Direct Numerical Simulations (DNS) are performed to investigate the process of spontaneous ignition of hydrogen flames at laminar, turbulent, adiabatic and non-adiabatic conditions. Mixtures of hydrogen and vitiated air at temperatures representing gas-turbine reheat combustion are considered. Adiabatic spontaneous ignition processes are investigated first, providing a quantitative characterization of stable and unstable flames. Results indicate that, in hydrogen reheat combustion, compressibility effects play a key role in flame stability and that unstable ignition and combustion are consistently encountered for reactant temperatures close to the mixture's characteristic crossover temperature. Furthermore, it is also found that the characterization of the adiabatic processes is also valid in the presence of non-adiabaticity due to wall heat-loss. Finally, a quantitative characterization of the instantaneous fuel consumption rate within the reaction front is obtained and of its ability, at auto-ignitive conditions, to advance against the approaching turbulent flow of the reactants, for a range of different turbulence intensities, temperatures and pressure levels.
In the present work, premixed combustion in a turbulent boundary layer under auto-ignitive conditions is investigated using direct numerical simulation (DNS). The turbulent inflow of the reactive DNS is obtained by temporal sampling of a corresponding inert DNS of a turbulent boundary layer at a location with Reτ= 360, where Reτ is the friction Reynolds number. The reactants of the DNS are determined by mixing the products of lean natural gas combustion and a H2/N2 fuel jet, resulting in a lean mixture of high temperature with a short ignition delay time. In the free stream the reaction front is stabilized at a streamwise location which can be predicted using the free stream velocity U∞ and the ignition delay time τig. Inside the boundary layer, combustion modifies the near-wall coherent turbulent structures considerably and turbulence results in reaction front wrinkling. The combustion modes in various regions were examined based on the results of displacement velocity, species budget and chemical explosive mode analysis (CEMA). It was indicated that flame propagation prevails in the near-wall region and auto-ignition becomes increasingly important as the wall-normal distance increases. The interactions of turbulence and combustion were studied through statistics of reaction front normal vector and strain rate tensor. It was found that the reaction front normal preferentially aligns with the most compressive strain rate in regions where the effects of heat release on the strain rate are minor and with the most extensive strain rate where its effects are significant. Negative correlations between the wall heat flux and flame quenching distance were observed. A new quenching mode, back-on quenching, was identified. It was found that the heat release rate at the wall is the highest when head-on quenching occurs and lowest when back-on quenching occurs.
It has been previously demonstrated that thermal gas expansion might have a role in boundary layer flashback of premixed turbulent flames [Gruber et al., J Fluid Mech 2012], inducing local flow-reversal in the boundary layer's low-velocity streaks on the reactants’ side of the flame and facilitating its upstream propagation. We perform a two-dimensional numerical investigation of the interaction between a periodic shear flow and a laminar premixed flame. The periodic shear is a simplified model for the oncoming prolonged streamwise velocity streaks with alternating regions of high and low velocities found in turbulent boundary layers in the vicinity of the walls. The parametric study focuses on the amplitude and wavelength of the periodic shear flow and on the gas expansion ratio (unburnt-to-burnt density ratio). With the increase of the amplitudes of the periodic shear flow and of the gas expansion, the curved flame velocity increases monotonically. The flame velocity dependence on the periodic shear wavelength is non-monotonic, which is consistent with previous theoretical studies of curved premixed flame velocity. The flame shape that is initially formed by the oncoming periodic shear appears to be metastable. At a later stage of the flame propagation, the flame shape transforms into the stationary one dominated by the Darrieus-Landau instability.
The evaporation and combustion of n-heptane droplets under high pressure (20 atm) conditions were investigated using three-dimensional direct numerical simulations (DNS) in the present work. Two ambient temperatures were considered, i.e. 860 K and 1180 K. Low-temperature combustion, which is controlled by low-temperature chemical pathways, occurs in the 860 K cases while single-stage combustion occurs in the 1180 K cases. Droplet evaporation and combustion in isotropic turbulence were considered. The interactions of turbulence, evaporation and combustion in a cloud of droplets were explored. It was found that turbulence promotes the evaporation process. The temperature and mixture-fraction are negatively correlated after evaporation, which impacts the subsequent ignition process. For both low-temperature and single-stage combustion, ignition occurs in regions with low scalar dissipation rate. Low-temperature ignition first appears in lean mixtures while single-stage ignition in rich mixtures. Ignition kernels evolve into reaction fronts, which propagate towards thermally and compositionally stratified mixture and consume the remaining fuel. Spontaneous ignition front is dominant in low-temperature combustion while deflagrative front is playing an important role in single-stage combustion. The interactions of turbulence and the reaction front structures were examined. Cylindrical elements are the most probable shape of the reaction front for turbulent droplet combustion. The reaction front normal is misaligned with the vorticity vector for both low-temperature and single-stage combustion, indicating that the vortical structures preferentially locate along the tangential plane of the reaction front, which promotes the observed cylindrical reaction front structures in turbulent combustion of droplets.
Ammonia has been identified as a promising energy carrier that produces zero carbon dioxide emissions when used as a fuel in gas turbines. Although the combustion properties of pure ammonia are poorly suited for firing of gas turbine combustors, blends of ammonia, hydrogen, and nitrogen can be optimized to exhibit premixed, unstretched laminar flame properties very similar to those of methane. There is limited data available on the turbulent combustion characteristics of such blends and important uncertainties exist related to their blow-out behavior. The present work reports experimental measurements of the blow-out limits in an axisymmetric unconfined bluff-body stabilized burner geometry of NH3/H2/N2-air flame, comprised of 40% NH3, 45% H2, and 15% N2 by volume in the “fuel” blend. Blow-out limits for the NH3/H2/N2-air flames are compared to those of methane–air flames. OH PLIF and OH chemiluminescence images of the flames just prior to blow-out are presented. Furthermore, two large-scale Direct Numerical Simulations (DNS) of temporally evolving turbulent premixed jet flames are performed to investigate differences in the turbulence-chemistry interaction and extinction behavior between the NH3/H2/N2-air and methane–air mixtures. The experiments reveal that the blow-out velocity of NH3/H2/N2-air flames is an order of magnitude higher than that of methane–air flames characterized by nearly identical unstretched laminar flame speed, thermal thickness and adiabatic flame temperature. Results from the DNS support the experimental observation and clearly illustrate that a methane–air mixture exhibits a stronger tendency towards extinction compared to the NH3/H2/N2-air blend for identical strain rates. Furthermore, the DNS results reveal that, even in the presence of intense sheared turbulence, fast hydrogen diffusion into the spatially distributed preheat layers of the fragmented and highly turbulent flame front plays a crucial role in the enhancement of the local heat release rate and, ultimately, in preventing the occurrence of extinction.
