Bulletin of the American Physical Society
67th Annual Meeting of the APS Division of Fluid Dynamics
Volume 59, Number 20
Sunday–Tuesday, November 23–25, 2014; San Francisco, California
Session H34: LES and Modeling of Turbulent Combustion |
Hide Abstracts |
Chair: Joe Ofelein, Sandia National Laboratories Room: 2024 |
Monday, November 24, 2014 10:30AM - 10:43AM |
H34.00001: Analysis of operator splitting errors for DNS of low Mach number turbulent reacting flows Jonathan MacArt, Michael E. Mueller A formally second-order accurate Strang splitting approach has been developed and applied to the solution of scalar transport/reaction equations for Direct Numerical Simulation (DNS) of low Mach number turbulent reacting flows. The temporal discretization errors of the scheme are analyzed in both the asymptotic and non-asymptotic regimes of convergence and compared with a formally first-order accurate Lie splitting approach in a series of one-dimensional test problems with real combustion chemistry. The Strang splitting scheme is demonstrated to achieve its theoretical accuracy when all relevant chemical time scales are resolved; however, with larger time steps representative of those utilized in practice for low Mach number DNS of turbulent reacting flows, a reduction in order is observed. Nonetheless, the Strang splitting approach exhibits a higher order of accuracy and smaller errors than Lie splitting for all time steps. Preliminary DNS results for a turbulent planar jet computed with this scheme will also be discussed. [Preview Abstract] |
Monday, November 24, 2014 10:43AM - 10:56AM |
H34.00002: Optimal Numerical Schemes for Compressible Large Eddy Simulations Ayaboe Edoh, Ann Karagozian, Venkateswaran Sankaran, Charles Merkle The design of optimal numerical schemes for subgrid scale (SGS) models in LES of reactive flows remains an area of continuing challenge. It has been shown that significant differences in solution can arise due to the choice of the SGS model's numerical scheme and its inherent dissipation properties, which can be exacerbated in combustion computations.\footnote{Cocks, Sankaran and Soterioiu, AIAA 2013-0170} This presentation considers the individual roles of artificial dissipation, filtering,\footnote{Kennedy and Carpenter, \textbf{App. Num. Math.},14, 397-433, 1994} secondary conservation\footnote{Subbareddy and Candler, \textbf{J. Comp. Phys.}, 228,1347-1364, 2009} (Kinetic Energy Preservation), and collocated versus staggered grid arrangements with respect to the dissipation and dispersion characteristics and their overall impact on the robustness and accuracy for time-dependent simulations of relevance to reacting and non-reacting LES. We utilize von Neumann stability analysis in order to quantify these effects and to determine the relative strengths and weaknesses of the different approaches. [Preview Abstract] |
Monday, November 24, 2014 10:56AM - 11:09AM |
H34.00003: A Trust-Region Constrained Fidelity Adaptive Combustion Model Yee Chee See, Hao Wu, Qing Wang, Matthias Ihme A general framework is developed to dynamically adapt the local fidelity of the combustion model for reacting flows. This framework combines a hierarchy of combustion models with different fidelity, and the adaptation is achieved by dynamically assigning a combustion model under consideration of their accuracy and computational cost. The usage of each model is confined to the trust region whose size is specified by the user. The applicability of a certain combustion model is determined by the compatibility between its manifold and the local flow field. By doing so, it becomes possible to conduct a reacting flow simulation, in which fidelity and cost are subject to user-specific requirements, and prior knowledge about the combustion regime is not necessary. This fidelity-adaptive model is applied to a triple flame to demonstrate its capability and the model performance is assessed through direct comparisons against a detailed numerical simulation. [Preview Abstract] |
Monday, November 24, 2014 11:09AM - 11:22AM |
H34.00004: Consideration of Turbulence Effects in One-Dimensional Laminar Flamelet Equations Wai Lee Chan, Matthias Ihme The laminar flamelet formulation has been used as a fundamental building block for the construction of turbulent combustion closures. By assuming that turbulence only leads to a deformation and straining of the local flame structure, the turbulence/chemistry interaction is then considered through a presumed shape probability density function (PDF) approach. However, the consistency of this approach remains unclear in the context of large-eddy simulations (LES), and the objective of this study is to examine the representation of turbulent scalar fluxes and turbulence/chemistry coupling on the flame structure. To this end, a detailed numerical simulation of a turbulent counterflow diffusion flame is performed, and the simulation results are used to analyze the limitations of the classic laminar flamelet formulation and explore a possible alternative approach. [Preview Abstract] |
Monday, November 24, 2014 11:22AM - 11:35AM |
H34.00005: Chemical Source Term Closure in Turbulent Combustion using Approximate Deconvolution Methods Qing Wang A closure model for the chemical source term in Large Eddy Simulation (LES) using the Approximate Deconvolution Method (ADM) is proposed. The model recovers the scalar field that is discarded by the LES filter from the information retained in the large-scale structures using an approximate deconvolution operator. The nonlinear chemical source term is then evaluated based on the de-convoluted scalar field. Since this formulation makes no presumptions on the combustion regime, it is applicable to complex combustion configurations and detailed chemistry. The capability of this sub-grid closure model is examined in an a priori study, and the performance, accuracy, and computational cost are characterized through a posteriori simulations. [Preview Abstract] |
Monday, November 24, 2014 11:35AM - 11:48AM |
