Bulletin of the American Physical Society
77th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 24–26, 2024; Salt Lake City, Utah
Session C30: Interact: Reacting Flows |
Hide Abstracts |
Chair: Jacqueline Chen, Sandia National Laboratories Room: 255 B |
Sunday, November 24, 2024 10:50AM - 11:20AM |
C30.00001: INTERACT FLASH TALKS: Reacting Flows Each Interact Flash Talk will last around 1 minute, followed by around 30 seconds of transition time. |
|
C30.00002: Lean premixed hydrogen-air flames in intense sheared turbulence and their scaling Martin Rieth, Andrea Gruber, Jacqueline H Chen Hydrogen is a potential carbon-free natural gas replacement for gas turbines. However, combustion properties for lean premixed hydrogen flames can differ significantly from natural gas due to preferential diffusion effects leading to thermo-diffusive instabilities. In this work, we investigate the burning rate scaling with pressure and turbulence intensity, which is not well understood. To this end, we present a series of direct numerical simulations in a temporal shear layer configuration at atmospheric and elevated pressure spanning multiple turbulence intensities. The dataset shows how local turbulent burning rates are significantly amplified at elevated pressure, and how increased turbulence intensity further increases local burning rates. We will highlight main features of the turbulent burning behavior of the flames at different conditions. In addition, we will discuss potential scaling of turbulent burning rates with non-dimensional parameters including this new dataset and additional experimental datasets. |
|
C30.00003: Combustion reduced-order modeling with nonlinear projections Dallin Littlewood, James Sutherland Reduced-order models (ROMs) of high-dimensional combustion systems are used to decrease the computational costs of simulations by reducing the number of equations needed to describe a system. These ROMs are constructed through dimensionality reduction (projections) to re-parameterize the system creating a low-dimensional manifold (LDM) where the corresponding LDM parameters are transported. Projections have typically been obtained through linear methods, such as principal component analysis (PCA) or linear encoder neural networks, as linear projections allow low-dimensional transport equations to be trivially derived. However, as the complexity of reaction mechanisms increases for different fuels, more LDM parameters are needed to provide a LDM with good topological quality. Nonlinear projections achieve better topological quality than linear projections for the same number of LDM parameters, but the derivation of low-dimensional transport equations becomes nontrivial. This work derives low-dimensional laminar flamelet transport equations for nonlinear projections and demonstrates the ability to converge steady solutions and move between dissipation rates on the LDM, and is demonstrated for hydrogen, syngas, and methane flamelets that have been reduced to two latent dimensions. |
|
C30.00004: Local PCA transport model for reduced-order combustion modeling Rafi Malik, Hong G Im In the field of reduced-order models (ROM) for reacting flow simulations, data-driven Principal Component Analysis (PCA) has shown its effectiveness in reducing the dimensionality without losing accuracy. In particular, the PC-transport approach has been applied to different type of systems, ranging from homogeneous 0D reactors to 3D turbulent jet flames. In the PC-transport approach, the thermo-chemical state-space is parameterized using a reduced number of latent variables, the PC scores, which are transported in flow simulations instead of the original physical variables. However, those scores are defined using a global PCA basis matrix, the latter being unique for the entire PCA manifold. This global definition of the PC scores limits the achievable degree of reduction, especially in highly nonlinear data sets where a large number of scores are needed for an acceptable accuracy. The present work seeks to enhance the PC-transport model using a local approach, where the PCA low-dimensional manifold is divided into clusters using an unsupervised algorithm based on Vector Quantization PCA (VQPCA), a PCA analysis is performed separately in each cluster providing an optimal local basis in each of them, and thereafter those local scores are transported in the simulation instead of the global scores, allowing for a higher degree of reduction. The approach is tested in homogeneous 0D reactors for different fuels using detailed chemistry. |
|