A flamelet analysis of a highly resolved direct numerical simulation (DNS) of a multi-injection flame with both auto-ignition and ignition induced by flame-flame interaction was conducted. A novel method was proposed to identify the different combustion modes of ignition processes using generalized flamelet equations. A state-of-the-art DNS database for a multi-injection n-dodecane flame in a diesel engine environment was investigated. Three-dimensional flamelets were extracted from the DNS at different time instants with a focus on auto-ignition and interaction-ignition processes. The influences of mixture field interactions and the scalar dissipation rate on the ignition process were examined by varying the species composition boundary conditions of the transient flamelet equations. Results showed that auto-ignition is delayed if the burned products are added to the oxidizer side of the flamelet, and the ignition delay time is sensitive to the scalar dissipation rate. The significance of mass diffusion in the flame-normal direction is reduced due to the existence of burned products in the oxidizer stream. Budget analyses of the generalized flamelet equations revealed that the transport along the mixture fraction iso-surface is insignificant during the auto-ignition process, but becomes important when interaction-ignition occurs, which is further confirmed through a flamelet regime classification method.
In the present work, three-dimensional turbulent non-premixed oblique slot-jet flames impinging at a wall were investigated using direct numerical simulation (DNS). Two cases are considered with the Damköhler number (Da) of case A being twice that of case B. A 17 species and 73-step mechanism for methane combustion was employed in the simulations. It was found that flame extinction in case B is more prominent compared to case A. Reignition in the lower branch of combustion for case A occurs when the scalar dissipation rate relaxes, while no reignition occurs in the lower branch for case B due to excessive scalar dissipation rate. A method was proposed to identify the flame quenching edges of turbulent non-premixed flames in wall-bounded flows based on the intersections of mixture fraction and OH mass fraction iso-surfaces. The flame/wall interactions were examined in terms of the quenching distance and the wall heat flux along the quenching edges. There is essentially no flame/wall interaction in case B due to the extinction caused by excessive turbulent mixing. In contrast, significant interactions between flames and the wall are observed in case A. The quenching distance is found to be negatively correlated with wall heat flux as previously reported in turbulent premixed flames. The influence of chemical reactions and wall on flow topologies was identified. The FS/U and FC/U topologies are found near flame edges, and the NNN/U topology appears when reignition occurs. The vortex-dominant topologies, FC/U and FS/S, play an increasingly important role as the jet turbulence develops.
Accurate modeling of mixing in large-eddy simulation (LES)/transported probability density function (PDF) modeling of turbulent combustion remains an outstanding issue. The issue is particularly salient in turbulent premixed combustion under extreme conditions such as high-Karlovitz number Ka. The present study addresses this issue by conducting an a priori analysis of a power-law scaling based mixing timescale model for the transported PDF model. A recently produced DNS dataset of a high-Ka turbulent jet flame is used for the analysis. A power-law scaling is observed for a scaling factor used to model the sub-filter scale mixing timescale in this high-Ka turbulent premixed DNS flame when the LES filter size is much greater than the characteristic thermal thickness of a laminar premixed flame. The sensitivity of the observed power-law scaling to the different viewpoints (local or global) and to the different scalars for the data analysis is examined and the dependence of the model parameters on the dimensionless numbers Ka and Re (the Reynolds number) is investigated. Different model formulations for the mixing timescale are then constructed and assessed in the DNS flame. The proposed model is found to be able to reproduce the mixing timescale informed by the high-Ka DNS flame significantly better than a previous model.
In the present work, three-dimensional turbulent non-premixed oblique slot-jet flames impinging at a wall were investigated using direct numerical simulation (DNS). Two cases are considered with the Damköhler number (Da) of case A being twice that of case B. A 17 species and 73-step mechanism for methane combustion was employed in the simulations. It was found that flame extinction in case B is more prominent compared to case A. Reignition in the lower branch of combustion for case A occurs when the scalar dissipation rate relaxes, while no reignition occurs in the lower branch for case B due to excessive scalar dissipation rate. A method was proposed to identify the flame quenching edges of turbulent non-premixed flames in wall-bounded flows based on the intersections of mixture fraction and OH mass fraction iso-surfaces. The flame/wall interactions were examined in terms of the quenching distance and the wall heat flux along the quenching edges. There is essentially no flame/wall interaction in case B due to the extinction caused by excessive turbulent mixing. In contrast, significant interactions between flames and the wall are observed in case A. The quenching distance is found to be negatively correlated with wall heat flux as previously reported in turbulent premixed flames. The influence of chemical reactions and wall on flow topologies was identified. The FS/U and FC/U topologies are found near flame edges, and the NNN/U topology appears when reignition occurs. The vortex-dominant topologies, FC/U and FS/S, play an increasingly important role as the jet turbulence develops.
Krisman, Alex; Meagher, Patrick; Zhao, Xinyu; Park, Ji-Woong; Lu, Tianfeng; Chen, Jacqueline H.
The safe operation of aeronautical engines requires an understanding of flame ignition, propagation and extinction. In this study, direct numerical simulations are performed using a 29 species reduced chemical mechanism for jet fuel surrogate Jet A to understand the flame quenching process. Here, initially laminar spherical flames of varying sizes and equivalence ratios are subject to an identical periodic domain of decaying and isotropic high intensity turbulence with a turbulent Reynolds number of 2400. All cases become quenched, except for the larger kernel with lower Karlovitz number. An analysis of the flame structure shows broadened preheat zone, flame shortening on the product side, differential species diffusion and partial fuel pyrolysis in the fresh mixture. Two extinction mechanisms are identified arising from flame shortening and high flame stretch. Flame shortening occurs due to turbulence-chemistry interactions that resemble the flame–flame interaction in a laminar counterflow reactant-to-reactant configuration, which contorts and breaks up the ignition kernel. Flame stretch is a local effect that attenuates the heat release rate and causes the flame to retreat towards the product mixtures, similar to what has been observed for reactant-to-product laminar counterflow flames. Chemical explosive mode analysis was also performed to quantify the flame structure and local combustion mode. The diffusion–reaction balance in pinched-off flame islands favors extinction of these smaller structures, while auto-ignition modes are observed within the flame kernel after fresh mixture is engulfed and preheated in the product kernel. Statistics of the density-weighted displacement speed conditional on local combustion mode indicates strong correlation between the local extinction mode and negative displacement speed. The local balance between diffusion and reaction ultimately determines the propensity for local extinction in both laminar and turbulent flames, the extent of which has an impact on global flame propagation.
In this paper, the first flamelet analysis is conducted of a highly resolved DNS of a multi-injection flame with both auto-ignition and ignition induced by flame-flame interaction. A novel method is proposed to identify the different combustion modes of ignition processes using generalized flamelet equations. The state-of-the-art DNS database generated by Rieth et al. (US National Combustion Meeting, 2019) for a multi-injection flame in a Diesel engine environment is investigated. Three-dimensional flamelets are extracted from the DNS at different time instants with a focus on auto-ignition and interaction-ignition processes. The influences of mixture field interactions and the scalar dissipation rate on the ignition process are investigated by varying the species composition boundary conditions of the transient flamelet equations. Budget analyses of the generalized flamelet equations show that the transport along the mixture fraction iso-surface is insignificant during the auto-ignition process, but becomes important when interaction-ignition occurs, which is further confirmed through a flamelet regime classification method.
