Direct numerical simulation of microjets for turbulent boundary layer control
A direct numerical simulation approach is used to simulate an array of MEMS slot jets in a turbulent boundary layer for the purposes of flow control. Initial studies were used to first ascertain the correctness of the model and give insight into the performance characteristics of the devices. To this end, a flow and geometric parameter study was conducted on a 2-D simulation of the devices and the results used in the design of 3-D devices. Then, once the performance characteristics of the 3-D devices were determined, a series of parametric studies were conducted involving quasi-steady, periodic and single pulses into a turbulent boundary layer. Results indicate that low Re devices can substantially affect the flow but the arbitrary actuation tended to introduce new disturbances in the flow. Hybrid methods involving passive surface texture elements and active devices were briefly examined in an attempt to extract some form of useful flow control from the actuators but the results were still insufficient. Instead, a real-time, adaptive feedback method was selected to calibrate the strength of the actuators. The feedback method was first tested by itself to verify that the numerical approach used here was capable of implementing such control. The feedback method was then modified and used with a single, isolated slot. Subsequent simulations increased the number of slots to form a row of actuators and eventually actuator arrays. Results showed that an actively-controlled actuator array provided a small turbulent drag decrease despite having only a small impact on the turbulence levels in the flow. This result suggested a different mechanism of drag reduction from the original feedback method based on the presence of cavities which are seen as low-shear stress regions by the mean flow.