C30.00005: Phy-ChemNODE: A Physics-Enhanced Neural Ordinary Differential Equations Approach for Accelerating Stiff Chemical Kinetic Computations Pinaki Pal, Tadbhagya Kumar, Anuj Kumar A first-of-its-kind neural ordinary differential equations (NODE) based approach, known as ChemNODE, was developed at Argonne National Laboratory (Owoyele and Pal, Energy & AI 2021) to accelerate detailed chemistry computations. In this framework, chemical source terms predicted by a neural network are integrated during training, and by computing the required derivatives, the neural network weights are optimized to minimize the difference between the predicted and ground-truth thermochemical state solutions. ChemNODE was enhanced by incorporating elemental mass conservation constraints directly into the loss function during training (Kumar et al. 2023, https://arxiv.org/abs/2312.00038). Both a-priori and a-posteriori tests (for hydrogen-air 0D autoignition) demonstrated that the enhanced Phy-ChemNODE framework not only improves physical consistency of the data-driven model, but also enables faster model training. A-posteriori studies performed by coupling Phy-ChemNODE with a CFD solver exhibited robustness and generalizability to unseen initial conditions from within (interpolative capability) as well as outside (extrapolative capability) the training regime. Lastly, Phy-ChemNODE was recently extended to hydrocarbon chemistry by incorporating a non-linear autoencoder (AE) for dimensionality reduction, wherein the NODE learns the temporal evolution of the dynamical system in a reduced-order latent space obtained from the AE. Both the AE and NODE were trained together in an end-to-end manner. Numerical studies performed for methane-oxygen chemistry (32 chemical species; 266 chemical reactions) demonstrated 10X speedup achieved by Phy-ChemNODE compared to solving for the full system of stiff ODEs, while preserving prediction fidelity. |
|
C30.00006: Structure and Dynamics of Three Dimensional Spray Detonations in Jet Fuels Sai Sandeep Dammati, Alexei Y Poludnenko Recent years have seen a surge in numerical studies of spray detonations due to their relevance in hypersonic propulsion devices and energy conversion systems. Having said that, all such prior studies are currently limited to two-dimensional (2D) channels with ideal boundary conditions. In this work, for the first time, we carry out three dimensional (3D) numerical simulations of detonations propagating in a liquid jet fuel spray. In particular, the simulations are carried out in dodecane/air mixtures using an Eulerian-Lagrangian formulation with complex chemistry, complete molecular transport, realistic boundary conditions and state-of-the art spray sub-models for droplet drag, atomization and evaporation. The obtained detonation properties are then contrasted with those of purely gaseous detonations. A sensitivity study is conducted by varying the droplet break-up time in the droplet atomization model. The results show that in case of fast atomization, the detonation cells are slightly irregular but are qualitatively and quantitatively similar to those of purely gaseous detonations. However, when the atomization is slow, the cells become more regular in shape but smaller in size when compared to the purely gaseous case. Overall, these results demonstrate that detonation properties of jet fuels in 3D geometries are both qualitatively and quantitatively different from those obtained from 2D geometries. |
|
C30.00007: Numerical Investigation of Detonation Cell Size in Hydrogen and Hydrocarbon Fuels Abeetath Ghosh, Sai Sandeep Dammati, Alexei Y Poludnenko Detonation cell size is one of the main characteristic scales of a multi-dimensional detonation, which plays an important role in the design of practical systems due to its empirical correlation with essential design metrics, such as minimum channel size and critical tube diameter. Despite the advances in modeling of multidimensional detonations, accurate prediction of the experimental detonation cell size remains elusive, except under specific conditions and for certain fuel mixtures. The underlying causes of this discrepancy and the fundamental mechanisms, which determine the cell size, are not fully understood. This study systematically compares numerical and experimental cell sizes across a wide range of pressures, temperatures, equivalence ratios and fuel mixtures. The simulation are performed using a fully compressible, adaptive mesh refinement, reacting flow solver Athena-RFX++ with complex chemical kinetics and mixture-averaged molecular transport. The effect of the ignition time and its temperature sensitivity on the cell size is investigated, with particular emphasis on the relative importance of different shock overdrives present in a multidimensional detonation front. Further, the impact of the simulation dimensionality and numerical resolution on the cell size is explored. Finally, the findings are used to identify potential limitations in the existing physio-chemical models, which may contribute to the inaccuracies in the predicted detonation structure. |
