Bulletin of the American Physical Society
70th Annual Meeting of the APS Division of Fluid Dynamics
Volume 62, Number 14
Sunday–Tuesday, November 19–21, 2017; Denver, Colorado
Session F2: Reacting Flows: LESCFD Reacting
|
Hide Abstracts |
Chair: Yuan Xuan, Pennsylvania State University Room: 402 |
Monday, November 20, 2017 8:00AM - 8:13AM |
F2.00001: Relationships between Physical and Fourier Space for Large-Eddy Simulation of Premixed Turbulent Combustion: Multi-Dimensional Fourier Decomposition. Paulo Paes, James Brasseur, Yuan Xuan Large Eddy Simulation (LES) is a powerful formulation to model turbulent reacting flows with tradeoffs between complexity and resolution. LES assumes that all energy-dominated turbulence motions are well-resolved and forward cascade-dominant so that modeled effects of Sub-Filter-Scale (SFS) motion are second order. However, the application of this scale-based decomposition to reacting turbulent flows is not straightforward since dynamically important kinetics within thin flame regions are mostly SFS. We aim to systematically refine understanding of the relationships between physical and scale space for LES of premixed turbulent combustion, beginning with reduced-physics simulations of the interactions between single-scale vortex arrays and laminar premixed flames. We apply Fourier scale-based decomposition where interpretation in inhomogeneous directions is unclear and where periodic extension of the finite domain in those directions produce boundary discontinuities. We present a strategy to remove the pollution to the signal from these discontinuities with minimal modification of the original signal by systemically isolating their spectral content. The procedure is applied to a 2D vortex-flame interaction where we demonstrate 2D Fourier-physical space relationships in homogeneous and inhomogeneous directions. \textit{Supported by AFSOR.} [Preview Abstract] |
Monday, November 20, 2017 8:13AM - 8:26AM |
F2.00002: Relationships between Physical and Fourier Space for Large-Eddy Simulation of Premixed Turbulence Combustion: Transition from Weak to Strong Vortices. Yuan Xuan, Paulo Paes, James Brasseur Large Eddy Simulation (LES) is a powerful formulation for turbulent reacting flows with tradeoff between complexity and resolution. LES assumes that all energy-dominant motions are well-resolved and forward cascade-dominant. But the application of this scale-based decomposition to reacting turbulent flows is not straightforward since dynamically important thin flame regions are mostly SFS. We aim to systematically refine understanding of the relationships between physical and scale space for LES of premixed turbulent combustion, beginning with reduced-physics simulations of the interaction between laminar premixed flame and single-scale vortex arrays with ``eddy'' strength systematically increased to create flame response from ``wrinkled'' to ``pinched''. We apply the Fourier description to these simulations using a newly developed procedure to remove the polluting content from discontinuities in inhomogeneous directions. Using wavelet-like filters we identify four characteristic structural features concurrently in physical/Fourier space, with relative significance estimated from the variance contained in the spectrally filtered fields. Different variables display distinct dominant structural features that undergo systematic transition from weak to strong eddy-flame interaction. [Preview Abstract] |
Monday, November 20, 2017 8:26AM - 8:39AM |
F2.00003: Large eddy simulations of a reactive jet in hot vitiated crossflow: flame stabilisation mechanism Oliver Schulz, Emile Piccoli, Anne Felden, Gabriel Staffelbach, Nicolas Noiray This numerical study investigates the flame stabilization mechanism of a reactive jet in crossflow (RJICF). A premixed ethylene-air jet is injected into a hot vitiated crossflow at 1500K. Compressible 3-d large eddy simulations (LES) were performed with the dynamic thickened flame (DTF) model and an analytically reduced chemistry (ARC) mechanism. Comparisons between LES and experiments are in good agreement for jet trajectories and velocities along the trajectories. LES capture the most common flame-flow field interactions observed in the experiments. A detailed flame analysis utilizing chemical explosive mode analysis (CEMA) identifies autoignition as the dominant flame stabilisation mechanism on the windward side of the jet. Heat is released at an optimum mixture for fast ignition. This heat is transferred to richer mixtures, which are characterized by even smaller minimum autoignition times. This so-called most reactive mixture fraction determines the position of the windward flame. A 3-d flame analysis shows autoignition regions that form close to structures resulting from the jet shear layer vortex shedding. These regions expand towards the flame tip and the side of the RJICF. Finally, they merge with the leeward propagating flame. [Preview Abstract] |
