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 L29: Turbulent Wakes 
Hide Abstracts 
Chair: Laurent Mydlarski, McGill University Room: Georgia World Congress Center B401 
Monday, November 19, 2018 4:05PM  4:18PM 
L29.00001: ScaleResolving Simulations (SRS) of a Square Cylinder undergoing VortexInduced Motion Thomas S. Fowler, IV, Sharath S. Girimaji The flow past a fixed square cylinder is an ideal configuration for benchmarking and validating scaleresolving simulations (SRS) and other computational turbulence approaches. This is due to the fact that the fixed flow separation location enables expedited analysis of unsteady vortex shedding and coherent structure generation. Flow past a square cylinder which is oscillating due to vortexinduced motion (flowstructure interaction) is of practical significance in a variety of ocean engineering applications. It is an ideal test case for studying fluidstructure interactions and the effects of vortexinduced motion. In this work, we perform simulations of flow past a square cylinder with and without free oscillations using the partiallyaveraged NavierStokes (PANS) SRS kω turbulence approach. We examine the effects of varying degrees of freedom of cylinder oscillations on vortex shedding and coherent structures in the wake. We also investigate the effects of different types of oscillations on flow statistics and integral quantities, such as Strouhal number, unsteady recirculation length, and coefficients of lift and drag. These simulations are performed at low to moderate Reynolds number regime. 
Monday, November 19, 2018 4:18PM  4:31PM 
L29.00002: ScaleResolving Simulations (SRS) of flow past a sphere at subcritical Reynolds number Chetna P Kamble, Sharath S Girimaji Partiallyaveraged NavierStokes (PANS) twolayer turbulence scheme offers a robust near wall closure for ScaleResolving Simulations (SRS). This technique couples a scaleresolved oneequation model near wall with a PANS kε model in the outer layer. In this study, we employ the PANS twolayer closure to examine flow over a sphere in the subcritical regime (Re = 3700). This flow embodies complex characteristics including laminar separation, transition to turbulence in the shear layer and vortex shedding. Comparison with Direct Numerical Simulation (DNS) reveals the dependence of the mean and integral flow parameters on numerical (spatiotemporal) as well as physical resolution (cutoff scales). Optimal physical resolution for capturing the vortexshedding frequency (f_{vs}) and KelvinHelmholtz instability (f_{KH}) is established. Furthermore, the largescale coherent structures shed in a helical configuration at random azimuthal locations on the sphere are visualized by means of velocity gradient invariants. 
Monday, November 19, 2018 4:31PM  4:44PM 
L29.00003: Decay of stratified turbulent wakes behind a disk at a moderately high Reynolds number Karu Chongsiripinyo, Sutanu Sarkar Largeeddy simulations of flow past a disk with Reynolds number of 50,000 and Froude number of 0.1,2,10,50,∞ are presented. The results are analyzed with a focus on the decay of turbulent wakes that are subject to strong, intermediate, and weak stratification. The axisymmetric wake is also investigated as it sets the initial flow state where buoyancy becomes important in Fr ≥ O(10) wakes. Re is sufficiently large so that turbulent production dominates over mean viscous dissipation in the mean kinetic energy budget. Nevertheless, it is found that the characteristic scales of turbulence (both r.ms. velocity and integral length scale) evolve differently from those of the mean deficit velocity. Therefore, contrary to classical selfsimilarity theory, turbulence and mean quantities obey different power laws. We will report on how buoyancy changes the relative behavior of mean and turbulence power laws and characterize the transition from 3D to the nonequilibrium (NEQ) regime. Energetics and mixing properties will be quantified in the regime of stratified turbulence which is wider than in previous bodyinclusive simulations. Changes, as a function of Fr, in the wave field generated by the body and by the wake will be described. 
Monday, November 19, 2018 4:44PM  4:57PM 
L29.00004: Stratified flow over a prolate spheroid Jose Luis OrtizTarin, Karu Chongsiripinyo, Sutanu Sarkar The effect of stratification on the boundary layer and the near wake of a 4:1 prolate spheroid is investigated using highlyresolved largeeddy simulations. The Reynolds number based on the minor axis is $Re=Ud/\nu=10^4$. The effect of stratification, characterized by the Froude number $Fr=U/Nd$, is studied by comparing unstratified ($Fr=\infty$), intermediately stratified ($Fr=3$) and strongly stratified ($Fr=1$) cases . 
