Bulletin of the American Physical Society
67th Annual Meeting of the APS Division of Fluid Dynamics
Volume 59, Number 20
Sunday–Tuesday, November 23–25, 2014; San Francisco, California
Session G18: Vortex Dynamics: Vortex Identification and Mechanisms |
Hide Abstracts |
Chair: Karen Mulleners, Das Institut fuer Turbomaschinen und Fluid-Dynamik Room: 2004 |
Monday, November 24, 2014 8:00AM - 8:13AM |
G18.00001: Quantifying the reconnection process of two vortices Guillaume Beardsell, Louis Dufresne, Guy Dumas In this work, we use DNS to study the reconnection of two vortices. The Navier-Stokes equations are solved using a Fourier pseudospectral algorithm with triply periodic boundary conditions. The zero-circulation constraint, which was found to be problematic by Pradeep \& Hussain (2004), is circumvented by solving the governing equations in a proper rotating frame. To quantify the reconnection of two vortices, an approach using vortex filaments is considered. This approach is first validated against the results of Hussain \& Duraisamy (2011) for two parallel counter-rotating vortices. In this latter case, symmetries in the initial flow provide a simple way to compute the instantaneous rate of reconnection. Next, we study the interaction of orthogonal, unequal strength vortices for which only partial reconnection can occur. Typically, the weak vortex ($\Gamma_2$) is seen to deform and wrap itself around the strong one ($\Gamma_1$) to (partially) reconnect. For Reynolds numbers ($\Gamma_1/\nu$) of the order of $10^3$ and circulation ratios $0.1\leq\Gamma_2/\Gamma_1\leq0.9$, we compute the instantaneous reconnection rate and observe the propagating vorticity structures. Particularly, we look at some of the topological features that can be well visualized with vortex filaments. [Preview Abstract] |
Monday, November 24, 2014 8:13AM - 8:26AM |
G18.00002: Spectrum and Structure of Evolving Vortex Sheet V. Rajesh, O.N. Ramesh Shear layer structure and dynamics play an important role in understanding turbulent flows. The evolution of the Vortex sheet, considered as the inviscid/infinite Reynolds number approximation to the shear layer, has been studied in the literature for its many facets like Kelvin-Helmholtz instability, finite-time singularity and spiral structures formation. In the present work, a two-dimensional vortex sheet evolution is simulated using Krasny's vortex blob method in high precision. The nonlinear stage of evolution and the resulting spectrum are validated against the recent results of Abid {\&} Verga (2011). The obtained energy spectrum, showing similarities to two-dimensional turbulence spectrum, is investigated in detail. An attempt at simulating the evolution of three-dimensional disturbances on the vortex sheet and the effect of imposed strain is also presented. [Preview Abstract] |
Monday, November 24, 2014 8:26AM - 8:39AM |
G18.00003: Eulerian and Lagrangian methods for vortex tracking in 2D and 3D flows Yangzi Huang, Melissa Green Coherent structures are a key component of unsteady flows in shear layers. Improvement of experimental techniques has led to larger amounts of data and requires of automated procedures for vortex tracking. Many vortex criteria are Eulerian, and identify the structures by an instantaneous local swirling motion in the field, which are indicated by closed or spiral streamlines or pathlines in a reference frame. Alternatively, a Lagrangian Coherent Structures (LCS) analysis is a Lagrangian method based on the quantities calculated along fluid particle trajectories. In the current work, vortex detection is demonstrated on data from the simulation of two cases: a 2D flow with a flat plate undergoing a 45$^o$ pitch-up maneuver and a 3D wall-bounded turbulence channel flow. Vortices are visualized and tracked by their centers and boundaries using $\Gamma_1$, the $Q$ criterion, and LCS saddle points. In the cases of 2D flow, saddle points trace showed a rapid acceleration of the structure which indicates the shedding from the plate. For channel flow, saddle points trace shows that average structure convection speed exhibits a similar trend as a function of wall-normal distance as the mean velocity profile, and leads to statistical quantities of vortex dynamics. [Preview Abstract] |
Monday, November 24, 2014 8:39AM - 8:52AM |
