Bulletin of the American Physical Society
69th Annual Meeting of the APS Division of Fluid Dynamics
Volume 61, Number 20
Sunday–Tuesday, November 20–22, 2016; Portland, Oregon
Session L34: Turbulence: Compressible Flow |
Hide Abstracts |
Chair: Diego Donzis, Texas A&M University Room: Oregon Ballroom 203 |
Monday, November 21, 2016 4:30PM - 4:43PM |
L34.00001: Compressibility effects on turbulent mixing John Panickacheril John, Diego Donzis We investigate the effect of compressibility on passive scalar mixing in isotropic turbulence with a focus on the fundamental mechanisms that are responsible for such effects using a large Direct Numerical Simulation (DNS) database. The database includes simulations with Taylor Reynolds number ($R_{\lambda}$) up to 100, turbulent Mach number ($M_{t}$) between 0.1 and 0.6 and Schmidt number ($Sc$) from 0.5 to 1.0. We present several measures of mixing efficiency on different canonical flows to robustly identify compressibility effects. We found that, like shear layers, mixing is reduced as Mach number increases. However, data also reveal a non-monotonic trend with $M_{t}$. To assess directly the effect of dilatational motions we also present results with both dilatational and soleniodal forcing. Analysis suggests that a small fraction of dilatational forcing decreases mixing time at higher $M_{t}$. Scalar spectra collapse when normalized by Batchelor variables which suggests that a compressive mechanism similar to Batchelor mixing in incompressible flows might be responsible for better mixing at high $M_{t}$ and with dilatational forcing compared to pure solenoidal mixing. We also present results on scalar budgets, in particular on production and dissipation. [Preview Abstract] |
Monday, November 21, 2016 4:43PM - 4:56PM |
L34.00002: DNS study on shock/turbulence interaction in homogeneous isotropic turbulence at low turbulent Mach number Kento Tanaka, Tomoaki Watanabe, Koji Nagata, Akihiro Sasoh, Yasuhiko Sakai, Toshiyuki Hayase The interaction between homogeneous isotropic turbulence and normal shock wave is investigated by direct numerical simulations (DNSs). In the DNSs, a normal shock wave with a shock Mach number 1.1 passes through homogeneous isotropic turbulence with a low turbulent Mach number and a moderate turbulent Reynolds number. The statistics are calculated conditioned on the distance from the shock wave. The results showed that the shock wave makes length scales related to turbulence small. This effect is significant for the Taylor microscale defined with the velocity derivative orthogonal to the shock wave. The decrease in the Kolmogorov scale is also found. Statistics of velocity derivative are found to be changed by the shock wave propagation. The shock wave causes enstrophy amplification due to the dilatation/vorticity interaction. By this interaction, the vorticity components parallel to the shock wave is more amplified than the normal component. The strain rate is also amplified by the shock wave. [Preview Abstract] |
Monday, November 21, 2016 4:56PM - 5:09PM |
L34.00003: Effects of Compressibility on Turbulent Relative Particle Dispersion Bhimsen Shivamoggi In this paper, phenomenological developments are used to explore the effects of compressibility on the relative particle dispersion (RPD) in 3D fully-developed turbulence (FDT). The role played by the compressible FDT cascade physics underlying this process is investigated. Compressibility effects are found to lead to reduction of RPD, development of the ballistic regime and particle clustering, corroborating the laboratory experiment and numerical simulation results (Cressman et al., 2004) on the motion of Lagrangian tracers on a surface flow that constitutes a 2D compressible subsystem. These formulations are developed from the scaling relations for compressible FDT and are validated further via an alternative dimensional/scaling development for compressible FDT similar to the one given for incompressible FDT by Batchelor and Townsend (1956). The rationale for spatial intermittency effects is legitimized via the nonlinear scaling dependence of RPD on the kinetic energy dissipation rate. [Preview Abstract] |
Monday, November 21, 2016 5:09PM - 5:22PM |
L34.00004: Universality and scaling in compressible turbulence Diego Donzis, Shriram Jagannathan A large database of Direct Numerical Simulations (DNS) of stationary compressible isotropic turbulence at a range of Taylor Reynolds numbers ($R_\lambda \approx 38-450$) and turbulent Mach numbers ($M_t \approx 0.1-0.6$) is used to explore universality. While in incompressible turbulence self-similarity analysis leads to a single scaling parameter ($R_\lambda$), compressible turbulence expands the parameter space due to the coupling between hydrodynamics and thermodynamics, and the dependence on the mode of external forcing. While for the former it is common to use $M_t$ as a scaling parameter, the effects of the latter are harder to quantify, and their consequences may have been attributed to a certain lack of universality. For instance, when the dilatational mode is forced, the variance and skewness of pressure shows significant scatter when plotted against $M_t$. Using a Helmholtz decomposition, we split the velocity field into solenoidal and dilatational modes, and propose scaling parameters that include the contribution from both modes. When expressed against these parameters, we observe a universal scaling regime regardless of the mode of excitation of forcing. Other quantities that follow this behavior are also discussed. [Preview Abstract] |
Monday, November 21, 2016 5:22PM - 5:35PM |
L34.00005: The role of bulk viscosity on the decay of compressible, homogeneous, isotropic turbulence Eric Johnsen, Shaowu Pan The practice of neglecting bulk viscosity in studies of compressible turbulence is widespread. While exact for monatomic gases and unlikely to strongly affect the dynamics of fluids whose bulk-to-shear viscosity ratio is small and/or of weakly compressible turbulence, this assumption is not justifiable for compressible, turbulent flows of gases whose bulk viscosity is orders of magnitude larger than their shear viscosities (e.g., CO2). To understand the mechanisms by which bulk viscosity and the associated phenomena affect compressible turbulence, we conduct DNS of freely decaying compressible, homogeneous, isotropic turbulence for ratios of bulk-to-shear viscosity ranging from 0-1000. Our simulations demonstrate that bulk viscosity increases the decay rate of turbulent kinetic energy; while enstrophy exhibits little sensitivity to bulk viscosity, dilatation is reduced by an order of magnitude within the two eddy turnover time. Via a Helmholtz decomposition of the flow, we determined that bulk viscosity damps the dilatational velocity and reduces dilatational-solenoidal exchanges, as well as pressure-dilatation coupling. In short, bulk viscosity renders compressible turbulence incompressible by reducing energy transfer between translational and internal modes. [Preview Abstract] |
