Bulletin of the American Physical Society
62nd Annual Meeting of the APS Division of Fluid Dynamics
Volume 54, Number 19
Sunday–Tuesday, November 22–24, 2009; Minneapolis, Minnesota
Session HC: Turbulence Simulations V |
Hide Abstracts |
Chair: Pino Martin, Princeton University Room: 101C |
Monday, November 23, 2009 10:30AM - 10:43AM |
HC.00001: Direct numerical simulation of transonic shock wave/boundary layer interaction Matteo Bernardini, Sergio Pirozzoli, Francesco Grasso The interaction of a normal shock wave with a turbulent boundary layer over a flat plate at $M_{\infty} = 1.3$, $Re_{\theta} = 1300$ is investigated by means of DNS. The mean flow pattern consists of an upstream fan of compression waves associated with the thickening of the boundary layer and the supersonic region is terminated by a nearly-normal shock foot, which is observed to be significantly bent away from the wall. At the selected conditions the flow does not exhibit separation in the mean. However, the interaction region is characterized by unsteady instantaneous flow reversal in a very large zone, extending for many (upstream) boundary layer thicknesses past the nominal location of the interacting shock, and by the unsteady release of large vortical structures. The adverse pressure gradient tends to suppress the span-wise fluctuations in the near wall region across the interaction, thus yielding two-component isotropic turbulence in the neighborhood of the wall. Frequency-spectra of the wall pressure signature in the interaction region do not reveal the existence of organized low-frequency motions. [Preview Abstract] |
Monday, November 23, 2009 10:43AM - 10:56AM |
HC.00002: Analysis of Low-Frequency Unsteadiness in Shockwave and Turbulent Boundary Layer Interaction Stephan Priebe, M. Pino Martin The direct numerical simulation (DNS) of a compression-ramp shockwave and turbulent boundary layer interaction is presented. The simulation covers significantly more periods of the characteristic low-frequency shock motion than a previous simulation, which has been validated against experiments.\footnote{Wu, M. \& Martin, M. P., AIAAJ, vol.45, pp. 879-889, 2007}$^,$\footnote{Wu, M. \& Martin, M. P., JFM, vol.594, pp. 71-83, 2008} We perform a spectral analysis of the flowfield, investigate the behavior of the separation bubble and possible physical mechanisms driving the low-frequency unsteadiness. [Preview Abstract] |
Monday, November 23, 2009 10:56AM - 11:09AM |
HC.00003: DNS of transition in supersonic boundary layers Suman Muppidi, Krishnan Mahesh We are developing the numerical capability to simulate laminar-to-turbulent transition in high speed external flows in complex geometries. Transition to turbulence is associated with increased aerodynamic and thermal loads, and there is a need to reliably predict the behavior and better understand transition mechanisms. We use Direct Numerical Simulation to study transition induced by blowing and suction in a flat plate boundary layer flow at Mach 2.25. We use an unstructured compressible solver with a novel shock capturing scheme that is active only in regions of discontinuities. Preliminary results show good agreement with past work. We will use the simulation results to discuss the time-averaged behavior, transition mechanism, and the importance of simulation details (computational technique, domain size, mesh and timestep). [Preview Abstract] |
Monday, November 23, 2009 11:09AM - 11:22AM |
HC.00004: A Hybrid central difference/WENO scheme to simulate compressible turbulence with shocks and interfaces Eric Johnsen, Johan Larsson, Sanjiva Lele Numerical simulations of the late-time turbulent multi-material mixing in the Richtmyer-Meshkov instability (RMI) are challenging due to the contradictory requirements to treat turbulence, for which numerical dissipation must be avoided, and flow discontinuities, for which dissipation is introduced to stabilize the solution. In order to overcome this problem, a hybrid method was developed and is used to selectively apply a non-dissipative scheme in smooth regions, shock capturing at shocks, and interface capturing at interfaces. Results from 1D multicomponent Riemann problems and from 2D single-mode RMI will be presented and compared to other formulations to show that the present multifluid hybrid code does not generate spurious oscillations at discontinuities, that numerical dissipation is contained, and that the total mass, momentum and energy of the system are conserved. Detailed 3D simulations of the interaction of a shock in air with a curtain of SF6 will be presented, and the implementation of diffusive effects will be discussed. [Preview Abstract] |
Monday, November 23, 2009 11:22AM - 11:35AM |
HC.00005: Interaction of a spherical shock wave with compressible isotropic turbulence Ankit Bhagatwala, Sanjiva Lele There have been several studies on the interaction of a planar shock wave with a turbulent inflow. However, such canonical problems do not address the more practical cases of interest, wherein the shock is spherical. We study the interaction of a spherical shock wave expanding outwards into a field of compressible isotropic turbulence. The shock is initiated by depositing a large amount of energy at the center of the domain. As it progresses through the domain, it modifies and is modified by the turbulent flow field. The turbulent field has a high enough turbulent Mach number and Reynolds number to have significant dilatational, vortical and entropic fluctuations. We present two cases, one in which turbulence is significantly modified by the shock and another in which the shock is significantly modified by the turbulence. We identify the physical mechanisms behind these observations and the parameters that are important to this problem. Finally, we highlight some of the crucial differences and challenges of simulating spherical shock-turbulence interaction compared to the planar case. [Preview Abstract] |
Monday, November 23, 2009 11:35AM - 11:48AM |
