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 R34: Extinction-Reignition, Autoignition and Flashback |
Hide Abstracts |
Chair: Jackie Chen, Sandia National Laboratories Room: 2024 |
Tuesday, November 25, 2014 1:05PM - 1:18PM |
R34.00001: Tomographic PIV study of boundary-layer flashback in swirl flames Dominik Ebi, Noel Clemens Preventing boundary layer flashback in swirl combustors is a key challenge for gas turbines intended to burn high hydrogen content fuels. We are studying this type of flashback by investigating the upstream flame propagation of lean-premixed methane/hydrogen-air flames inside the mixing tube of our model swirl combustor. Experiments are conducted at atmospheric pressure. Flashback is triggered by increasing the equivalence ratio. Previous studies employing planar measurements have shown that the flame strongly alters the upstream flow field and thus its own propagation path. Volumetric measurement techniques are needed to further increase understanding of this highly three-dimensional coupled flow-flame interaction. Flashback is an inherently transient event with duration on the order of a few hundred milliseconds. Time-resolved tomographic PIV together with high-speed chemiluminescence imaging is therefore applied to investigate the velocity field in the vicinity of the flame. [Preview Abstract] |
Tuesday, November 25, 2014 1:18PM - 1:31PM |
R34.00002: Flashback behavior in a model swirl combustor at elevated pressure Rakesh Ranjan, Dominik Ebi, Noel Clemens Understanding of combustion physics at high pressure is essential for safe and efficient operation of gas turbine combustors. A new optically-accessible elevated pressure combustion facility has been developed for this purpose. The modular design of the chamber allows applying various optical diagnostic techniques and the installation of different types of combustors. In the current study, the effect of pressure on boundary layer flashback in lean-premixed swirl flames is investigated. Mixtures of hydrogen and methane at different equivalence ratios are tested. High-speed chemiluminescence imaging is employed to study the upstream flame propagation inside the mixing tube, which allows comparison to previous results of flashback at atmospheric pressure. [Preview Abstract] |
Tuesday, November 25, 2014 1:31PM - 1:44PM |
R34.00003: Large Eddy Simulation of Flame Flashback in Swirling Premixed Flames Christopher Lietz, Venkatramanan Raman In the design of high-hydrogen content gas turbines for power generation, flashback of the turbulent flame by propagation through the low velocity boundary layers in the premixing region is an operationally dangerous event. Predictive models that could accurately capture the onset and subsequent behavior of flashback would be indispensable in gas turbine design. The large eddy simulation (LES) approach is used here to model this process. The goal is to examine the validity of a probability distribution function (PDF) based model in the context of a lean premixed flame in a confined geometry. A turbulent swirling flow geometry and corresponding experimental data is used for validation. A suite of LES calculations are performed on a large unstructured mesh for varying fuel compositions operating at several equivalence ratios. It is shown that the PDF based method can predict some statistical properties of the flame front, with improvement over other models in the same application. [Preview Abstract] |
Tuesday, November 25, 2014 1:44PM - 1:57PM |
R34.00004: Large eddy simulation of flame flashback in a turbulent channel Malik Hassanaly, Christopher Lietz, Venkat Raman, Hemanth Kolla, Jacqueline Chen, Andrea Gruber In high-hydrogen content gas turbines, the propagation of a premixed flame along with boundary layers on the combustor walls is a source of failure, whereby the flame could enter the fuel-air premixing region that is not designed to hold high-temperature fluid. In order to develop models for predicting this phenomenon, a large eddy simulation (LES) based study is carried out here. The flow configuration is based on a direct numerical simulation (DNS) of a turbulent channel, where an initial planar flame is allowed to propagate upstream in a non-periodic channel. The LES approach uses a flamelet-based combustion model along with standard models for the unresolved subfilter flux terms. It is found that the LES are very accurate in predicting the structure of the turbulent flame front. However, there was a large discrepancy for the transient evolution of the flame, indicating that the flame-boundary layer interaction modulates flame propagation significantly, and the near-wall flame behavior may be non-flamelet like due to the anisotropic of the flow in this region. [Preview Abstract] |
Tuesday, November 25, 2014 1:57PM - 2:10PM |