G18.00004: Reconstruction of three-dimensional coherent structures in turbulent wakes using planar measurements Serhiy Yarusevych, Chris Morton The present study is focused on reconstructing the dynamics of dominant three-dimensional coherent structures in turbulent wakes of complex cylindrical geometries using time-resolved, planar Particle-Image-Velocimetry data. As a test case, the turbulent wake of a low aspect ratio dual step cylinder model is considered. The model consists of a large diameter cylinder ($D)$ of low aspect ratio ($L/D)$ attached to the mid-span of a small diameter cylinder ($d)$. Experiments are performed in a water flume facility for \textit{Re}$_{D}=$2100, $D/d=$2, and $L/D=$1. The investigated model produces cellular vortex shedding, with distinct variations in the average shedding frequency along the span of the model, and the associated complex vortex interactions. Time-resolved velocity measurements are acquired simultaneously in two mutually orthogonal planes at multiple planes along the span of the model. The technique involves conditional averaging of the planar results to produce three-dimensional reconstructions of wake topology for a given planar alignment of the dominant spanwise vortex filaments. This is achieved by identifying velocity fields matching a given flow-based template. The results demonstrate that the proposed technique can successfully reconstruct the dominant wake vortex interactions and can be extended to other flows where traditional phase-averaging approaches are not applicable. [Preview Abstract] |
Monday, November 24, 2014 8:52AM - 9:05AM |
G18.00005: The Effect of Phase Averaging Techniques on Lagrangian Coherent Structures in the Wake of a Circular Cylinder Matthew Rockwood, Melissa Green Experimental results of the wake of a circular cylinder were studied using Lagrangian coherent structures (LCS). The planar velocity data was collected at multiple Reynolds numbers over a range from 3,000 to 12,000 using a two-component DPIV measurement system. The data was phase averaged by binning velocity fields based on two reference quantities: vorticity centroid location in each snapshot, and pressure measurements on the cylinder surface. A Proper Orthogonal Decomposition (POD) was also applied to the velocity data to extract portions of the velocity field containing the most energy. Another set of phase averaged velocity fields were then generated using the vorticity centroid location of the POD reconstructed fields. The change in the LCS locations and the vortices identified using the Eulerian Q-criterion were found to be minimal. This investigation ensures that the most accurate, efficient phase averaging techniques are being used to study the LCS in the wake of the circular cylinder. [Preview Abstract] |
Monday, November 24, 2014 9:05AM - 9:18AM |
G18.00006: Identification of Vortex Breakdown in Bio-Inspired Wakes Using Proper Orthogonal Decomposition Zachary Berger, Justin King, Melissa Green In this investigation, the flow field of a bio-inspired wake is studied using stereoscopic PIV at the mid-span and quarter-span of a trapezoidal pitching panel in a water channel. Three-component planar velocity fields are generated immediately downstream of the panel. Standard (4 Hz) PIV measurements require phase-averaging to extract relevant flow features with respect to the time scales of the flow. In order to gain more insight into the energy content of the flow field as well as quantitatively identify the vortex breakdown, reduced-order modeling in the form of proper orthogonal decomposition (POD) is applied to the data. Two component vector POD allows for the extraction of the most energetic, large scale structures which can be used to reconstruct a low-dimensional representation of the flow field. This can then be compared to phase-averaged data to readily quantify vortex breakdown as a function of spanwise location in order to construct a model of the three-dimensional wake structure. [Preview Abstract] |
Monday, November 24, 2014 9:18AM - 9:31AM |
G18.00007: Helical vortices: viscous dynamics and instability Maurice Rossi, Can Selcuk, Ivan Delbende Understanding the dynamical properties of helical vortices is of great importance for numerous applications such as wind turbines, helicopter rotors, ship propellers. Locally these flows often display a helical symmetry: fields are invariant through combined axial translation of distance $\Delta z$ and rotation of angle $\theta = \Delta z/L$ around the same $z$-axis, where $2\pi L$ denotes the helix pitch. A DNS code with built-in helical symmetry has been developed in order to compute viscous quasi-steady basic states with one or multiple vortices. These states will be characterized (core structure, ellipticity, ...) as a function of the pitch, without or with an axial flow component. The instability modes growing in the above base flows and their growth rates are investigated by a linearized version of the DNS code coupled to an Arnoldi procedure. This analysis is complemented by a helical thin-cored vortex filaments model. [Preview Abstract] |