Monday, November 20, 2017 8:39AM - 8:52AM |
F2.00004: Numerical simulation of turbulent burning velocity of CH$_{\mathrm{4}}$/H$_{\mathrm{2}}$/air flame using LES/FGM approach Masaya Muto, Hiroaki Nagai, Ryoichi Kurose, Fumiteru Akamatsu, Kei Inoue, Kenji Miyamoto The turbulent burning velocity, $s_{T}$, of hydrogen/methane/air mixture is numerically investigated by large-eddy simulation (LES) with flamelet generated manifold (FGM) method of turbulent jet flow. Volume ratio of the hydrogen/methane in the mixture, $\alpha$ is varied from 0 to 0.6 for the ambient pressure range of $P=$0.1-0.9 MPa. Equivalence ratio of the mixture is fixed to be unity. The results show that the ratio of $s_{T}$ to laminar burning velocity $s_{L}$, $s_{T}$/$s_{L}$ increases with increasing $\alpha$ and $P$ This is considered to be due to the facts that $s_{L}$ decreases with increasing $P$, and that the flame thickness decreases with increasing $\alpha$ and $P$, which causes to increase the surface area of flame sheet. [Preview Abstract] |
Monday, November 20, 2017 8:52AM - 9:05AM |
F2.00005: ABSTRACT WITHDRAWN |
Monday, November 20, 2017 9:05AM - 9:18AM |
F2.00006: Role of Unsteady Effects in Radiation Heat Losses in Turbulent Nonpremixed Flames A. Cody Nunno, Jeffry K. Lew, Michael E. Mueller Accounting for heat losses due to radiation in turbulent nonpremixed flames is critical for predicting pollutants such as nitrogen oxides and soot. In reduced-order manifold approaches (“flamelet” models), the effects of heat losses require the computation of thermochemical states with reduced enthalpy. In this work, the role of unsteady effects in radiation heat losses is assessed by examining two methods for computing thermochemical states at reduced enthalpy. In the first method, unsteady “flamelet” equations including a radiation heat loss source term are solved at constant scalar dissipation rate, initialized with adiabatic solutions of the steady “flamelet” equations. In the second method, only steady “flamelet” equations with a radiation heat loss source term are solved but with a variable coefficient on the source term range from zero to unity. Both approaches are applied to the Sandia D flame. A priori analysis of the two methods indicates that the two methods are equivalent at larger scalar dissipation rates and/or smaller heat losses. A posteriori comparisons of Large Eddy Simulation (LES) with experimental measurements will be used to determine the relative effects of the two approaches on predictions of temperature, major species, and pollutants. [Preview Abstract] |
Monday, November 20, 2017 9:18AM - 9:31AM |
F2.00007: Modeling Soot Formation in High-Pressure Turbulent Spray Flames Daniel Haworth, Sebastian Ferreyro-Fernandez Most soot models are based on physical understanding derived from experiments at atmospheric or moderately elevated pressures, compared to the pressures that are of interest in engines and other applications. The emphasis in model development has been on kinetic processes rather than on turbulent hydrodynamics and mixing, but there is evidence that transport and mixing become relatively more rate-controlling with increasing pressure. Here simulations are performed of transient high-pressure turbulent spray flames under engine-relevant conditions. An unsteady RANS formulation is adopted, with various gas-phase chemical mechanisms and soot models, and a transported composition probability density function method to account for unresolved turbulent fluctuations in composition and temperature. Computed total soot mass and soot spatial distributions are highly sensitive to the modeling of unresolved turbulent fluctuations. To achieve agreement between model and experiment and to capture the highly intermittent nature of soot in the turbulent flame, it is necessary to accurately represent mixing and the low diffusivity of soot particles. The results suggest that mixing is at least as important as kinetics in controlling soot formation and evolution in high-pressure turbulent flames. [Preview Abstract] |