HC.00006: DNS and LES of Shock / Isotropic Turbulence Interaction Nathan Grube, Ellen Taylor, Pino Martin We use direct numerical simulation (DNS) and large-eddy simulation (LES) to investigate the interaction of highly-compressible isotropic turbulence with nominally planar shock waves. The upstream isotropic turbulence is characterized by turbulence Mach numbers ranging from 0.14 to 0.94 and Taylor microscale Reynolds numbers ranging from 16 to 78. The convection speed of the turbulence through the shock ranges from Mach 1.5 to Mach 5. Streamwise profiles of mean and fluctuating thermodynamic and turbulence quantities are computed along with budgets for Reynolds stresses and fluctuating vorticity. Approximate three-dimensional turbulence energy spectra are computed using Taylor's Hypothesis. Visualization is aided by numerical schlieren animations. The DNS and LES are run using a WENO shock-capturing method in a finite difference code. Subgrid-scale terms are modeled using a dynamic mixed model and an approximate deconvolution model. The LES results are compared with DNS data. [Preview Abstract] |
Monday, November 23, 2009 11:48AM - 12:01PM |
HC.00007: Numerical simulation of aero-optical distortions by turbulent boundary layers Kan Wang, Meng Wang Compressible large-eddy simulations are carried out to study the aero-optical distortions caused by flat-plate turbulent boundary layers at $Re_{\theta} = 1400$ and 2800 and $M=0.5$. The fluctuating index-of-refraction field is calculated from the density field, and ray tracing is employed to compute the optical path differences (OPD). It is found that optical wavefront distortions are predominantly caused by the logarithmic layer and wake region. Consistent with previous experimental findings, the distortion magnitude is dependent on the direction of propagation due to anisotropy of the boundary-layer vortical structures. An optical beam is distorted more severely when it is tilted toward downstream than upstream. This is explained by a correlation analysis of the fluctuating density field, which shows that the correlation length is larger along downstream-tilted optical paths than upstream-tilted ones. The predicted OPD magnitude and structure at both Reynolds numbers are compared to clarify the Reynolds number dependence and effect of small flow scales. [Preview Abstract] |
Monday, November 23, 2009 12:01PM - 12:14PM |
HC.00008: LES of an inclined jet into a supersonic turbulent crossflow: synthetic inflow conditions Antonino Ferrante, Georgios Matheou, Paul Dimotakis The transition and spatial development of a helium sonic jet into a supersonic crossflow ($M$=3.6) were found to be strongly dependent on crossflow inflow conditions in the Large-Eddy Simulation (LES) of Ferrante {\em et al.} (AIAA-ASM, 2009-1511). These results indicate that correct turbulent inflow conditions are necessary to predict the main flow characteristics, dispersion and mixing, of a gaseous jet in a supersonic turbulent crossflow. The objective of this work is to provide a methodology for the generation of realistic synthetic inflow conditions for LES of spatially developing, supersonic, turbulent wall-bounded flows. The methodology is applied to the supersonic turbulent flow over a flat wall interacting with an inclined jet matching the experimental conditions of Maddalena {\em et al.} (JPP, 2006). The sub-grid scale stretched vortex model of turbulent momentum and scalar transport developed by Pullin and co-workers is employed. Inflow turbulence fluctuations are generated by modifying the methodology of Ferrante \& Elghobashi (JCP, 2004) to high-Reynolds number, supersonic flows. The results show that the main flow features generated by the gas-dynamic interactions of an inclined jet with the turbulent supersonic crossflow, such as unsteady bow shock, barrel shock, shear layer, and counter-rotating vortex pair, were captured. [Preview Abstract] |
Monday, November 23, 2009 12:14PM - 12:27PM |
HC.00009: Direct simulation of heat transfer in a turbulent swept flow over a wire in a channel Reetesh Ranjan, Carlos Pantano, Paul Fischer, Andrew Siegel We present results from direct numerical simulations of heat transfer (considered as a passive scalar) in a turbulent swept flow across a thin, cylindrical wire in a channel. This model mimics the flow through the wire-wrapped fuel pins typical of fast neutron reactor designs. Mean flow develops both along the wire and across the wire, leading to the formation of a turbulent cross-flow regime in the channel. This leads to improvement in heat transfer properties of the channel surface due to enhancement in mixing. The friction Reynolds number in the axial direction is approximately 305. Cross-flow friction Reynolds numbers ranging from 0 to 115 are examined. Two passive scalars at Prandtl number of 1.0 and 0.01 respectively, are simulated in this study. Constant flux boundary conditions are used along the walls of the channel and adiabatic conditions are used along the surface of the wire. The numerical method uses spectral elements in the plane perpendicular to the wire axis and Fourier decomposition in the direction of the axis of the wire. The simulations use up to 107 million collocation points and were performed at the Argonne Leadership BG/P supercomputer. The passive scalar field statistics are investigated, including mean scalar field, turbulence statistics and instantaneous surface scalar distribution. [Preview Abstract] |
Monday, November 23, 2009 12:27PM - 12:40PM |
HC.00010: LES of Supersonic Turbulent Channel Flow at Mach Numbers 1.5 and 3 Sriram Raghunath, Giles Brereton LES of compressible, turbulent, body-force driven, isothermal-wall channel flows at $Re_{\tau}$ of 190 and 395 at moderate supersonic speeds (Mach 1.5 and 3) are presented. Simulations are fully resolved in the wall-normal direction without the need for wall-layer models. SGS models for incompressible flows, with appropriate extensions for compressibility, are tested {\it a priori\/} with DNS results and used in LES. Convergence of the simulations is found to be sensitive to the initial conditions and to the choice of model (wall-normal damping) in the laminar sublayer. The Nicoud--Ducros wall adapting SGS model, coupled with a standard SGS heat flux model, is found to yield results in good agreement with DNS. [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