It has been predicted that several changes will occur when premixed flames are subjected to the extreme levels of turbulence that can be found in practical combustors. This paper is a review of recent experimental and DNS results that have been obtained for the range of extreme turbulence, and it includes a discussion of cases that agree or disagree with predictions. “Extreme turbulence” is defined to correspond to a turbulent Reynolds number (ReΤ, based on integral scale) that exceeds 2800 or a turbulent Karlovitz number that exceeds 100, for reasons that are discussed in Section 2.1. Several data bases are described that include measurements made at Lund University, the University of Sydney, the University of Michigan and the U.S. Air Force Research Lab. The data bases also include DNS results from Sandia National Laboratory, the University of New South Wales, Newcastle University, the California Institute of Technology and the University of Cambridge. Several major observations are: (a) DNS now can be achieved for a realistic geometry (of the Lund University jet burner) even for extreme turbulence levels, (b) state relations (conditional mean profiles) from DNS and experiments do tend to agree with laminar profiles, at least for methane-air and hydrogen-air reactants that are not preheated, and (c) regime boundaries have been measured and they do not agree with predicted boundaries. These findings indicate that the range of conditions for which flamelet models should be valid is larger than what was previously believed. Additional parameters have been shown to be important; for example, broken reactions occur if the “back-support” is insufficient due to the entrainment of cold gas into the product gas. Turbulent burning velocity measurements have been extended from the previous normalized turbulence levels (u’/SL) of 24 up to a value of 163. Turbulent burning velocities no longer follow the trend predicted by Shchelkin but they tend to follow the trend predicted by Damköhler. The boundary where flamelet broadening begins was measured to occur at ReTaylor = 13.8, which corresponds to an integral scale Reynolds number (ReT) of 2800. This measured regime boundary can be explained by the idea that flame structure is altered when the turbulent diffusivity at the Taylor scale exceeds a critical value, rather than the idea that changes occur when Kolmogorov eddies just fit inside a flamelet. A roadmap to extend DNS to complex chemistry and to higher Reynolds numbers is discussed. Exascale computers, machine learning, adaptive mesh refinement and embedded DNS show promise. Some advances are reviewed that have extended the use of line and planar PLIF and CARS laser diagnostics to studies that consider complex hydrocarbon fuels and harsh environments.
Shin, Dong H.; Richardson, Edward S.; Aparece-Scutariu, Vlad; Minamoto, Yuki; Chen, Jacqueline H.
The link between the distribution of fluid residence time and the distribution of reactive scalars is analysed using Direct Numerical Simulation data. Information about the reactive scalar distribution is needed in order to model the reaction terms that appear in Large Eddy and Reynolds-Averaged simulations of turbulent reacting flows. The lifted flame is simulated taking account of multi-step chemistry for dimethyl-ether fuel. Due to autoignition and flame propagation, the reaction progress increases with residence time. The variation of fluid residence time is evaluated by solving an Eulerian transport equation for the fluid age. The fluid age is a passive scalar with a spatially-uniform source term, meaning that its moments and dissipation rates in turbulent flows can be modelled using closures already established for conserved scalars such as mixture fraction. In combination with the mixture fraction, the fluid age serves as a useful mapping variable to distinguish younger less-reacted fluid near the inlet from older more-reacted fluid downstream. The local fluctuations of mixture fraction and fluid age have strong negative correlation and, building upon established presumed-pdf models for mixture fraction, this feature can be used to construct an accurate presumed-pdf model for the joint mixture fraction/fluid age pdf. It is demonstrated that the double-conditional first-order moment closure combined with the proposed presumed model for the joint pdf of mixture fraction and fluid age gives accurate predictions for unconditional reaction rates - both for pre-ignition radical species produced by low-temperature processes upstream of the flame base, and for major species that are produced at the flame front.
A three-dimensional direct numerical simulation (DNS) is performed for a turbulent hydrogen-air flame, represented with detailed chemistry, stabilized in a model gas-turbine combustor. The combustor geometry consists of a mixing duct followed by a sudden expansion and a combustion chamber, which represents a geometrically simplified version of Ansaldo Energia's GT26/GT36 sequential combustor design. In this configuration, a very lean blend of hydrogen and vitiated air is prepared in the mixing duct and convected into the combustion chamber, where the residence time from the inlet of the mixing duct to the combustion chamber is designed to coincide with the ignition delay time of the mixture. The results show that when the flame is stabilized at its design position, combustion occurs due to both autoignition and flame propagation (deflagration) modes at different locations within the combustion chamber. A chemical explosive mode analysis (CEMA) reveals that most of the fuel is consumed due to autoignition in the bulk-flow along the centerline of the combustor, and lower amounts of fuel are consumed by flame propagation near the corners of the sudden expansion, where the unburnt temperature is reduced by the thermal wall boundary layers. An unstable operating condition is also identified, wherein periodic auto-ignition events occur within the mixing duct. These events appear upstream of the intended stabilization position, due to positive temperature fluctuations induced by pressure waves originating from within the combustion chamber. The present DNS investigation represents the initial step of a comprehensive research effort aimed at gaining detailed physical insight into the rate-limiting processes that govern the sequential combustor behavior and avoid the insurgence of the off-design auto-ignition events.
The ignition process in diesel engines is highly complex and incompletely understood. In the present study, two-dimensional direct numerical simulations are performed to investigate the ignition dynamics and their sensitivity to thermochemical and mixing parameters. The thermochemical and mixing conditions are matched to the benchmark Spray A experiment from the Engine Combustion Network. The results reveal a complex ignition process with overlapping stages of: low-temperature ignition (cool flames), rich premixed ignition, and nonpremixed ignition, which are qualitatively consistent with prior experimental and numerical investigations, however, this is the first time that fully-resolved simulations have been reported at the actual Spray A thermochemical condition. Parametric variations are then performed for the Damkohler number Da, oxidiser temperature, oxygen concentration, and peak mixture fraction (a measure of premixedness), to study their effect on the ignition dynamics. It is observed that with both increasing oxidiser temperature and decreasing oxygen concentration, that the cool flame moves to richer mixtures, the overlap in the ignition stages decreases, and the (nondimensional) time taken to reach a fully burning state increases. With increasing Da, the cool-flame speed is decreased due to lower mean mixing rates, which causes a delayed onset of high-temperature ignition. With increasing peak mixture fraction, the onset of each stage of ignition is not affected, but the overall duration of the ignition increases leading to a longer burn duration. Overall, the results suggest that turbulence-chemistry interactions play a significant role in determining the timing and location in composition space of the entire ignition process.