|
C30.00008: The Influence of Thermodiffusive Flame-front Instabilities on Deflagration to Detonation Transition Kayden J Jenkins, Alex G Novoselov The need to reduce carbon emissions has inspired a thorough investigation into alternative fuels. One of the most promising alternatives being researched currently is hydrogen due to its high energy density per mass compared to fuels like gasoline. Some promising ideas for propulsion advancements employ the use of detonations (pulse detonation engines/rotating detonation engines), and as such, understanding the behavior and inception of detonations for fuels like hydrogen is incredibly important. Much research has been conducted concerning how the use of obstructions in the combustion chamber can trigger deflagration to detonation transition (DDT), but little has been focused on how flame-front instabilities in hydrogen can contribute to DDT. This work is focused on quantifying how thermo-diffusive instabilities affect DDT. Direct Numerical Simulations of flames with unity and non-unity Lewis numbers are performed to understand the role instabilities play in DDT. Comparisons of flame-front dynamics between these otherwise identical flames shed light on the impact of thermo-diffusive instabilities. |
|
C30.00009: Numerical Investigation of Shock-Turbulence Interactions in a Chemically Reactive Flow Ibrahem Alshybani, Zhaorui Li, Farhad A. Jaberi Shock-Turbulence-Combustion Interactions (STCI) in a fundamental flow configuration are studied in detail by the direct numerical simulations (DNS) method. DNS computations are conducted by solving fully compressible flow equations with a high-order hybrid monotonicity-preserving compact finite-difference scheme together with a single-step chemical reaction model for a non-premixed reaction. Robustness is ensured with a damping/convective buffer zone at the end of computational domain and grid convergence plus statistical convergence tests. We explore the effects of density variations, turbulence intensity, and initial scalar length scales on the post-shock turbulence, mixing and combustion in the STCI set up. The presented work also considers the influence of an intensely exothermic reaction and a high-intensity compressible isotropic turbulence on the shock properties and its movement. Our findings offer insights into the complex interplay between turbulence, shock waves, and chemical reactions in a well-characterized fundamental flow configuration. |
|
C30.00010: Flame structure of the Second Stage in Low-Emission Ammonia Rich-Quench-Lean Combustion Martin Rieth, Andrea Gruber, Evatt Hawkes, Jacqueline H Chen Ammonia is a promising carbon-free alternative to natural gas in dispatchable power generation. While carbon dioxide emissions are eliminated from the combustion process with ammonia, nitrogen oxide emissions (NOx) pose a significant challenge. One of the most promising strategies reducing such emissions in gas turbines is the rich-quench-lean (RQL) staged combustion system, where the combustion process is divided into two stages. With ammonia-based fuels, typically the first stage features a fuel-rich flame while the second stage comprises air injection to consume the remaining fuel. In this presentation, we will focus on the second stage combustion process by means of direct numerical simulations (DNS) of a simplified and downscaled combustor. The analysis reveals details of the interaction between the cold turbulent air jets and the hot hydrogen-containing flow from the first stage. We will discuss the impact of the air split ratio on the flame structure, relate the combustion process to local flame features and topologies, and shed light on processes such as preferential diffusion. We will also discuss NO and N2O emissions, in particular for an additional case with ammonia slip. |
|