H34.00006: \textit{A priori} DNS evaluation of the shadow-position mixing model in turbulent reactive flows Xinyu Zhao, Ankit Bhagatwala, Jacqueline Chen, Daniel Haworth, Stephen Pope The modeling of molecular diffusion of chemical species is an important aspect of modeling turbulent reactive flows, especially for transported probability density function based methods. In this work, shadow-position mixing model (SPMM) is examined, using the DNS database of a temporally-evolving di-methyl ether jet flame undergoing local extinction and re-ignition. SPMM is similar to the conventional interaction by exchange with the mean (IEM) model, with the exception that there is an additional conditioning variable, the so-called ``shadow displacement.'' Turbulent statistics and the shadow displacement are first extracted from the DNS database. Based on the position, time and shadow displacement, the conditional species diffusion from DNS and from SPMM are calculated and compared for several major and minor species. Possible values of model constants are then derived from the comparison of the conditional diffusion. Finally, the relation of SPMM with the IEM model, and the relation of SPMM with interaction by exchange with the conditional mean model, are explored and discussed. [Preview Abstract] |
Monday, November 24, 2014 11:48AM - 12:01PM |
H34.00007: Large Eddy Simulations of Colorless Distributed Combustion Systems Husam F. Abdulrahman, Farhad Jaberi, Ashwani Gupta Development of efficient and low-emission colorless distributed combustion (CDC) systems for gas turbine applications require careful examination of the role of various flow and combustion parameters. Numerical simulations of CDC in a laboratory-scale combustor have been conducted to carefully examine the effects of these parameters on the CDC. The computational model is based on a hybrid modeling approach combining large eddy simulation (LES) with the filtered mass density function (FMDF) equations, solved with high order numerical methods and complex chemical kinetics. The simulated combustor operates based on the principle of high temperature air combustion (HiTAC) and has shown to significantly reduce the NOx, and CO emissions while improving the reaction pattern factor and stability without using any flame stabilizer and with low pressure drop and noise. The focus of the current work is to investigate the mixing of air and hydrocarbon fuels and the non-premixed and premixed reactions within the combustor by the LES/FMDF with the reduced chemical kinetic mechanisms for the same flow conditions and configurations investigated experimentally. The main goal is to develop better CDC with higher mixing and efficiency, ultra-low emission levels and optimum residence time. The computational results establish the consistency and the reliability of LES/FMDF and its Lagrangian-Eulerian numerical methodology. [Preview Abstract] |
Monday, November 24, 2014 12:01PM - 12:14PM |
H34.00008: DNS and LES/FMDF of turbulent jet ignition and combustion AbdoulAhad Validi, Farhad Jaberi The ignition and combustion of lean fuel-air mixtures by a turbulent jet flow of hot combustion products injected into various geometries are studied by high fidelity numerical models. Turbulent jet ignition (TJI) is an efficient method for starting and controlling the combustion in complex propulsion systems and engines. The TJI and combustion of hydrogen and propane in various flow configurations are simulated with the direct numerical simulation (DNS) and the hybrid large eddy simulation/filtered mass density function (LES/FMDF) models. In the LES/FMDF model, the filtered form of the compressible Navier-Stokes equations are solved with a high-order finite difference scheme for the turbulent velocity and the FMDF transport equation is solved with a Lagrangian stochastic method to obtain the scalar field. The DNS and LES/FMDF data are used to study the physics of TJI and combustion for different turbulent jet igniter and gas mixture conditions. The results show the very complex and different behavior of the turbulence and the flame structure at different jet equivalence ratios. [Preview Abstract] |
Monday, November 24, 2014 12:14PM - 12:27PM |
H34.00009: Study of Entropy Generation in Turbulent Jet Flames Using Large Eddy Simulation Mehdi Safari, Reza H. Sheikhi Analysis of local entropy generation is an effective means to investigate sources of irreversibility in turbulent combustion. Large eddy simulation (LES) is employed to describe transport of entropy in turbulent reacting flows. The filtered form of this equation includes entropy production due to viscous dissipation, heat conduction, mass diffusion and chemical reaction, all of which appear as unclosed terms. The SGS effects are taken into account using a methodology based on the filtered density function (FDF). This methodology, entitled entropy FDF (En-FDF), is developed and utilized in the form of scalar-entropy FDF transport equation. This equation is modeled by a set of stochastic differential equations. The modeled En-FDF transport equation is solved by a Lagrangian Monte Carlo method. The methodology is employed for LES of a turbulent nonpremixed jet flame at several flow parameters. The main advantage of the En-FDF is that it provides closure for all individual entropy generation effects. It also includes the effect of chemical reaction in a closed form. Predictions show good agreements with the experimental data. Entropy generation effects are predicted by the En-FDF and analyzed. The sensitivity of entropy generation to flow parameters are investigated. [Preview Abstract] |
Monday, November 24, 2014 12:27PM - 12:40PM |
H34.00010: CFD Modeling of a Laser-Induced Ethane Pyrolysis in a Wall-less Reactor Olga Stadnichenko, Valeriy Snytnikov, Junfeng Yang, Omar Matar Ethylene, as the most important feedstock, is widely used in chemical industry to produce various rubbers, plastics and synthetics. A recent study found the IR-laser irradiation induced ethane pyrolysis yields 25\% higher ethylene production rates compared to the conventional steam cracking method. Laser induced pyrolysis is initiated by the generation of radicals upon heating of the ethane, then, followed by ethane/ethylene autocatalytic reaction in which ethane is converted into ethylene and other light hydrocarbons. This procedure is governed by micro-mixing of reactants and the feedstock residence time in reactor. Under mild turbulent conditions, the turbulence enhances the micro-mixing process and allows a high yield of ethylene. On the other hand, the high flow rate only allows a short residence time in the reactor which causes incomplete pyrolysis. This work attempts to investigate the interaction between turbulence and ethane pyrolysis process using large eddy simulation method. The modelling results could be applied to optimize the reactor design and operating conditions. [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