Laminar natural gas flames are investigated at engine-relevant thermochemical conditions where the ignition delay time τ is short due to very high ambient temperatures and pressures. At these conditions, it is not possible to measure or calculate well-defined values for the laminar flame speed sl, laminar flame thickness δl, and laminar flame time scale due to the explosive thermochemical state. The corresponding reference values, sR, δR, and that account for the effects of autoignition, are numerically estimated to investigate the enhancement of flame propagation, and the competition with autoignition that arises under nominally autoignitive conditions (characterised here by the number τ/τR). Large values of τ/τR indicate that autoignition is unimportant, values near or below unity indicate that flame propagation is not possible, and intermediate values indicate that a combination of both flame propagation and autoignition may be important, depending upon factors such as device geometry, turbulence, stratification, et cetera. The reference quantities are presented for a wide range of temperatures, equivalence ratios, pressures, and hydrogen concentrations, which includes conditions relevant to stationary gas turbine reheat burners and boosted spark ignition engines. It is demonstrated that the transition from flame propagation to autoignition is only dependent on residence time, when the results are non-dimensionalised by the reference values. The temporal evolution of the reference values are also reported for a modelled boosted SI engine. It is shown that the nominally autoignitive conditions enhance flame propagation, which may be an ameliorating factor for the onset of engine knock. The calculations are performed using a recently-developed, detailed 177 species mechanism for C0–C3 chemistry that is derived from theoretical chemistry and is suitable for a wide range of thermochemical conditions as it is not tuned or optimised for a particular operating condition.
Direct numerical simulations are performed to investigate the transient upstream flame propagation (flashback) through homogeneous and fuel-stratified hydrogen-air mixtures transported in fully developed turbulent channel flows. Results indicate that, for both cases, the flame maintains steady propagation against the bulk flow direction, and the global flame shape and the local flame characteristics are both affected by the occurrence of fuel stratification. Globally, the mean flame shape undergoes an abrupt change when the approaching reactants transition from an homogeneous to a stratified mixing configuration. A V-shaped flame surface, whose leading-edge is located in the near-wall region, characterizes the nonstratified, homogeneous mixture case, while a U-shaped flame surface, whose leading edge propagates upstream at the channel centerline, distinguishes the case with fuel stratification (fuel-lean in the near-wall region and fuel-rich away from the wall). The characteristic thickness, wrinkling, and displacement speed of the turbulent flame brush are subject to considerable changes across the channel due to the dependence of the turbulence and mixture properties on the distance from the channel walls. More specifically, the flame transitions from a moderately wrinkled, thin-flamelet combustion regime in the homogeneous mixture case to a strongly wrinkled flame brush more representative of a thickened-flame combustion regime in the near-wall region of the fuel-stratified case. The combustion regime may be related to the Karlovitz number, and it is shown that a nominal channel-flow Karlovitz number, Kainch, based on the wall-normal variation of canonical turbulence (tη=(ν/ϵ)1/2) and chemistry (tl=δl/Sl) timescales in fully developed channel flow, compares well with an effective Karlovitz number, Kaflch, extracted from the present DNS datasets using conditionally sampled values of tη and tl in the immediate vicinity of the flame (0.1
We present a direct numerical simulation of a temporal jet between n-dodecane and diluted air undergoing spontaneous ignition at conditions relevant to low-temperature diesel combustion. The jet thermochemical conditions were selected to result in two-stage ignition. Reaction rates were computed using a 35-species reduced mechanism which included both the low- and high-temperature reaction pathways. The aim of this study is to elucidate the mechanisms by which low-temperature reactions promote high-temperature ignition under turbulent, non-premixed conditions. We show that low-temperature heat release in slightly rich fuel regions initiates multiple cool flame kernels that propagate towards very rich fuel regions through a reaction-diffusion mechanism. Although low-temperature ignition is delayed by imperfect mixing, the propagation speed of the cool flames is high: as a consequence, high-temperature reactions in fuel-rich regions become active early during the ignition transient. Because of this early start, high-temperature ignition, which occurs in fuel-rich regions, is faster than homogeneous ignition. Following ignition, the high-temperature kernels expand and engulf the stoichiometric mixture-fraction iso-surface which in turn establish edge flames which propagate along the iso-surface. The present results indicate the preponderance of flame folding of existing burning surfaces, and that ignition due to edge-flame propagation is of lesser importance. Finally, a combustion mode analysis that extends an earlier classification [1] is proposed to conceptualize the multi-stage and multi-mode nature of diesel combustion and to provide a framework for reasoning about the effects of different ambient conditions on diesel combustion.
Direct numerical simulation (DNS) of a high Karlovitz number (Ka) CH4/air stratified premixed jet flame was performed and used to provide insights into fundamentals of turbulent stratified premixed flames and their modelling implications. The flame exhibits significant stratification where the central jet has an equivalence ratio of 0.4, which is surrounded by a pilot flame with an equivalence ratio of 0.9. A reduced chemical mechanism for CH4/air combustion based on GRI-Mech3.0 was used, including 268 elementary reactions and 28 transported species. Over five billion grid points were employed to adequately resolve the turbulence and flame scales. The maximum Ka of the flame in the domain approaches 1400, while the jet Damköhler number (Dajet) is as low as 0.0027. The flame shows early stages of CH4/air combustion in the near field and later stages in the far field; the separation of combustion stages can be largely attributed to the small jet flow timescale and the low Dajet. The gradient of equivalence ratio in the flame normal direction, dϕ/dn, is predominantly negative, and small-scale stratification was found to play an important role in determining the local flame structure. Notably, the flame is thinner, the burning is more intense, and the levels of the radical pools, including OH, O and H, are higher in regions with stronger mixture stratification. The local flame structure is more strained and less curved in these regions. The mean flame structure is considerably influenced by the strong shear, which can be reasonably predicted by unity Lewis number stratified premixed flamelets when the thermochemical conditions of the reactant and product are taken locally from the DNS and the strain rates close to those induced by the mean flow are used in the flamelet calculation. An enhanced secondary reaction zone behind the primary reaction zone was observed in the downstream region, where the temperature is high and the fuel concentration is negligible, consistent with the observed separation of combustion stages. The main reactions involved in the secondary reaction zone, including CO + OH⇔CO2 + H (R94), H + O2 + M⇔HO2 + M (R31), HO2 + OH⇔H2O + O2 (R82) and H2 + OH⇔H2O + H (R79), are related to accumulated intermediate species including CO, H2, H, and OH. The detailed mechanism of intermediate species accumulation was explored and its effect on chemical pathways and heat release rate was highlighted.
