The PELskin project: part IV—control of bluff body wakes using hairy filaments

Abstract

The passive control of bluff body wakes using a sparse layer of elastic hairy filaments has been investigated via a series of numerical simulations and compared to selected experiments under well-controlled boundary conditions. It has been found that a distribution of filaments spaced half of the dominant three dimensional instability and resonating with the main shedding frequency can drastically delay the three dimensional transition of the wake behind a circular cylinder. It will also be shown that when using a pair of rows of filaments symmetrically spaced by an azimuthal angle, the wake topology can be deeply affected as well as the value of the integral force coefficients of the cylinder. In the most favourable case, a coupled three dimensional transition delay and strongly reduced values of the drag and of the lift fluctuation can be simultaneously achieved. These results hold also for higher Reynolds-number flows as shown in experiments on a cylinder with hairy flaps attached to the aft part. The lock-in effect of structural vibration of the flaps with the vortex shedding is assumed to be the reason for a sudden change in the shedding cycle as soon as the motion amplitude is high enough to modify the wake. In line with this hypothesis, it has been demonstrated that a long elastic filament pinned on the centerline of a forced spatially developing mixing layer can interact with the vortex dynamics delaying the pairing process-leading to a reduced thickness of the layer. These findings show that a properly designed fluid structure interaction can indeed lead to technological benefits in terms of wake control: drag reduction, vibration control and possibly palliation of aeroacoustic emissions.

Numerical study of filament suspensions at finite inertia

“…Both the frequency and the amplitude of the oscillations match the computational and experimental data reported in Pinelli et al. (2016). We observe that the difference between the two numerical approaches is small; in particular, the slightly different amplitude may be due to the different numerical approaches and the uncertainty of the experiments.…”

“…( a ) Time history of the displacement of the filament free-end in an oscillatory flow (solid line) compared to the experimental (dashed line) and numerical (triangles) results by Pinelli et al. (2016). ( b ) Instantaneous visualisation of the row of filaments in the half-channel.…”

“…Note that, in the case of neutrally buoyant filaments, i.e. (ρ f /A f − ρ) = 0, the left-hand side of (2.3) vanishes and in order to avoid the singularity of the coefficient matrix, the discretised equation (23) is modified as in Pinelli et al (2016) as The red circles denote the Lagrangian points for which the position of the filaments is defined and the blue diamonds show the Lagrangian points used to compute tension.…”

Flowing fibers as a proxy of turbulence statistics

“…Fig. 2b shows the streamwise displacement of the tip of the last filament with respect to time: our results (solid line) are compared with the experimental measurements (red dots) and with the simulations (blue dots) reported by Pinelli et al [34]. Both the frequency of oscillation and the magnitude of the displacement match the literature results.…”

Effect of Inclination on Vortex Shedding Frequency Behind a Bent Cylinder: An Experimental Study

“…Despite the fact that actual slender structures are frequently not in cross-flow, but otherwise inclined with respect to the free-stream, most of the available literature on vortex shedding and vortex-induced vibration is concerned with circular cylinders in cross-flow, and comparatively fewer studies have investigated the effect of the cylinder inclination [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. Recently, studies of inclined cylinders in flow at relatively low Reynolds numbers (Re < 1000) have experienced renewed interest in connection with flow sensing [20,21], flow control [22][23][24], and energy harvesting [25,26] applications involving slender flexible structures that reconfigure or bend due to fluid loads. The cylinder inclination angle α is normally defined as the angle between the incoming flow direction and the cylinder axis, so that α = 90 • corresponds to normal incidence and therefore to cross-flow, whilst when α = 0 • the cylinder axis is aligned with the free-stream velocity and the flow is purely axial.…”