Active flow control of the turbulent boundary layer over a NACA4412 wing profile for skin friction drag reduction

Sammanfattning: In the context of building a framework for active flow control of turbulent boundary layers in wings, a set of large-eddy simulation (LES) are implemented in OpenFOAM. The flow around a NACA4412 wing profile is simulated at 5° angle of attack and Re_c = 400˙000. Validation of the uncontrolled flow results is performed with respect to the dataset generated by Vinuesa et al. (2018) at the same aerodynamic configuration. Afterwards, two different flow control strategies are analyzed over the suction side (SS) of the wing to yield skin friction drag reduction and an overall improvement of the aerodynamic efficiency. The region subject to the actuation spans 0.25 x_ss/c to 0.:86 x_ss/c, where c is the chord length of the wing. In the current setup, uniform blowing (BLW) and suction (SCT) control schemes show close agreement with the trends presented by Atzori (2021). Indeed, BLW decreases the viscous drag, but increases its pressure contribution and penalizes the lift, thus lowering the global efficiency of the wing, while SCT has an opposite effect. Thus, these methods behave similarly to pressure gradients (PGs) conditions, as BLW enhances the APG, whereas SCT damps it. The streamwise travelling waves strategy is then assessed for three set-ups characterized by different phase speeds. A consistent skin friction drag reduction and efficiency improvement are observed for two cases, while milder benefits are recorded even when drag increase was expected. Trends which have already been reported in the literature by Quadrio et al. (2009) and Skote (2014) are identified, i.e. the effects of this actuation to be mainly enclosed in the viscous sub-layer and the gross amount of drag reduction to be dependent on the wave relative speed; however, it is believed that the PGs conditions over the SS of the wing significantly alters the outcomes of the chosen parameters. Eventually, Reynolds averaged Navier-Stokes (RANS) simulations are performed to assess their accuracy with respect to the generated LES set-up, in the effort to enable a multi-fidelity approach for future works.