Bulletin of the American Physical Society
71st Annual Meeting of the APS Division of Fluid Dynamics
Volume 63, Number 13
Sunday–Tuesday, November 18–20, 2018; Atlanta, Georgia
Session G39: Turbulence: General II |
Hide Abstracts |
Chair: Maziar Hemati, University of Minnesota Room: Georgia World Congress Center Ballroom 3/4 |
Monday, November 19, 2018 10:35AM - 10:48AM |
G39.00001: Model-based fusion of PIV and hot-wire measurements for reconstruction of wall-bounded turbulent flows C Vamsi Krishna, Mengying Wang, Maziar Sam Hemati, Mitul Luhar Laboratory instrumentation is often limited in its ability to fully resolve the spatio-temporal fluctuations associated with turbulent flows. Particle Image Velocimetry (PIV) systems capable of fine spatial resolution are often limited to low temporal resolution, while instruments capable of high temporal resolution (e.g., hot wire anemometers) are restricted to point measurements. In this study, we employ models derived directly from the governing Navier-Stokes equations to reconstruct velocity fields in the intermediate time horizon between non-time resolved PIV snapshots, with input from time-resolved point measurements. Models grounded in rapid distortion theory are used to integrate the velocity field both forwards and backwards in time from the PIV snapshots to generate flow field predictions at high spatio-temporal resolution. Point measurements are then used to estimate the relative weights of the forward- and backward-time predictions. Direct numerical simulation data for turbulent channel flow from the Johns Hopkins Database is used to validate the sensor fusion approach, and to evaluate reconstruction accuracy. As proof-of-concept for real-world systems, this framework is also being used to reconstruct flow fields for ongoing turbulent boundary layer experiments. |
Monday, November 19, 2018 10:48AM - 11:01AM |
G39.00002: Physics-based multi-sensor fusion for statistically optimal reconstruction of wall-bounded turbulence Mengying Wang, C Vamsi Krishna, Mitul Luhar, Maziar Sam Hemati High-resolution spatiotemporal measurements of wall-bounded turbulence can be challenging to obtain in experiments. Instrumentation that can achieve the requisite temporal resolution (e.g., hot-wire anemometry) is typically confined to point measurements that restrict spatial fidelity; likewise, systems capable of obtaining spatially-resolved field measurements (e.g., PIV) usually lack the sampling rates required to achieve adequate temporal fidelity. In this study, we present a Bayesian estimation framework to fuse noisy multi-rate and multi-fidelity sensor measurements with uncertain predictions from a physics-based fluid dynamics model--derived using Rapid Distortion Theory. A “fast” Kalman filter is designed to fuse model predictions with high-rate point measurements; a “slow” Kalman filter is then used to fuse these time-resolved estimates with sub-Nyquist-rate field measurements to maintain spatial fidelity of the reconstruction. The method is demonstrated on a turbulent channel flow using direct numerical simulation data from the Johns Hopkins Turbulence Database. Optimal point-sensor placement is also investigated. Overall, the physics-based multi-sensor fusion approach yields unbiased and minimum variance spatiotemporal reconstructions of wall-bounded turbulent flows. |
Monday, November 19, 2018 11:01AM - 11:14AM |
G39.00003: Active grid excitation for tailored turbulence generation Lars Neuhaus, Lars Kröger, Joachim Peinke, Gerd Gülker, Michael Hölling Wind tunnel experiments are essential to understand flow phenomena or investigate new aerodynamic concepts. Especially for investigations of the interaction of wind turbines with the flow it is crucial to create realistic turbulence in the wind tunnel. To do so active grids are used, which allow to modulate the flow velocity and hence to reproduce the stochastics of atmospheric turbulence on a wide range of time scales. Herein a simple approach is shown to create various turbulent wind fields with different stochastic behaviours, on e.g. one-point and increment probability density functions and power spectral density. In this approach the excitation of the individual active grids shafts is defined by a Langevin time series. By varying the drift and diffusion coefficient of the Langevin time series the stochastics of the flow can be varied. In combination with a characterisation of the active grid behaviour a prediction of the flows one point and increment statistics on the computer is possible. This allows for an easy a priori tailoring of different turbulent wind fields for wind tunnel experiments. |
Monday, November 19, 2018 11:14AM - 11:27AM |
G39.00004: Axially varying sand grain roughness in turbulent Taylor-Couette flow Dennis Bakhuis, Pim Bullee, Rodrigo Ezeta, Dominic Tai, Sander Huisman, Detlef Lohse, Chao Sun We investigate the effect of axially alternating rough and smooth surfaces in a Taylor-Couette apparatus with Reynolds numbers up to 2×10^{6}. We employ realistic roughness by applying industrial, waterproof sandpaper (grit P36 with a measured standard deviation of the roughness height k_{rms} = 170 μm) on the inner cylinder of the Twente Turbulent Taylor-Couette facility. We vary s/d from 0.42 to 5.05 where d is the gap width of the Taylor-Couette and s the axial length scale of the rough and smooth patches. By varying s/d we can tune the size of the Taylor rolls and thereby the overall drag (C_{f} or, equivalently, Nu_{ω}) the inner cylinder experiences. We will discuss how the drag as a function of the driving speed depends on s/d and we will connect these results to our local PIV measurements. |
Monday, November 19, 2018 11:27AM - 11:40AM |