Monday, November 21, 2016 5:35PM - 5:48PM |
L34.00006: ABSTRACT WITHDRAWN |
Monday, November 21, 2016 5:48PM - 6:01PM |
L34.00007: Detailed thermodynamic analyses of high-speed compressible turbulence Colin Towery, Ryan Darragh, Alexei Poludnenko, Peter Hamlington Interactions between high-speed turbulence and flames (or chemical reactions) are important in the dynamics and description of many different combustion phenomena, including autoignition and deflagration-to-detonation transition. The probability of these phenomena to occur depends on the magnitude and spectral content of turbulence fluctuations, which can impact a wide range of science and engineering problems, from the hypersonic scramjet engine to the onset of Type Ia supernovae. In this talk, we present results from new direct numerical simulations (DNS) of homogeneous isotropic turbulence with turbulence Mach numbers ranging from $0.05$ to $1.0$ and Taylor-scale Reynolds numbers as high as $700$. A set of detailed analyses are described in both Eulerian and Lagrangian reference frames in order to assess coherent (structural) and incoherent (stochastic) thermodynamic flow features. These analyses provide direct insights into the thermodynamics of strongly compressible turbulence. Furthermore, presented results provide a non-reacting baseline for future studies of turbulence-chemistry interactions in DNS with complex chemistry mechanisms. [Preview Abstract] |
Monday, November 21, 2016 6:01PM - 6:14PM |
L34.00008: Turbulent jumps and shock-structure in shock-turbulence interactions using shock-resolving direct numerical simulations Chang-Hsin Chen, Diego Donzis Substantial efforts have been made to understand the canonical interaction between isotropic turbulence and a normal shock. Evidence from theories, experiments and simulations, however, has shown that the interaction is complex and that the outcome is determined not only by mean flow behavior, as suggested by early theories, but also by characteristics of turbulence fluctuations typically quantified by parameters such as the Reynolds ($R_\lambda$) and the turbulent Mach number ($M_t$). An important, yet unresolved, issue is the accurate determination of departures from Rankine-Hugoniot relations due to turbulent fluctuations upstream of the shock. We present an analytic study, based on the quasi-equilibrium assumption, that yield turbulent jumps that depend not only on the mean flow but also on turbulence characteristics. In particular, the focus will be on thermodynamic jumps. Our analytical results agree well with new shock-resolving simulations at a range of Reynolds and Mach numbers. In the context of these results we also present a comparison of previous theory on the dilatation at the shock with the new DNS data. This is further discussed in the context of the transition from wrinkled to broken regimes and the difficulties associated with identifying a shock for very vigorous turbulence. [Preview Abstract] |
Monday, November 21, 2016 6:14PM - 6:27PM |
L34.00009: Spectral Behavior of Weakly Compressible Aero-Optical Distortions Edwin Mathews, Kan Wang, Meng Wang, Eric Jumper In classical theories of optical distortions by atmospheric turbulence, an appropriate and key assumption is that index-of-refraction variations are dominated by fluctuations in temperature and the effects of turbulent pressure fluctuations are negligible. This assumption is, however, not generally valid for aero-optical distortions caused by turbulent flow over an optical aperture, where both temperature and pressures fluctuations may contribute significantly to the index-of-refraction fluctuations. A general expression for weak fluctuations in refractive index is derived using the ideal gas law and Gladstone-Dale relation and applied to describe the spectral behavior of aero-optical distortions. Large-eddy simulations of weakly compressible, temporally evolving shear layers are then used to verify the theoretical results. Computational results support theoretical findings and confirm that if the log slope of the 1-D density spectrum in the inertial range is $-m_{\rho}$, the optical phase distortion spectral slope is given by $-(m_{\rho}+1)$. The value of $m_{\rho}$ is then shown to be dependent on the ratio of shear-layer free-stream densities and bounded by the spectral slopes of temperature and pressure fluctuations. [Preview Abstract] |
Monday, November 21, 2016 6:27PM - 6:40PM |
L34.00010: Predicting the mean fields of compressible turbulent boundary layer via a symmetry approach. Wei-Tao Bi, Bin Wu, Zhen-Su She A symmetry approach for canonical wall turbulence is extended to develop mean-field predictions for compressible turbulent boundary layer (CTBL). A stress length and a weighted heat flux length are identified to obey the multilayer dilation symmetry of canonical flows, giving rise to predictions of the mean velocity and temperature profiles for a range of Reynolds number (Re), Mach number (Ma) and wall temperature (Tw). Also predicted are the streamwise developments of the shape factor, the boundary layer edge velocity and the boundary layer thicknesses, etc. Only three parameters are involved in the predictions, which have sound physics and organized behaviors with respect to the Re, Ma and Tw effects. The predictions are extensively validated by direct numerical simulation and experimental data, showing better accuracies than the previous theories. The results provide new quantifications that can be used to assess computations, measurements and turbulence models of CTBL, as well as to provide new insights for the CTBL physics. [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. |
© 2025 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