Exascale computing promises quantities of data too large to efficiently store and transfer across networks in order to be able to analyze and visualize the results. We investigate compressed sensing (CS) as an in situ method to reduce the size of the data as it is being generated during a large-scale simulation. CS works by sampling the data on the computational cluster within an alternative function space such as wavelet bases and then reconstructing back to the original space on visualization platforms. While much work has gone into exploring CS on structured datasets, such as image data, we investigate its usefulness for point clouds such as unstructured mesh datasets often found in finite element simulations. We sample using a technique that exhibits low coherence with tree wavelets found to be suitable for point clouds. We reconstruct using the stagewise orthogonal matching pursuit algorithm that we improved to facilitate automated use in batch jobs. We analyze the achievable compression ratios and the quality and accuracy of reconstructed results at each compression ratio. In the considered case studies, we are able to achieve compression ratios up to two orders of magnitude with reasonable reconstruction accuracy and minimal visual deterioration in the data. Our results suggest that, compared to other compression techniques, CS is attractive in cases where the compression overhead has to be minimized and where the reconstruction cost is not a significant concern.
Considering that simulations of turbulent combustion are computationally expensive, this chapter takes a decidedly different perspective, that of high-performance computing (HPC). The cost scaling arguments of non-reacting turbulence simulations are revisited and it is shown that the cost scaling for reacting flows is much more stringent for comparable conditions, making parallel computing and HPC indispensable. Hardware abstractions of typical parallel supercomputers are presented which show that for design of an efficient and optimal program, it is essential to exploit both distributed memory parallelism and shared-memory parallelism, i.e. hierarchical parallelism. Principles of efficient programming at various levels of parallelism are illustrated using archetypal code examples. The vast array of numerical methods, particularly schemes for spatial and temporal discretization, are examined in terms of tradeoffs they present from an HPC perspective. Aspects of data analytics that invariably result from large feature-rich data sets generated by combustion simulations are covered briefly.
This supplementary material complements the article and provides additional information to the chemical mechanism used in this work, boundary conditions for the LES con guration and table generation, comparisons of axial velocities, results from a LES/ nite-rate chemistry (FRC) approach, and results from the LES/DTF/SPF approach with a particular chemistry table that is generated using a single strained premixed amelet solution.
In order to achieve exascale systems, application resilience needs to be addressed. Some programming models, such as task-DAG (directed acyclic graphs) architectures, currently embed resilience features whereas traditional SPMD (single program, multiple data) and message-passing models do not. Since a large part of the community's code base follows the latter models, it is still required to take advantage of application characteristics to minimize the overheads of fault tolerance. To that end, this paper explores how recovering from hard process/node failures in a local manner is a natural approach for certain applications to obtain resilience at lower costs in faulty environments. In particular, this paper targets enabling online, semitransparent local recovery for stencil computations on current leadership-class systems as well as presents programming support and scalable runtime mechanisms. Also described and demonstrated in this paper is the effect of failure masking, which allows the effective reduction of impact on total time to solution due to multiple failures. Furthermore, we discuss, implement, and evaluate ghost region expansion and cell-to-rank remapping to increase the probability of failure masking. To conclude, this paper shows the integration of all aforementioned mechanisms with the S3D combustion simulation through an experimental demonstration (using the Titan system) of the ability to tolerate high failure rates (i.e., node failures every five seconds) with low overhead while sustaining performance at large scales. In addition, this demonstration also displays the failure masking probability increase resulting from the combination of both ghost region expansion and cell-to-rank remapping.
Obtaining multi-process hard failure resilience at the application level is a key challenge that must be overcome before the promise of exascale can be fully realized. Previous work has shown that online global recovery can dramatically reduce the overhead of failures when compared to the more traditional approach of terminating the job and restarting it from the last stored checkpoint. If online recovery is performed in a local manner further scalability is enabled, not only due to the intrinsic lower costs of recovering locally, but also due to derived effects when using some application types. In this paper we model one such effect, namely multiple failure masking, that manifests when running Stencil parallel computations on an environment when failures are recovered locally. First, the delay propagation shape of one or multiple failures recovered locally is modeled to enable several analyses of the probability of different levels of failure masking under certain Stencil application behaviors. Our results indicate that failure masking is an extremely desirable effect at scale which manifestation is more evident and beneficial as the machine size or the failure rate increase.
A three-dimensional direct numerical simulation is conducted for a temporally evolving planar jet of n-heptane at a pressure of 40 atmospheres and in a coflow of air at 1100 K. At these conditions, n-heptane exhibits a two-stage ignition due to low- and high-temperature chemistry, which is reproduced by the global chemical model used in this study. The results show that ignition occurs in several overlapping stages and multiple modes of combustion are present. Low-temperature chemistry precedes the formation of multiple spatially localised high-temperature chemistry autoignition events, referred to as 'kernels'. These kernels form within the shear layer and core of the jet at compositions with short homogeneous ignition delay times and in locations experiencing low scalar dissipation rates. An analysis of the kernel histories shows that the ignition delay time is correlated with the mixing rate history and that the ignition kernels tend to form in vortically dominated regions of the domain, as corroborated by an analysis of the topology of the velocity gradient tensor. Once ignited, the kernels grow rapidly and establish edge flames where they envelop the stoichiometric isosurface. A combination of kernel formation (autoignition) and the growth of existing burning surface (via edge-flame propagation) contributes to the overall ignition process. An analysis of propagation speeds evaluated on the burning surface suggests that although the edge-flame speed is promoted by the autoignitive conditions due to an increase in the local laminar flame speed, edge-flame propagation of existing burning surfaces (triggered initially by isolated autoignition kernels) is the dominant ignition mode in the present configuration.
The effect of differential molecular diffusion (DMD) in turbulent non-premixed flames is studied by examining two previously reported DNS of temporally evolving planar jet flames, one with CO/H2 as the fuel and the other with C2H4 as the fuel. The effect of DMD in the CO/H2 DNS flames in which H2 is part of fuel is found to behave similar to laminar flamelet, while in the C2H4 DNS flames in which H2 is not present in the fuel it is similar to laminar flamelet in early stages but becomes different from laminar flamelet later. The scaling of the effect of DMD with respect to the Reynolds number Re is investigated in the CO/H2 DNS flames, and an evident power law scaling (∼Re−a with a a positive constant) is observed. The scaling of the effect of DMD with respect to the Damköhler number Da is explored in both laminar counter-flow jet C2H4 diffusion flames and the C2H4 DNS flames. A power law scaling (∼Daa with a a positive constant) is clearly demonstrated for C2H4 nonpremixed flames.
This article reports an analysis of the first detailed chemistry direct numerical simulation (DNS) of a high Karlovitz number laboratory premixed flame. The DNS results are first compared with those from laser-based diagnostics with good agreement. The subsequent analysis focuses on a detailed investigation of the flame area, its local thickness and their rates of change in isosurface following reference frames, quantities that are intimately connected. The net flame stretch is demonstrated to be a small residual of large competing terms: The positive tangential strain term and the negative curvature stretch term. The latter is found to be driven by flame speed-curvature correlations and dominated in net by low probability highly curved regions. Flame thickening is demonstrated to be substantial on average, while local regions of flame thinning are also observed. The rate of change of the flame thickness (as measured by the scalar gradient magnitude) is demonstrated, analogously to flame stretch, to be a competition between straining tending to increase gradients and flame speed variations in the normal direction tending to decrease them. The flame stretch and flame thickness analyses are connected by the observation that high positive tangential strain rate regions generally correspond with low curvature regions; these regions tend to be positively stretched in net and are relatively thinner compared with other regions. High curvature magnitude regions (both positive and negative) generally correspond with lower tangential strain; these regions are in net negatively stretched and thickened substantially.