G39.00005: Fluctuation-induced forces in homogenous isotropic turbulence Vamsi Spandan, Daniel Putt, Alpha A Lee, Rodolfo Ostilla Monico In this talk, we uncover a fluctuation-induced force between two plates immersed in homogeneous isotropic turbulence using Direct Numerical Simulations. The force is a non-monotonic function of plate separation. The mechanism of force generation reveals an intriguing analogy with fluctuation-induced forces: energy in the fluid is localised in regions of high vorticity, or "worms", which have a characteristic length scale. The magnitude of the force depends on the packing of worms inside the plates, with the maximal force attained when the plate separation is comparable to the characteristic worm length. We explore the effect of different plate sizes and shapes on the force, as well as in more generalised objects. We ask the question whether these "worms" can induce other fluctuation forces on asymmetrical objects. |
Monday, November 19, 2018 11:40AM - 11:53AM |
G39.00006: An investigation of the structure-conditioned turbulence kinetic energy budget. Pawel Baj, James R Dawson, Nicholas A Worth, John M Lawson, Eberhard Bodenschatz An attempt is made to identify links between the turbulence kinetic energy (TKE) balance in homogeneous isotropic turbulence (HIT) and the underlying local flow topology. This is achieved through conditioning terms of the TKE budget on invariants of the velocity gradient tensor (Q_{A}, R_{A}), the strain-rate tensor (Q_{S}, R_{S}) and the rotation-rate tensor (Q_{Ω}). These conditional quantities reveal preferences of the particular TKE budget’s terms to be low or high depending on the specific areas of the Q_{A}-R_{A}, Q_{S}-R_{S} and Q_{S}-Q_{Ω} diagrams. In order to present the evolution of the TKE balance during the mean temporal development of structures, terms of the former are calculated along the conditional mean trajectories evaluated in Q_{A}-R_{A}, Q_{S}-R_{S} and Q_{S}-Q_{Ω} spaces. The data used to produce the statistics was gathered in a centre region of a von Karman mixing tank, which forms the suitable experimental approximation of HIT. The dataset consisted of nearly 200 000 independent samples of a cubic box of (40η)^{3} ensuring satisfactory convergence. The DNS data from the John Hopkins University turbulence database is used for comparison. |
Monday, November 19, 2018 11:53AM - 12:06PM |
G39.00007: Assessment of Taylor's hypothesis in the presence of mean shear using multi-camera particle image velocimetry Yanchong Duan, Danxun Li, Ellen K. Longmire To quantify the validity of Taylor's hypothesis and its effects on estimates of coherent motions in flows with mean shear, experiments were performed in streamwise-wall-normal planes in a turbulent open channel flow over a smooth wall at two conditions (Re_{τ}=500, 1000). Time-resolved particle image velocimetry measurements were obtained with multiple cameras covering a long streamwise field of view of 12 water depths. This streamwise extent enabled visualization of long streamwise structures in the flow and facilitated the investigation of Taylor's hypothesis. By taking the local mean velocity and global convection velocity (estimated by space-time correlation) as the convection velocity for Taylor's hypothesis, two kinds of reconstructed velocity fields can be obtained and compared with real spatial velocity fields. Results of individual fields and correlation analysis will be presented and discussed including correlation coefficients and scale estimation of coherent motions. |
Monday, November 19, 2018 12:06PM - 12:19PM |
G39.00008: Distinguishing coherence and randomness in turbulence using excess entropy Huixuan Wu, Xingtian Tao Turbulent flow involves both coherent motion and random fluctuation, and emphasizing only one of these two aspects is inadequate to understand the flow’s behavior. In this research, we utilize excess entropy to study the relationship between coherence and probability in a turbulent wake flow downstream of a cylinder, where the Re number is 8100. The unpredictability is quantified by entropy-per-signal and the coherence is characterized using excess entropy. In the stream-wise direction, the entropy-per-signal of the vertical velocity component keeps increasing, and at the same time, the excess entropy of that velocity component, Ev, decreases. This observation is consistent with the fact that the large-scale organized structures dissociate and the flow becomes more random. On the other hand, the wake flow is highly anisotropic. The excess entropy of the horizontal velocity fluctuation, Eu, has no significant variation in the stream-wise direction because this velocity component is largely controlled by small-scale structures, which do not undergo a significant change. Quantification of the coherence and randomness in a flow provides a new perspective to study turbulence. |
Monday, November 19, 2018 12:19PM - 12:32PM |
G39.00009: An experimental investigation of the turbulent cascade in a von Kármán mixing tank Anna Knutsen |
Monday, November 19, 2018 12:32PM - 12:45PM |
G39.00010: Abstract Withdrawn We present the results of local small-scale measurements of a vertical forced plume ejecting into a linear stably stratified environment, with a stratification strength of N= 0.32s^{-1}. High-resolution measurements of velocity and density fields were acquired using simultaneous Particle Image Velocimetry (PIV) and Planar Laser Induced Fluorescence (PLIF). The refractive indices of the ambient and the plume fluids were matched to avoid optical aberrations. The energetics of an evolving forced plume were studied by measuring the terms in kinetic energy budget terms, such as mechanical production (P), buoyancy flux(B), and dissipation (ε). The P and ε magnitude decreases with increase in the height, while the magnitude of buoyancy flux, B, decreases with height and becomes zero at the neutral buoyant layer (Z_{s}). Below the Z_{s}, the buoyancy flux, B, shows presence of unstable motions and its sign changes as the plume intrudes into the neutral buoyant layer, indicating stability. The flux Richardson's number (or mixing efficiency), Ri_{f}, which quantifies the local mixing in the plume region with the ambient, was also measured. Near the source, Ri_{f} revealed that the plume is dominated by momentum mixing, which transitions to scalar mixing as the plume moves vertically upwards. |
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. |
© 2019 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
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700