are summarised. They provide the basis for future studies on the implementation of slip boundary conditions in more realistic, and three-dimensional geometries, such as air-wings and hulls. The outlook is to explore the use of surface roughness for flow control and skin friction reduction, hence achieve a more sustainable transport industry and deliver on future emissions reduction targets.