R34.00005: Numerical investigation of spontaneous flame propagation under Reactivity Controlled Compression Ignition (RCCI) conditions Ankit Bhagatwala, Ramanan Sankaran, Sage Kokjohn, Jacqueline Chen Results from one and two-dimensional direct numerical simulations under dual-fuel Reactivity Controlled Compression Ignition (RCCI) conditions will be presented. These simulations employ an improved model of compression heating through mass source/sink terms developed in a previous work, which incorporates feedback from the flow to follow a predetermined experimental pressure trace. One-dimensional simulations explored the effect of temperature and fuel concentration gradients on the combustion mode. Two-dimensional simulations explored parametric variation in temperature stratification, pressure profiles and n-heptane concentration. Statistics derived from analysis of local diffusion/reaction balances were used to elucidate combustion characteristics for the different cases. Both deflagration and spontaneous ignition modes were observed to co-exist. Higher n-heptane concentration and higher level of thermal stratification resulted in a greater degree of flame propagation, whereas lower n-heptane concentration (higher fraction of iso-octane) and higher pressure resulted in more prevalent autoignition. Starting with a uniform initial temperature and a stratified n-heptane concentration also resulted in a large fraction of combustion occurring through flame propagation. [Preview Abstract] |
Tuesday, November 25, 2014 2:10PM - 2:23PM |
R34.00006: Autoignition: Modes of reaction front propagation evolving from hot spots with defined temperature gradients David R. Kassoy An asymptotic mathematical model, based on the compressible reactive, conservation equations, including transport terms and an arbitrary energy source, is used to quantify the thermo-mechanical consequences of an imposed temperature gradient, $\Delta $T/ l. The mathematical model explains the physics of the gradient system in terms of the local conduction time l$^{2}$/(kappa), where kappa is the characteristic thermal diffusivity, the local acoustic time l/a$_{0}$, where a$_{0}$ is the characteristic acoustic time scale, the characteristic time scale of energy deposition from the source, t$_{\mathrm{ds}}$, and the characteristic energy deposition into and through the gradient region on that time scale. The primary objectives are to predict the magnitude of the induced gas motion and determine when and if transport effects are important. The methodology, related to that in several earlier studies [1-5], can be used to distinguish between detonation and deflagration initiation as well as spontaneous reaction wave propagation. This analysis will help to explain the somewhat enigmatic results in Refs. 6-8. 1,2. Clarke, J.F, Kassoy, D.R. and Riley, N. (1984) Proc. Roy. Soc. A393, 309-351; 3. Kassoy, D.R. (2010), J. Eng Math, 68, 249-262. Kassoy, D.R. (2013), CTM, 18, 101-116. Kassoy, D.R. (2014), AIAA J., doi10.2514, /1J052807. Zeldovich, Y.B. (1980), Combust. Flame, 39, 211-214.. Gu, X.J., Emerson, D.R., Bradley, D. (2003), Comb. Flame, 133, 63-74. Sankaran, R., Hong, G. Hawkes, E.R. Chen J. H., (2005) Proc. Combustion Inst., 30, 875-882. [Preview Abstract] |
Tuesday, November 25, 2014 2:23PM - 2:36PM |
R34.00007: DNS of autoigniting turbulent jet flame Rajapandiyan Asaithambi, Krishnan Mahesh Direct numerical simulation of a round turbulent hydrogen jet injected into vitiated coflowing air is performed at a jet Reynolds number of 10,000 and the results are discussed. A predictor-corrector density based method for DNS/LES of compressible chemically reacting flows is developed and used on a cylindrical grid. A novel strategy to remove the center-line stiffness is developed. A fully developed turbulent pipe flow simulation is prescribed as the velocity inlet for the fuel jet. The flame base is observed to be stabilized primarily by autoignition. Further downstream the flame exhibits a diffusion flame structure with regions of rich and lean premixed regimes flanking the central diffusion flame. The lift-off height is well predicted by a simple relation between the ignition delay of the most-reactive mixture fraction and the streamwise velocity of the jet and coflow. [Preview Abstract] |
Tuesday, November 25, 2014 2:36PM - 2:49PM |
R34.00008: Flame hole dynamics simulation of Sandia Flame F Robert Knaus, John Hewson, Stefan Domino, Carlos Pantano The Sandia Flame ``F'' is a piloted methane/air diffusion flame containing high levels of local extinction. These regions of local extinction reduce the efficiency of combustion and can increase the production of certain pollutants (e.g. carbon monoxide) as well as limit the overall stability of the flame. We present a flame hole dynamics model describing evolution of local extinction zones (flame holes) in a turbulent diffusion flame and apply it to perform a direct numerical simulation of the Sandia Flame F using Sandia's ``SIERRA low Mach Module, Nalu.'' The flame hole dynamics model is a phase-field model that describes the state of the flame (burning or extinguished) through a surface partial differential equation modeling extinction, reignition and advection of the flame state on the moving stoichiometric surface using edge flame properties. The solution of the surface equation is then extended away from the surface and used for state evaluations using a flamelet library with steady flamelets in the burning region and a transient solution in the quenched regions. The flame hole dynamics approach allows tracking extinction and reignition in turbulent diffusion flames without using the computationally costly detailed chemistry explicitly. [Preview Abstract] |