A two-dimensional direct numerical simulation of the ignition at diesel engine-relevant conditions was performed at 40 atm and at 900 K ambient temperature using dimethyl ether (DME) as the fuel with a 30 species reduced chemical mechanism. At these conditions similar to diesel fuel DME exhibited two-stage ignition. The low-temperature chemistry analysis revealed a "spotty" first-stage autoignition in lean regions which transitions to a diffusively supported cool-flame and then propagates up the local mixture fraction gradient towards richer regions. The cool-flame speed was much faster than can be attributed to spatial gradients in first-stage ignition delay time in homogeneous reactors and it caused a shortening of the second-stage ignition delay times compared to a homogeneous reactor wherein the shortening became more pronounced at richer mixtures. Multiple high-temperature ignition kernels were noted over a range of rich mixtures that are much richer than the homogeneous most reactive mixture and most kernels formed much earlier than suggested by the homogeneous ignition delay time of the corresponding local mixture.
The flame structure corresponding to lean hydrogen–air premixed flames in intense sheared turbulence in the thin reaction zone regime is quantified from flame thickness and conditional scalar dissipation rate statistics, obtained from recent direct numerical simulation data of premixed temporally-evolving turbulent slot jet flames [1]. It is found that, on average, these sheared turbulent flames are thinner than their corresponding planar laminar flames. Extensive analysis is performed to identify the reason for this counter-intuitive thinning effect. The factors controlling the flame thickness are analyzed through two different routes i.e., the kinematic route, and the transport and chemical kinetics route. The kinematic route is examined by comparing the statistics of the normal strain rate due to fluid motion with the statistics of the normal strain rate due to varying flame displacement speed or self-propagation. It is found that while the fluid normal straining is positive and tends to separate iso-scalar surfaces, the dominating normal strain rate due to self-propagation is negative and tends to bring the iso-scalar surfaces closer resulting in overall thinning of the flame. The transport and chemical kinetics route is examined by studying the non-unity Lewis number effect on the premixed flames. The effects from the kinematic route are found to couple with the transport and chemical kinetics route. In addition, the intermittency of the conditional scalar dissipation rate is also examined. It is found to exhibit a unique non-monotonicity of the exponent of the stretched exponential function, conventionally used to describe probability density function tails of such variables. The non-monotonicity is attributed to the detailed chemical structure of hydrogen-air flames in which heat release occurs close to the unburnt reactants at near free-stream temperatures.
In the present work, a direct numerical simulation (DNS) of an experimental high Karlovitz number (Ka) CH4/air piloted premixed flame was analyzed to study the inner structure and the stabilization mechanism of the turbulent flame. A reduced chemical mechanism for premixed CH4/air combustion with NOx based on GRI-Mech3.0 was used, including 268 elementary reactions and 28 transported species. The evolution of the stretch factor, I0, indicates that the burning rate per unit flame surface area is considerably reduced in the near field and exhibits a minimum at x/D = 8. Downstream, the burning rate gradually increases. The stretch factor is different between different species, suggesting the quenching of some reactions but not others. Comparison between the turbulent flame and strained laminar flames indicates that certain aspects of the mean flame structure can be represented surprisingly well by flamelets if changes in boundary conditions are accounted for and the strain rate of the mean flow is employed; however, the thickening of the flame due to turbulence is not captured. The spatial development of displacement speeds is studied at higher Ka than previous DNS. In contrast to almost all previous studies, the mean displacement speed conditioned on the flame front is negative in the near field, and the dominant contribution to the displacement speed is normal diffusion with the reaction contribution being secondary. Further downstream, reaction overtakes normal diffusion, contributing to a positive displacement speed. The negative displacement speed in the near field implies that the flame front situates itself in the pilot region where the inner structure of the turbulent flame is affected significantly, and the flame stabilizes in balance with the inward flow. Notably, in the upstream region of the turbulent flame, the main reaction contributing to the production of OH, H+O2⇔O+OH (R35), is weak. Moreover, oxidation reactions, H2+OH⇔H+H2O (R79) and CO+OH⇔CO2+H (R94), are influenced by H2O and CO2 from the pilot and are completely quenched. Hence, the entire radical pool of OH, H and O is affected. However, the fuel consumption layer remains comparably active and generates heat, mainly via the reaction CH4+OH⇔CH3+H2O (R93).
Transported probability density function (TPDF) methods are an attractive modeling approach for turbulent flames as chemical reactions appear in closed form. However, molecular micro-mixing needs to be modeled and this modeling is considered a primary challenge for TPDF methods. In the present study, a new algebraic mixing rate model for TPDF simulations of turbulent premixed flames is proposed, which is a key ingredient in commonly used molecular mixing models. The new model aims to properly account for the transition in reactive scalar mixing rate behavior from the limit of turbulence-dominated mixing to molecular mixing behavior in flamelets. An a priori assessment of the new model is performed using direct numerical simulation (DNS) data of a lean premixed hydrogen–air jet flame. The new model accurately captures the mixing timescale behavior in the DNS and is found to be a significant improvement over the commonly used constant mechanical-to-scalar mixing timescale ratio model. An a posteriori TPDF study is then performed using the same DNS data as a numerical test bed. The DNS provides the initial conditions and time-varying input quantities, including the mean velocity, turbulent diffusion coefficient, and modeled scalar mixing rate for the TPDF simulations, thus allowing an exclusive focus on the mixing model. The new mixing timescale model is compared with the constant mechanical-to-scalar mixing timescale ratio coupled with the Euclidean Minimum Spanning Tree (EMST) mixing model, as well as a laminar flamelet closure by Pope and Anand (1984). It is found that the laminar flamelet closure is unable to properly capture the mixing behavior in the thin reaction zones regime while the constant mechanical-to-scalar mixing timescale model under-predicts the flame speed. The EMST model coupled with the new mixing timescale model provides the best prediction of the flame structure and flame propagation among the models tested, as the dynamics of reactive scalar mixing across different flame regimes are appropriately accounted for.