C30.00011: Assessment of manifold-based combustion models for combustion of sustainable aviation fuel in a gas turbine combustor Bruce Alan Perry, Sreejith Nadakkal Appukuttan, Mohammad J Rahimi, David Montgomery, Marc Day, Shashank Yellapantula Uncertainty and risk associated with how differing sustainable aviation fuel (SAF) properties affect combustor performance relative to petroleum-based jet fuel is a major barrier to certification and widespread adoption of these fuels. Predictive simulations of SAF combustion in realistic aviation gas turbine combustors can help to hasten the adoption of these fuels by providing confidence that the fuels will meet certification requirements. In this talk, we present progress toward this capability by assessing candidate reduced-order manifold combustion models including flamelet generated manifolds and flamelet/progress variable models for use in large eddy simulations of the multi-modal combustion in a lean premixed prevaporized gas turbine combustor. These comparisons are based on results using the PeleLMeX solver, which incorporates mutliphysics modeling capability with adaptive mesh refinement to allow for high-fidelity simulation of realistic combustors. Using these models, the combustion of Hydrotreated Esters and Fatty Acids (HEFA), a leading SAF candidate, is compared to the combustion of Jet-A fuel. The HEFA simulations take advantage of recent physical properties measurements and chemical kinetics surrogate development in order to accurately compare combustion characteristics. |
|
C30.00012: A physics embedded neural-network sub-grid-scale model for simulating flames in turbulence Seung Won Suh, Jonathan F MacArt, Luke Olson, Jonathan Ben Freund A neural network for large-eddy simulation (LES) of turbulent reacting flows is embedded in the governing equations as a sub-grid-scale closure. Its training is fully coupled with the physics of the filtered governing equations: the network parameters are optimized based on model-predicted observables, which entails coupling numerical solutions of the adjoint governing equations with the usual backpropagation training gradient. It is also designed to preserve mass fraction boundedness with a total variation diminishing (TVD) property. We demonstrate the method for a statistically planar premixed flame in isotropic turbulence with a single-species, single-step, and irreversible chemical reaction. It is trained to correct the short-time deviation from a spatially filtered solution of direct numerical simulation. The trained model is then applied to a longer LES and is evaluated based on predicted statistical quantities of interest. Finally, the trained model is demonstrated on the out-of-sample configuration of an expanding spherical flame in isotropic turbulence. |
|
C30.00013: Directional considerations of diffusive fluxes in manifold modeling for lean ammonia/hydrogen/nitrogen-air laminar premixed flames Sydney L Rzepka, Katie E VanderKam, Michael E Mueller Ammonia is a promising hydrogen carrying fuel that, when partially cracked into mixtures of ammonia/hydrogen/nitrogen, can have similar combustion properties as methane. Containing hydrogen, under fuel-lean conditions, ammonia/hydrogen/nitrogen-air premixed flames can be thermodiffusively unstable, affecting flame propagation speeds as well as the local formation of nitrogen oxides and nitrous oxide. Understanding and accurate modeling of these pollutants are critical for the viability of ammonia as a zero-carbon fuel. Manifold-based combustion models significantly decrease computational cost by mapping the high dimensional thermochemical state to a lower dimensional manifold. Previous work evaluated a premixed manifold model in progress variable that included differential diffusion and flame curvature effects, but the manifold model did not accurately predict pollutants. Strong differential diffusion effects leading to significant diffusive transport in the direction orthogonal to the manifold coordinate were hypothesized to be responsible for this model failure. The current work investigates this hypothesis by computing the magnitude of all diffusive fluxes as well as those along the manifold coordinate and offers further insight into how these effects can be modeled. |
|
C30.00014: Transport and reignition of n-Dodecane fuel droplets in a transverse jet in supersonic crossflow Carson Ramm, Prashant Tarey, John Boles, Tanner Nielsen, Matthew Goodson, Jacob A McFarland, Mesbah Uddin, Praveen K Ramaprabhu
|
|
C30.00015: Numerical investigation of soot formation in sustainable aviation fuels Bruno S. Soriano, Jacqueline H Chen The increasing demand in the aviation sector motivates the study of soot formation and alternative fuels to minimize the environmental impact on climate change. The formation of condensation trails (contrails) by aircraft engines corresponds to approximately 4-5 % of the total net effective radiative forcing causing global warming. The contrails are composed of ice crystals formed primarily from the interaction of soot particles and water vapor emitted by the engine. The objective of this work is to analyze the processes related to soot formation for the commercially available Jet-A, and the C1 fuel representative of a