Monday, November 24, 2014 9:31AM - 9:44AM |
G18.00008: Evolution of mean flow and disturbances in strained vortices Yuji Hattori Evolution of disturbed strained vortices is studied by direct numerical simulation. We choose 2D flattened Taylor-Green vortices as a base flow and add a small wave packet which grows exponentially due to the elliptical instability. The evolution consists of three stages: the linear, non-linear, and turbulent stages. At the linear stage the wave packet located initially at the center of a vortex grows exponentially without significant change of the shape. At the nonlinear stage the wave packet collapses and small-scale structures develop. Concentration of vorticity in the mean flow, which is similar to the ``expulsion of vorticity'' in rotating turbulence, is observed before the transition to turbulence. Finally the flow becomes turbulent exhibiting the Kolmogorov energy spectrum although the mean flow is not far from the initial state. The mechanism behind the concentration of vorticity will be discussed in connection with angular momentum transfer and selective decay of inviscid invariants. [Preview Abstract] |
Monday, November 24, 2014 9:44AM - 9:57AM |
G18.00009: Tomographic PIV Study of Hairpin Vortices Daniel Sabatino, Tobias Rossmann Tomographic PIV is used in a free surface water channel to quantify the flow behavior of hairpin vortices that are artificially generated in a laminar boundary layer. Direct injection from a 32:1 aspect ratio slot at low blowing ratios ($ 0.1 < BR < 0.2$) is used to generate an isolated hairpin vortex in a thick laminar boundary layer ($485 < Re_{\delta^*} < 600$). Due to the large dynamic range of length and velocity scales (the resulting vortices have advection velocities 5X greater than their tangential velocities), a tailored optical arrangement and specialized post processing techniques are required to fully capture the small-scale behavior and long-time development of the flow field. Hairpin generation and evolution are presented using the $\lambda_2$ criterion derived from the instantaneous, three-dimensional velocity field. The insight provided by the tomographic data is also compared to the conclusions drawn from 2D PIV and passive scalar visualizations. Finally, the three-dimensional behavior of the measured velocity field is correlated with that of a simultaneously imaged, passive scalar dye that marks the boundary of the injected fluid, allowing the examination of the entrainment behavior of the hairpin. [Preview Abstract] |
Monday, November 24, 2014 9:57AM - 10:10AM |
G18.00010: Comparing wake structures behind a finite aspect ratio and an infinite span normal thin flat plate Arman Hemmati, David H. Wood, Robert J. Martinuzzi The wake of an infinite span (2D) thin flat plate and that of a finite aspect ratio, AR = 3.2, plate, both normal to a uniform stream, are compared using DNS at Re=1200. For the 2D plate, three wake flow regimes are observed. Intervals of typical anti-symmetric Karman shedding (Regime M) are interrupted by intervals where the shear layer folding process first delayed (Regime L) and then accelerated, Regime H. The distinct flow patterns in these regimes have significant impact on lift and drag values, wake structure and instantaneous pressure loads. In contrast, only Regime M is observed for the AR=3.2 plate. The presence of two lateral shear layers appears to maintain the Karman shedding. Compared to the infinite plate: the mean recirculation region shrinks by 45$\%$ to 1.57H; the magnitude of the Reynolds Stresses drops significantly; Turbulent kinetic energy levels along the wake centerline and peak production and dissipation rates are significantly lower. Further, the three normal Reynolds stresses are comparable in magnitude. To better understand the impact of additional shear layers on the wake stability and resultant wake structures, the 3D structures will be reconstructed using DNS results. Pressure and stress distribution along the plate surfaces will also be examined. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700