Turbulent flow control via nature inspired surface modifications

Abstract: Many of the flows in nature are turbulent. To modify turbulent flows, nature serves itself with different types of coatings. Sharks have riblets-like structures on their skin, fishes have slime with polymers and the surface of the lotus flower has superhydrophobic properties. However many times these naturally occurring coatings also serve other purposes. Due to millions of years of adaption, there are anyway many reasons to be inspired by these. The present work is an investigation of nature inspired coatings with the aim of passive flow manipulations. The goal of the investigation has not been to achieve drag reduction, but to achieve a better understanding of the effect of these coatings on turbulent flows. Simulations have been performed in a channel flow configuration, where the boundary condition on one wall has been modified. A macroscopic description has been used to simulate superhydrophobic and porous-like surfaces and a microscopic description has been used to simulate suspended fibers, both rigid and flexible, attached to the channel wall. For the macroscopic description, a pseudo-spectral method was used and for the microscopic description a lattice-Boltzmann method was used. The superhydrophobic modification was implemented using a general slip tensor formulation. In agreement with earlier results, drag reduction was achieved with slip in the streamwise direction and slip in the spanwise direction resulted in drag increase. Non-zero off-diagonal terms in the slip tensor resulted in a slight drag increase, but with rather similar flow behaviour. Transpiration, imitating a porous media, gave rise to drag increase and severely modified the turbulent structures, forming two-dimensional structures elongated in the spanwise direction. For the short fibers, neither rigid nor flexible fibers modified the velocity field to a large extent. The fibers gave rise to recirculation regions and these were seen to be stronger below high-speed streaks. Flexible fibers showed similarities to porous media through a coupling of wallnormal velocity and pressure fluctuations, and this was not seen for the rigid fibers. The fiber deflections were seen to correlate well with the pressure fluctuations.