Monday, November 19, 2018 4:57PM  5:10PM 
L29.00005: Flow over cylinders with porous outer layers: influences of freestream turbulence and porous layer depth and porosity Laurent Mydlarski, Alexandre Cohen, Donovan Blais, B. Rabi Baliga A better understanding of the mechanisms affecting the propagation of forest fires, which have increased in frequency and intensity over the past 1015 years, is essential for their effective mitigation and control. One such mechanism is spotting, in which burning leaves, branches, and other debris, termed firebrands, are transported away from the main fire by the wind and ignite new fires. Although spotting has been the subject of prior experimental and numerical studies, the influences on firebrand trajectory of both freestream turbulence and the porous char layer that forms on their outer surface have not been accounted for. As a first step of a more comprehensive study aimed at improving predictions of firebrand trajectories, wind tunnel experiments were conducted to quantify the effects of freestream turbulence and porous outer layers on the flow over stationary, nonburning cylinders. Passive and active grids were used to generate flows of different turbulence intensities, in which forcebalance and hotwire anemometry measurements were made. The dependence of the drag on the cylinder, as well as the flow in its wake, on the freestream turbulence and the porous outer layer (depth and porosity) will be presented. 
Monday, November 19, 2018 5:10PM  5:23PM 
L29.00006: Sidebyside cylinders in crossflow with freestream turbulence Raúl Bayoán Cal, Elizabeth Camp An experimental approach to examine the influence on freestream turbulence on the wakes issuing from a pair of sidebyside cylinders in crossflow is undertaken. These experiments are performed via particle image velocimetry in a closed loop wind tunnel. Both the ratio of transverse space to the diameter of the cylinder (T/D) as well as inflow turbulence intensity (Tu%) are varied in order to allow a thorough comparison of the changes due to freestream turbulence on differing canonical flow regimes. For each T/D chosen, a control case utilizing the ambient Tu% of the wind tunnel is measured and analyzed which is comparable to the low turbulence inflow conditions used in the literature to identify flow patterns that are characteristic of sidebyside cylinders in crossflow. Components of the timeaveraged mean velocity and the timeaveraged Reynolds stress tensor are compared in order to provide a detailed characterization of the similarities and differences between the flow fields in each of the twelve test cases. Proper orthogonal decomposition is also applied to identify relevant modes and organization in such flows. 
Monday, November 19, 2018 5:23PM  5:36PM 
L29.00007: Characterization of marine propeller model near wake in the presence of external (free stream) turbulence Bennitt Hermsen, Luksa Luznik High resolution, three dimensional PIV experiments were conducted in the in the US Naval Academy’s recirculating water tunnel (1.8m long test section with 0.41m x 0.41m cross sectional area) with threebladed marine propeller model (D=13cm). Spatial coverage in the near wake was accomplished via “tiling” of individual fields of view (FIV) of nominally 12x8 cm^{2} area covering downstream distance from the wake generator (propeller) plane up to approximately 9D downstream.Two separate set of measurements were made: phase locking, i.e. PIV image pairs were triggered at a specific blade location and “free runs” where no phase locking was performed. The resulting flow Reynolds number based on diameter is Re_{D}=2.6x10^{5}. Two parameters of the grid generated external (free stream) turbulence are controlled: the streamwise integral length scale of the turbulence and the turbulence intensity. Results will focus on characterizing spatial distributions of ensemble averaged mean wake velocities as well as spatial distribution of wake growth rates and normal and shear turbulent stresses up to 9D downstream from the rotor plane.

Monday, November 19, 2018 5:36PM  5:49PM 
L29.00008: PIV measurements of a Wake from a Rough Flat Plate Sean Lawrence, Julio Soria, Callum Atkinson The formation of wakes over smooth bodies has been extensively studied, however the same cannot be said for wakes that evolve from rough bodies. Surface roughness is known to greatly affect the behaviour of a turbulent boundary layer and the nature of the structures within. These rough turbulent boundary layers modify the formation of the wake and its spatialtemporal evolution. The present study used 2component  2dimensional particle image velocimetry (2C2D PIV) to measure the wake flow evolving from a smooth and rough flat plate with a blunt trailing edge in the LTRAC large horizontal water tunnel facility. Specifically, 2C2D PIV was acquired in the near wake, in addition to the turbulent boundary layer at the trailing edge of the plate. The boundary layer was tripped after the semielliptical leading edge and developed over the flat plate with a freestream velocity of 400 mm/s and Re_{θ }= 1.64×10^{8 } at the leading edge. The surface was roughened using graded particles adhered to the surface. First and second order statistics of these measurements will be presented. 