Sustainable Aviation Fuel with low aromatic content and low ignition propensity. First, a reduced chemical mechanism is developed for both fuels using experimental measurements in counter-flow configurations and predictions using the state-of-the-art detailed chemical mechanism from Lawrence Livermore National Laboratory. Soot predictions with the reduced mechanisms are obtained with the DNS code, PeleLMeX, and a state-of-the-art soot model (Hybrid Method of Moments). The predictions revealed that Jet-A and C1 have a reduction in soot volume fraction with the increase in fluid dynamic strain. However, the interaction between the soot nucleation, surface growth, condensation and oxidation, and their response to strain is different. The findings of this research provide us a better understanding on the fundamental aspects of soot emissions in sustainable aviation fuels. |
|
C30.00016: In-Situ Adaptive Manifolds for Soot and Emissions Predictions in Turbulent Reacting Flows Matthew X Yao, Michael E Mueller Manifold-based combustion models reduce the computational cost of turbulent reacting flow simulations by projecting the thermochemical state onto a low-dimensional manifold, which can be computed separately from the flow solver. Traditionally, the model is pretabulated by precomputing solutions to a set of manifold equations. For soot and emissions, additional dimensions are required to account for heat losses, so these pretabulated databases can become very memory intensive, and many of the states may not even be accessed. In-Situ Adaptive Manifolds (ISAM) has recently been developed, in which necessary manifold solutions are computed on-the-fly and stored for reuse with In-Situ Adaptive Tabulation (ISAT). In this work, ISAM is coupled to a soot model based on the Hybrid Method of Moments (HMOM), and a new approach for incorporating heat losses into ISAM has been developed. The model is demonstrated and validated against the Sandia Sooting Flame. The computational efficiency of the sooting ISAM framework will be discussed. |
|
C30.00017: Inherently high-speed manifold-based modeling of supersonic turbulent combustion John Benno Boerchers, Michael E Mueller Manifold-based turbulent combustion models were developed in the low Mach number limit where thermodynamic pressure variation and viscous heating are assumed to be negligible. Various approaches have attempted to account for compressibility effects through the addition of ad hoc correction terms that introduce thermodynamic inconsistencies between model and flow solver, and, although a more recent iterative approach allows for complete thermodynamic consistency between model and flow solver, all of these models still ultimately rely on solutions to the low Mach manifold equations. In this work, high-fidelity data from a high-speed turbulent reacting flow in a scramjet-like geometry is used to inform an inherently high-speed manifold model that includes the high-speed terms directly in the manifold equations, avoiding the need for ad hoc correction terms or iterative approaches. The inclusion of these terms also allows the model to capture local compressibility effects such as shockwaves. Multiple datasets are then analyzed to understand how errors in accounting for compressibility effects compare to other manifold-based modeling errors, specifically the modality of the combustion processes. |
|
C30.00018: Measurements of transient sequences for fully premixed hydrogen/air swirling flames. C.P. Premchand, Paul P PALIES Decarbonization based on hydrogen represents a pivotal strategy in the transition to cleaner energy and transport systems. Towards this goal, transient sequences of lean blowout, blowoff, and flashback are studied with high-speed direct flame imaging. These transient sequences are documented from an identical starting point: a 100% H₂/Air swirl-stabilized condition also described. The sequences are depicted on a regime map function of equivalence ratio and bulk velocity, considering the evolution of both the Karlovitz number and the Lewis number. Sequences are acquired on a fully premixed, hydrogen/air, swirl-stabilized laboratory scale experiment. Data acquisitions are made with a high-speed camera (sensitive to both UV and visible light wavelengths) to capture flame chemiluminescence, complemented by schlieren imaging to visualize and describe flame evolution during each transient sequence. The presentation also provides insights on the effects of preferential diffusion versus turbulence on flame wrinkling for jet flame. Finally, the variations in flame intensity amplitude by integrating flame images into a time-series signal across different cases are discussed. This study aims at investigating unsteady processes for hydrogen premixed flames. |
