Vladimir Parezanović

On experimental sensitivity analysis of the turbulent wake from an axisymmetric blunt trailing edge

The sensitivity to local disturbances of the turbulent wake over a 3D blunt body with an axisymmetric detachment is investigated at Re = 2.1·10 4. The flow presents a favored m = 2 azimuthal symmetry set by two wings. The instantaneous wake is measured either above or below the plane containing the wings but leads a statistical symmetric wake. Topology shifts are random but occur mostly after a large number of global mode periods. The statistical symmetry is highly sensitive to any asymmetric disturbance. As a consequence, depending on its position a small control cylinder in the close wake fixes the wake to one asymmetric topology affecting shedding activity and drag. The effect of an axisymmetric perturbation (m = 0) on flow topology and dynamics is also studied ; it induces significant drag reductions and global mode modifications when acting on mixing layers. Whatever the disturbance, the sensitivity of the wake seems concentrated into the mixing layers and may depend more on their local turbulent characteristics than on the inviscid dynamics of vorticity.

The impact of a local perturbation on global properties of a turbulent wake

Stationary perturbation techniques are used to investigate the sensitivity of the global properties of the wake behind a bluff body at moderate Re. The entire recirculation bubble is found to be a sensitive region for the global frequency selection and the quality of the synchronization. A striking position is found at its center where synchronization is destroyed and the frequency is increased. High speed particle image velocimetry sheds some light into the underlying mechanisms, which are interpreted in terms of vorticity interaction, flow reattachment, and flow deviation. Strong mean flow modifications are observed to correspond to drastic changes in drag and its fluctuations.

Closed-loop control of experimental shear flows using machine learning

We propose a novel closed-loop control strategy of turbulent flows using machine learning methods in a model-free manner. This strategy, called Machine Learning Control (MLC), allows – for the first time – to detect and exploit all enabling nonlinear actuation mechanisms in an un-supervised automatic manner. In this communication, we focus on MLC applications for in-time control of experimental shear flows and demonstrate how it outperforms state-of-the-art control. In particular, MLC is applied to three different experimental closed-loop control setups: (1) the TUCOROM mixing layer tunnel, (2) the Gortler PMMH water tunnel with a backward facing step, and (3) the LML Boundary Layer wind tunnel with a separating turbulent boundary layer. In all three cases, MLC finds a control which yields a significantly better performance with respect to the given cost functional as compared to the best previously tested open-loop actuation. We foresee numerous potential applications to most nonlinear multiple-input multiple-output (MIMO) flow control problems, particularly in experiments. In particular, the model-free architecture of MLC enables its application to a large class of complex nonlinear systems in all areas of science.

Modulated global mode of a controlled wake

The 3D properties of the Benard von Karman global mode of the wake behind a 2D bluff body are investigated in an experimental study when a smaller, circular control cylinder is placed in the wake. Previous investigations [1, 2, 3], have shown the influence of a small control cylinder on the global frequency of the wake. The main effect is related to the interaction of vorticity produced by the control cylinder with the vorticity created by the bluff body. However, for certain positions of control, the envelope of the local velocity signal shows a modulation of amplitude. The search in the span-wise direction has shown the effects of the control cylinder’s presence on existence and location of the modulation.