Tuesday, November 25, 2014 2:49PM - 3:02PM |
R34.00009: Estimates of the unsteady extinction frequency in turbulent non-premixed flames with simple stochastic models John Hewson The prediction of statistics for flame extinctions is a key challenge in predictive modeling for non-premixed turbulent combustion. Here we use a simplified measure of unsteady flame extinction based on the excess dissipation above a steady-state extinction scalar dissipation rate integrated over time, a quantity referred to as the dissipation impulse. The dissipation impulse that leads to extinction has been related to the shape of the so-called S-curve describing steady-state flame phase state. The statistics for extinction frequency given this model are studied using a simple Ornstein-Uhlenbeck process to describe dissipation rate evolution. This predicts an extinction rate as a function of (1) the characteristics of the steady-state flame behavior through the S-curve, (2) the probability that the extinction dissipation rate is exceeded and (3) the frequency with which the extinction dissipation rate is observed. The extinction frequency is further interpreted in the context of the analogous Fokker-Planck (FP) equation for the flame temperature-dissipation probability phase space. In the FP context the rate is a function of the phase-space advection fluxes and the probability densities. This suggests simplified estimates for rates of extinction in turbulent non-premixed flames may be possible with less computational effort than has typically been required. [Preview Abstract] |
Tuesday, November 25, 2014 3:02PM - 3:15PM |
R34.00010: Effects of small scale energy injection on large scales in turbulent reaction flows Yuan Xuan Turbulence causes the generation of eddies of various length scales. In turbulent non-reacting flows, most of the kinetic energy is contained in large scale turbulent structures and dissipated at small scales. This energy cascade process from large scales to small scales provides the foundation of a lot of turbulence models, especially for Large Eddy Simulations. However, in turbulent reacting flows, chemical energy is converted locally to heat and therefore deploys energy at the smallest scales. As such, effects of small scale energy injection due to combustion on large scale turbulent motion may become important. These effects are investigated in the case of auto-ignition under homogeneous isotropic turbulence. Impact of small scale heat release is examined by comparing various turbulent statistics (e.g. energy spectrum, two-point correlation functions, and structure functions) in the reacting case to the non-reacting case. Emphasis is placed on the identification of the most relevant turbulent quantities in reflecting such small-large scale interactions. [Preview Abstract] |
Tuesday, November 25, 2014 3:15PM - 3:28PM |
R34.00011: Unsteady plasma fluid simulations in reacting flow Tiernan Casey, Jyh-Yuan Chen, Jie Han, Fabrizio Bisetti, Paul Arias, Hong Im As partially ionized environments, flames possess the potential to admit plasma behavior when the degree of ionization is sufficiently high. As such, the flame behavior can be susceptible to augmentation by applied electric fields. Experiments observing flame stabilization and ignition enhancement under applied electric fields suggest that the dynamics and chemistry of electrons play an important role in catalyzing favorable reactions by exciting relevant neutral species to excited states. This is particularly true when the electric field strength is sufficient to accelerate electrons to such energies that they are no longer in thermal equilibrium with the surrounding neutrals. To investigate these processes, we formulate a multi-fluid governing system of transport equations to account for the non-equilibrium electrons by partitioning the mixture - whilst explicitly re-coupling the component fluids using data determined from solutions to the Boltzmann kinetic equation. A fully-compressible DNS reacting flow solver, S3D, is modified to solve for the charged species transport and time-resolved electric field for applied potentials. We present simulation results for laminar premixed methane flames with a view of informing investigations of microwave-assisted ignition. [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. |
© 2019 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
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700