|
C30.00019: Turbulent Premixed Flame Annihilation and Manifold-Based Models Michael D Walker, Katie E VanderKam, Michael E Mueller In premixed combustion, manifold-based models project the thermochemical state onto a one-dimensional space in progress variable, a scalar advancing from unburned reactants to products at thermodynamic equilibrium. Under conditions where turbulence intensity is high and the length scales of turbulence are small (comparable to the flame thickness), the curvature of the flame increases until the shape is so corrugated that it may interact with itself. These flame-flame interactions are important for flame propagation as a principal mechanism of local extinction and continuous annihilation of turbulent flame surfaces. Whether these interactions can be captured by a manifold-based model remains an open question since the flame-flame interactions give rise to local extrema in progress variable dissipation rate at non-zero/-unity progress variable. This work explores the influence of flame-flame interactions on the local thermochemical state using Direct Numerical Simulation of turbulent premixed planar hydrogen/air flames. The conditional statistics of this database are compared to predictions from the manifold model using local progress variable dissipation rates extracted from the databases. The work concludes by discussing what additional physics, if any, may be needed in the manifold model to capture flame-flame interactions. |
|
C30.00020: Molecular level simulations of hydrogen-air reacting flows under thermal and chemical non-equilibrium Shrey Trivedi, Ahren W Jasper, John K. Harvey, Jacqueline H Chen Molecular level simulations of hydrogen-air detonation waves are performed using the Direct Simulation Monte Carlo (DSMC) method [1]. The conditions presented are relevant to those in Rotating Detonation Engines or Scramjets [2]. One-dimensional (1-D) detonation waves simulated under near-equilibrium conditions have been shown to agree well with the equilibrium solution with only slight variations when using the standard equilibrium TCE model in DSMC code SPARTA [3]. The equilibrium results were obtained with the Zel'dovich-von Neumann-Döring (ZND) solution obtained from the Shock and Detonation (SDT) toolbox [4]. In this paper, an effective temperature (Teff) formulation is explored which can account for ro-vibrational non-equilibrium. The use of Teff within the TCE model results in a good agreement of state-specific reaction rates from a 0-D isothermal bath with those obtained from the Quasi-Classical Trajectory (QCT) calculations. A 1-D detonation case is simulated with preheated reactants at 900 K and 0.3 atm pressure and Mach number M = 3.0. The results from the effective temperature model are compared against the equilibrium DSMC result and the differences between the two are highlighted. Simulationd of 2-D detonations waves are currently in progress. |
Sunday, November 24, 2024 11:20AM - 12:50PM |
C30.00021: INTERACT DISCUSSION SESSION WITH POSTERS: Reacting Flows After each Flash Talk has concluded, the Interact session will be followed by interactive poster or e-poster presentations, with plenty of time for one-on-one and small group discussions. |
|
C30.00022: Scale-based energy transfer mechanisms in supersonic combustion Clara M Helm, Davy Brouzet, Brett Bornhoft, Timothy P Gallagher, David Peterson Many turbulence closure models for Large Eddy Simulation (LES) are based on the concept of the turbulence energy cascade in which the turbulence energy is described as being transferred in an inviscid manner from large scales to progressively smaller and smaller scales until it is dissipated at the Kolmogorov length scales. The validity of this assumption of net-downward energy transfer across the LES filter scale is uncertain in the regime of compressible reacting turbulence. In this work, we investigate the phenomena of scale-dependent turbulent kinetic energy backscatter in the DNS of both non-reacting and reacting spatially developing shear layers with the goal of understanding how the energy cascade is altered in the presence of strong shocks and heat release. A configuration in which an incident shock impinges on the shear layer is also considered. Results show that at certain filter-to-grid ratios, significant amounts of local backscatter do occur. Correlations between the backscatter and other flow phenomena, such as heat release, pressure dilatation, and vorticity, will be reported. Cross-scale transport of reacting species scalar variance is also investigated. |
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