Numerical and experimental investigation of boundary layer transition with active and passive flow control methods

Drag reduction is one of the important research topics of today, and its prominence is increasing day by day due to the concerns about limited oil resources and global climate change. A modern airliner's total drag is composed of various components, but the friction drag, which accounts for about 60% of total drag, is the component having the largest potential in drag reduction among the other components. Moreover, a reduction in overall friction drag can be achieved by delaying laminar-to-turbulent transition. Application of a suitable flow control method may help to delay transition-to-turbulence and thus reduces friction drag. Therefore, in this study, two flow control methods, one active and the other passive, are separately applied to a parallel flow over a flat plate for the purpose of delaying laminar-to-turbulent transition and reducing the total drag. In the first part of the study, the effects of wall heating and cooling, one of the active flow control methods, applied to the flow over a 2 m long flat plate on laminar-to-turbulent transition and the drag reduction are parametrically investigated through two- and three-dimensional Computational Fluid Dynamics (CFD) simulations. The simulations are performed using a finite volume based flow solver ANSYS-FLUENT. A grid sensitivity study is conducted to obtain grid independent and accurate results before performing the final simulations. Furthermore, the effects of various Reynolds-Averaged Navier-Stokes (RANS) based turbulence models on the results are examined, and Menter's four equations Transition SST turbulence model is preferred for the final simulations due to its superiority in determining the laminar-to-turbulent transition. The simulation results are validated against the well-known correlations and experimental data available in the literature. Effects of various flow and design parameters such as free stream turbulence intensity, surface temperature, heater strip length and various heating/cooling combinations such as discrete heating, multiple heating, dual heating and discrete cooling on laminar-to-turbulent transition and drag reduction are parametrically investigated. It is found that drag acting on a flat plate can be reduced at certain rates by wall heating and cooling provided that their design parameters are carefully determined. However, these methods require an energy supply and therefore these active flow control methods can be used as an alternative method in drag reduction applications where the cost is not so important. In the second part of the study, the effects of the use of a roughness element, which is one of the passive flow control methods, on the stability of the flow over a flat plate are investigated both experimentally and numerically. In this context, the stability of a boundary layer induced by a three-dimensional rectangular prism shaped roughness element placed on a flat plate is parametrically investigated by means of a series of the BiGlobal Linear Stability Theory (LST) simulations. In the simulations, the aspect ratio of the roughness element and the Reynolds number are changed to investigate their effects on the instability mechanism behind the roughness element. The base flow required to perform the stability analysis is obtained using a finite volume based flow solver ANSYS-FLUENT. A comprehensive grid sensitivity analysis is carried out to obtain grid independent and accurate results before the ultimate computations. The simulation results are validated using a series of velocity measurements and hydrogen bubble flow visualization experiments in the Laminar Water Channel at the University of Stuttgart. The BiGlobal Linear Stability analyses are carried out to investigate the instability mechanisms behind the roughness elements using the in-house developed stability analysis code at the Institute of Aerodynamics and Gasdynamics (IAG) after obtaining an agreement between experimentally and numerically obtained base flow results. A grid optimization study is conducted before performing the stability analysis as did for the base flow. It is concluded considering the base flow results obtained for the investigated cases that the change in the Reynolds number did not appear to have a significant effect on the shape and length of the reversed flow regions emerged in front of and behind the roughness element whereas the effect of aspect ratio is significant. In addition, horseshoe vortex system whose legs are responsible for the formation of low- and high-speed streaks are observed in all the cases investigated within the content of this study. Furthermore, the BiGlobal LST results revealed that presence of the roughness element leads to the formation of several instability modes called sinuous and varicose, which are asymmetric and symmetric with respect to the spanwise direction. It is found that the streamwise evolution of these unstable modes differs depending on the Reynolds number and the aspect ratio of the roughness element. It is also concluded that the varicose mode is dominant for thin roughness elements whereas the sinuous mode is dominant for large elements. For the cases investigated, the flow is found to be stable within the range of the parameters studied. Furthermore, temporal evolution of the sinuous and the varicose is numerically obtained using the BiGlobal LST results and then these results are consolidated with a series of hydrogen bubble flow visualization experiments.