The increasing complexity of both scientific simulations and high-performance computing system architectures are driving the need for adaptive workflows, in which the composition and execution of computational and data manipulation steps dynamically depend on the evolutionary state of the simulation itself. Consider, for example, the frequency of data storage. Critical phases of the simulation should be captured with high frequency and with high fidelity for postanalysis; however, we cannot afford to retain the same frequency for the full simulation due to the high cost of data movement. We can instead look for triggers, indicators that the simulation will be entering a critical phase, and adapt the workflow accordingly. In this paper, we present a methodology for detecting triggers and demonstrate its use in the context of direct numerical simulations of turbulent combustion using S3D. We show that chemical explosive mode analysis (CEMA) can be used to devise a noise-tolerant indicator for rapid increase in heat release. However, exhaustive computation of CEMA values dominates the total simulation, and thus is prohibitively expensive. To overcome this computational bottleneck, we propose a quantile sampling approach. Our sampling-based algorithm comes with provable error/confidence bounds, as a function of the number of samples. Most importantly, the number of samples is independent of the problem size, and thus our proposed sampling algorithm offers perfect scalability. Our experiments on homogeneous charge compression ignition and reactivity controlled compression ignition simulations show that the proposed method can detect rapid increases in heat release, and its computational overhead is negligible. Our results will be used to make dynamic workflow decisions regarding data storage and mesh resolution in future combustion simulations.
With the goal of providing a more detailed fundamental understanding of ignition processes in diesel engines, this study reports analysis of a direct numerical simulation (DNS) database. In the DNS, a pseudo turbulent mixing layer of dimethyl ether (DME) at 400 K and air at 900 K is simulated at a pressure of 40 atmospheres. At these conditions, DME exhibits a two-stage ignition and resides within the negative temperature coefficient (NTC) regime of ignition delay times, similar to diesel fuel. The analysis reveals a complex ignition process with several novel features. Autoignition occurs as a distributed, two-stage event. The high-temperature stage of ignition establishes edge flames that have a hybrid premixed/autoignition flame structure similar to that previously observed for lifted laminar flames at similar thermochemical conditions. A combustion mode analysis based on key radical species illustrates the multi-stage and multi-mode nature of the ignition process and highlights the substantial modelling challenge presented by diesel combustion.
Dissipation spectra of velocity and reactive scalars—temperature and fuel mass fraction—in turbulent premixed flames are studied using direct numerical simulation data of a temporally evolving lean hydrogen-air premixed planar jet (PTJ) flame and a statistically stationary planar lean methane-air (SP) flame. The equivalence ratio in both cases was 0.7, the pressure 1 atm while the unburned temperature was 700 K for the hydrogen-air PTJ case and 300 K for methane-air SP case, resulting in data sets with a density ratio of 3 and 5, respectively. The turbulent Reynolds numbers for the cases ranged from 200 to 428.4, the Damköhler number from 3.1 to 29.1, and the Karlovitz number from 0.1 to 4.5. The dissipation spectra collapse when normalized by the respective Favre-averaged dissipation rates. However, the normalized dissipation spectra in all the cases deviate noticeably from those predicted by classical scaling laws for constant-density turbulent flows and bear a clear influence of the chemical reactions on the dissipative range of the energy cascade.
This paper presents a statistical analysis of edge flames in a turbulent lifted flame using direct numerical simulation (DNS). To investigate the dynamics of edge flames, a theoretical framework describing the edge-flame propagation velocity as a function of propagation velocities of mixture-fraction and product-mass fraction iso-surfaces at the flame base is used. The correlations between these propagation velocities and several other variables are then studied, including iso-surface curvatures, iso-surface orientations, strain rates, scalar dissipation rate and gradients of product mass fraction. The contribution of these parameters to the overall behaviour of the edge flame is also investigated using conditional averaging on two-dimensional spatial locations at the flame base. The analysis reveals that the tangential and normal strain rates in addition to the curvatures and scalar dissipation rates have significant contributions to the overall behaviour of the edge flame. The elliptical motion of the flame base described in our earlier study [1] is extended to provide a clearer picture of how these various parameters affect the large fluctuations of edge-flame velocity observed at the flame base.
A three-dimensional direct numerical simulation (DNS) of a turbulent lifted dimethyl ether (DME) slot jet flame was performed at elevated pressure to study interactions between chemical reactions with low-temperature heat release (LTHR), negative temperature coefficient (NTC) reactions and shear generated turbulence in a jet in a heated coflow. By conditioning on mixture fraction, local reaction zones and local heat release rate, the turbulent flame is revealed to exhibit a "pentabrachial" structure that was observed for a laminar DME lifted flame [Krisman et al., (2015)]. The propagation characteristics of the stabilization and triple points are also investigated. Potential stabilization points, spatial locations characterized by preferred temperature and mixture fraction conditions, exhibit autoignition characteristics with large reaction rate and negligible molecular diffusion. The actual stabilization point which coincides with the most upstream samples from the pool of potential stabilization points fovr each spanwise location shows passive flame structure with large diffusion. The propagation speed along the stoichiometric surface near the triple point is compared with the asymptotic value obtained from theory [Ruetsch et al., (1995)]. At stoichiometric conditions, the asymptotic and averaged DNS values of flame displacement speed deviate by a factor of 1.7. However, accounting for the effect of low-temperature species on the local flame speed increase, these two values become comparable. This suggests that the two-stage ignition influences the triple point propagation speed through enhancement of the laminar flame speed in a configuration where abundant low-temperature products from the first stage, low-temperature ignition are transported to the lifted flame by the high-velocity jet.
Zhao, Xin Y.; Bhagatwala, Ankit; Chen, Jacqueline H.; Haworth, Daniel C.; Pope, Stephen B.
In this study, the modeling of mixing by molecular diffusion is a central aspect for transported probability density function (tPDF) methods. In this paper, the newly-proposed shadow position mixing model (SPMM) is examined, using a DNS database for a temporally evolving di-methyl ether slot jet flame. Two methods that invoke different levels of approximation are proposed to extract the shadow displacement (equivalent to shadow position) from the DNS database. An approach for a priori analysis of the mixing-model performance is developed. The shadow displacement is highly correlated with both mixture fraction and velocity, and the peak correlation coefficient of the shadow displacement and mixture fraction is higher than that of the shadow displacement and velocity. This suggests that the composition-space localness is reasonably well enforced by the model, with appropriate choices of model constants. The conditional diffusion of mixture fraction and major species from DNS and from SPMM are then compared, using mixing rates that are derived by matching the mixture fraction scalar dissipation rates. Good qualitative agreement is found, for the prediction of the locations of zero and maximum/minimum conditional diffusion locations for mixture fraction and individual species. Similar comparisons are performed for DNS and the IECM (interaction by exchange with the conditional mean) model. The agreement between SPMM and DNS is better than that between IECM and DNS, in terms of conditional diffusion iso-contour similarities and global normalized residual levels. It is found that a suitable value for the model constant c that controls the mixing frequency can be derived using the local normalized scalar variance, and that the model constant a controls the localness of the model. A higher-Reynolds-number test case is anticipated to be more appropriate to evaluate the mixing models, and stand-alone transported PDF simulations are required to more fully enforce localness and to assess model performance.