Monday, November 19, 2018 5:49PM  6:02PM 
L29.00009: Assessing intermittency characteristics via cumulant analysis in nearwake of offshore wind turbines. Hawwa Kadum, Stanislav Rockel, Bianca Viggiano, Tamara Dib, Michael Holling, Joachim Peinke, Raul Bayoan Cal Turbulence intermittency in the wake behind a single floating wind turbine (FWT) as well as merging wakes of a pair of floating turbines is investigated using cumulant analysis. Hotwire data are collected through wind tunnel experiments. First order cumulant cascades show the pitch motion effect to be apparent at the rotor tips and beyond the swept area of the rotor. At the rotor tips, a shorter inertial subrange is observed affecting the large scales and causing them to drift away from the inertial subrange. Intermittency in FWT is more pronounced in a fixed wind turbine wake. Intermittency coefficients for single FWT are higher than the typical range c2=−0.025±0.003. The pitch oscillations appear as fluctuations in the large scales portion of the cascade within the first cumulants. Also, other less frequent fluctuations exist in both fixed and floating WT tips caused by vortex shedding and formation of coherent structures. Coherent structures grow in size moving above the top tip (ATT) and beyond the horizontal tips (BHT). This observation is noted as larger fluctuations in the flow above and beyond the rotor tips in addition to longer inertial subrange. For two FWT, pitch fluctuations vanish in between the two turbines as wakes meet. 
Monday, November 19, 2018 6:02PM  6:15PM 
L29.00010: Wind Turbine Wake Merging Descriptions and Quantification Ryan Scott, Bianca Viggiano, Tamara S Dib, Naseem Ali, Michael Hölling, Joachim Peinke, Raúl Bayoán Cal Wind tunnel experiments were performed with three arrangements of model turbines in symmetric and asymmetric configurations. Asymmetric congurations included a single turbine upstream of two evenly spaced turbines and a single turbine downstream of two evenly spaced turbines. Hot wire anemometry was used to measure simultaneously instantaneous velocity signals across eleven spanwise locations at three distances downstream of the last row in the turbine array. Two inflow conditions with turbulence intensities of 4% and 10% were investigated. Arrangement symmetry is found to be a key parameter where the position of the furthest downstream turbine is the dominant factor when considering the downstream velocity field. Consideration of the energy thickness and shape factor revealed the wake region to extend as an energy and momentum deficit beyond the velocity deficit. The energy spectra near regions of wake interaction indicate production of turbulence at scales equivalent to the model rotor radius. In these areas, the flow is highly selfcorrelated and serves to enhance wake longevity downstream by injecting energy into the system. 
Monday, November 19, 2018 6:15PM  6:28PM 
L29.00011: Generalized wake model for dynamic wind turbine thrust and yaw actuation Charles Meneveau, Carl Shapiro, Dennice Gayme Simple dynamic wake models are needed to develop improved wind farm designs and for use in operational controllers to regulate wind farm power production and reduce structural loads. We develop an improved wake model for dynamic thrust and yaw actuation. From the Reynoldsaveraged NavierStokes, we derive a onedimensional partial differential equation model, where the forcing is specified by an inviscid lifting line approach describing the downstream evolution of the streamwise velocity deficit, transverse velocity, and centerline of the wake. We apply a mixinglength model for the eddy viscosity in the wake that leads to linear downstream wake expansion, with a rate specified using a topdown model for a developing wind turbine array boundary layer (Meneveau 2012, J. Turbulence) where the friction velocity evolves nonmonotonically downstream. The streamwise velocity deficits are distributed using a superGaussian function that smoothly transitions from a tophat profile close to the turbine to a Gaussian profile farther downstream. The resulting model reproduces relevant phenomena seen in wind tunnel experiments (Bastankhah & PorteAgel 2016, JFM) and largeeddy simulations. 
Monday, November 19, 2018 6:28PM  6:41PM 
L29.00012: Turbulent Cascade Features and Scale Imprint Characteristics in a Wind Plant NearWake Sarah E Smith, Naseem Ali, Raúl Bayoán Cal Scale evolution statistics are used to analyze the flow structure of the wake flow in the boundary layer of a wind turbine array. A 3 x 3 wind turbine array is tested experimentally via xtype hotwire anemometry, and provided a 9 x 21 of velocity signal downstream the rotor. It is observed that the structure evolution tends towards the non universality as the distance from the rotor is reduced. Flow structures of the wakes are characterized via correlation between the velocity and multifractal Holder exponent related to the velocity increment via FrischParisi conjecture. Flow events are also classified using the sign of the fluctuating streamwise velocity and the Holder exponent into four quadrants. A second and fourth quadrants of the velocity and the intermittency is dominant at the hub height and bottom tip. Holesize filtering shows the variations in the number of record per quadrant and finding the fourth quadrant dominant. Further, the statistical properties of velocityintermittency quadrant are interpreted in regards to the dependence between the velocity and velocity increments. 
Follow Us 
Engage
Become an APS Member 
My APS
Renew Membership 
Information for 
About APSThe American Physical Society (APS) is a nonprofit 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 207403844
(301) 2093200
Editorial Office
1 Research Road, Ridge, NY 119612701
(631) 5914000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 200452001
(202) 6628700