Monday, November 20, 2017 9:31AM - 9:44AM |
F2.00008: Evaluation of a strain-sensitive transport model in LES of turbulent nonpremixed sooting flames Jeffry K. Lew, Suo Yang, Michael E. Mueller Direct Numerical Simulations (DNS) of turbulent nonpremixed jet flames have revealed that Polycyclic Aromatic Hydrocarbons (PAH) are confined to spatially intermittent regions of low scalar dissipation rate due to their slow formation chemistry. The length scales of these regions are on the order of the Kolmogorov scale or smaller, where molecular diffusion effects dominate over turbulent transport effects irrespective of the large-scale turbulent Reynolds number. A strain-sensitive transport model has been developed to identify such species whose slow chemistry, relative to local mixing rates, confines them to these small length scales. In a conventional nonpremixed ``flamelet'' approach, these species are then modeled with their molecular Lewis numbers, while remaining species are modeled with an effective unity Lewis number. \textit{A priori} analysis indicates that this strain-sensitive transport model significantly affects PAH yield in nonpremixed flames with essentially no impact on temperature and major species. The model is applied with Large Eddy Simulation (LES) to a series of turbulent nonpremixed sooting jet flames and validated via comparisons with experimental measurements of soot volume fraction. [Preview Abstract] |
Monday, November 20, 2017 9:44AM - 9:57AM |
F2.00009: Method of moments comparison for soot population modeling in turbulent combustion Shao Teng Chong, Hong Im, Venkat Raman Representation of soot population is an important component in the efficient computational prediction of particulate emissions. However, there are a number of moments-based techniques with varying numerical complexity. In the past, development of such methods has been principally carried out on canonical laminar and 0-D flows. However, their applications in realistic solvers developed for turbulent combustion may face challenges from turbulence closure to selection of moment sets. In this work, the accuracy and relative computational expense of a few common soot method of moments are tested in canonical turbulent flames for different configurations. Large eddy simulation (LES) will be used as the turbulence modeling framework. In grid-filtered LES, the interaction of numerical and modeling errors is a first-order problem that can undermine the accuracy of soot predictions. In the past, special moments-based methods for solvers that transport high frequency content fluid with ability to reconstruct particle size distribution have been developed. Here, a similar analysis will be carried out for the moment-based soot modeling approaches above. Specifically, realizability of moments methods with nonlinear advection schemes will be discussed. [Preview Abstract] |
Monday, November 20, 2017 9:57AM - 10:10AM |
F2.00010: A revised NOMANI model for NO prediction using improved mechanism indicator Daehyun Han, Seongwon Kang The main objective of this study is to predict NO emission efficiently and accurately using a revised NOMANI model in laminar and turbulent flames. Although the prompt and thermal NO mechanisms have very different reaction time scales, many practical combustion models assume fast chemistry and have a limitation in predicting formation of thermal NO. The NOMANI model by Pecquery et al. (2014) is based on the flamelet-progress variable approach and suggests a new tabulation with NO composition as an axis instead of the progress variable. In the present study, various mechanism indicators other than the original progress variable are analyzed to represent local mechanism of NO production more accurately. In order to develop improved weight parameters for non-premixed flames and post-flame regions, we consider the species involved in different NO formation mechanisms. We analyze the proposed approach using DNS of laminar premixed and non-premixed flames. As validation cases, LES studies are performed for turbulent flames, which leads to improved results. [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