The diffusive-thermal (D-T) instability of opposed nonpremixed tubular flames near extinction is investigated using two-dimensional (2-D) direct numerical simulations together with the linear stability analysis. Two different initial conditions (IC), i.e. the perturbed IC and the C-shaped IC are adopted to elucidate the effects of small and large amplitude disturbances on the formation of flame cells, similar to conditions found in linear stability analysis and experiments, respectively. The characteristics of the D-T instability of tubular flames are identified by a critical Damköhler number, DaC, at which the D-T instability first occurs and the corresponding number of flame cells for three different tubular flames with different flame radii. It is found that DaC predicted through linear stability analysis shows good agreement with that obtained from the 2-D simulations performed with two different ICs. The flame cell number, Ncell, from the 2-D simulations with the perturbed IC is also found to be equal to an integer close to the maximum wavenumber, kmax, obtained from the linear stability analysis. However, Ncell from the 2-D simulations with the C-shaped IC is smaller than kmax and Ncell found from the simulations with the perturbed IC. This is primarily because the strong reaction at the edges of the horseshoe-shaped cellular flame developed from the C-shaped IC is more likely to produce larger flame cells and reduce Ncell. It is also found that for cases with the C-shaped IC, once the cellular instability occurs, the number of flame cells remains constant until global extinction occurs by incomplete reaction manifested by small Da. It is also verified through the displacement speed, Sd, analysis that the two edges of the horseshoe-shaped cellular flame are stationary and therefore do not merge due to the diffusion-reaction balance at the edges. Moreover, large negative Sd is observed at the local extinction points while small positive or negative Sd features in the movement of flame cells as they adjust their location and size towards steady state.
This paper presents results from one and two-dimensional direct numerical simulations under Reactivity Controlled Compression Ignition (RCCI) conditions of a primary reference fuel (PRF) mixture consisting of n-heptane and iso-octane. RCCI uses in-cylinder blending of two fuels with different autoignition characteristics to control combustion phasing and the rate of heat release. These simulations employ an improved model of compression heating through mass source/sink terms developed in a previous work by Bhagatwala et al. (2014), which incorporates feedback from the flow to follow a predetermined experimental pressure trace. Two-dimensional simulations explored parametric variations with respect to temperature stratification, pressure profiles and n-heptane concentration. Statistics derived from analysis of diffusion/reaction balances locally normal to the flame surface were used to elucidate combustion characteristics for the different cases. Both deflagration and spontaneous ignition fronts were observed to co-exist, however it was found that higher n-heptane concentration provided a greater degree of flame propagation, whereas lower n-heptane concentration (higher fraction of iso-octane) resulted in more spontaneous ignition fronts. A significant finding was that simulations initialized with a uniform initial temperature and a stratified n-heptane concentration field, resulted in a large fraction of combustion occurring through flame propagation. It was also found that the proportion of spontaneous ignition fronts increased at higher pressures due to shorter ignition delay when other factors were held constant. For the same pressure and fuel concentration, the contribution of flame propagation to the overall combustion was found to depend on the level of thermal stratification, with higher initial temperature gradients resulting in more deflagration and lower gradients generating more ignition fronts. Statistics of ignition delay are computed to assess the Zel'dovich (1980) theory for the mode of combustion propagation based on ignition delay gradients.
Application resilience is a key challenge that has to be addressed to realize the exascale vision. Online recovery, even when it involves all processes, can dramatically reduce the overhead of failures as compared to the more traditional approach where the job is terminated and restarted from the last checkpoint. In this paper we explore how local recovery can be used for certain classes of applications to further reduce overheads due to resilience. Specifically we develop programming support and scalable runtime mechanisms to enable online and transparent local recovery for stencil-based parallel applications on current leadership class systems. We also show how multiple independent failures can be masked to effectively reduce the impact on the total time to solution. We integrate these mechanisms with the S3D combustion simulation, and experimentally demonstrate (using the Titan Cray-XK7 system at ORNL) the ability to tolerate high failure rates (i.e., node failures every 5 seconds) with low overhead while sustaining performance, at scales up to 262144 cores.
Proceedings of ISAV 2015: 1st International Workshop on In Situ Infrastructures for Enabling Extreme-Scale Analysis and Visualization, Held in conjunction with SC 2015: The International Conference for High Performance Computing, Networking, Storage and Analysis
Next generation architectures necessitate a shift away from traditional workflows in which the simulation state is saved at prescribed frequencies for post-processing analysis. While the need to shift to in situ workflows has been acknowledged for some time, much of the current research is focused on static workflows, where the analysis that would have been done as a post-process is performed concurrently with the simulation at user-prescribed frequencies. Recently, research efforts are striving to enable adaptive workflows, in which the frequency, composition, and execution of computational and data manipulation steps dynamically depend on the state of the simulation. Adapting the workflow to the state of simulation in such a data-driven fashion puts extremely strict efficiency requirements on the analysis capabilities that are used to identify the transitions in the workflow. In this paper we build upon earlier work on trigger detection using sublinear techniques to drive adaptive workflows. Here we propose a methodology to detect the time when sudden heat release occurs in simulations of turbulent combustion. Our proposed method provides an alternative metric that can be used along with our former metric to increase the robustness of trigger detection. We show the effectiveness of our metric empirically for predicting heat release for two use cases.
Computing distance fields is fundamental to many scientific and engineering applications. Distance fields can be used to direct analysis and reduce data. In this paper, we present a highly scalable method for computing 3D distance fields on massively parallel distributed-memory machines. A new distributed spatial data structure, named parallel distance tree, is introduced to manage the level sets of data and facilitate surface tracking over time, resulting in significantly reduced computation and communication costs for calculating the distance to the surface of interest from any spatial locations. Our method supports several data types and distance metrics from real-world applications. We demonstrate its efficiency and scalability on state-of-the-art supercomputers using both large-scale volume datasets and surface models. We also demonstrate in-situ distance field computation on dynamic turbulent flame surfaces for a petascale combustion simulation. Our work greatly extends the usability of distance fields for demanding applications.
Post-Moore's law scaling is creating a disruptive shift in simulation workflows, as saving the entirety of raw data to persistent storage becomes expensive. We are moving away from a post-process centric data analysis paradigm towards a concurrent analysis framework, in which raw simulation data is processed as it is computed. Algorithms must adapt to machines with extreme concurrency, low communication bandwidth, and high memory latency, while operating within the time constraints prescribed by the simulation. Furthermore, in- put parameters are often data dependent and cannot always be prescribed. The study of sublinear algorithms is a recent development in theoretical computer science and discrete mathematics that has significant potential to provide solutions for these challenges. The approaches of sublinear algorithms address the fundamental mathematical problem of understanding global features of a data set using limited resources. These theoretical ideas align with practical challenges of in-situ and in-transit computation where vast amounts of data must be processed under severe communication and memory constraints. This report details key advancements made in applying sublinear algorithms in-situ to identify features of interest and to enable adaptive workflows over the course of a three year LDRD. Prior to this LDRD, there was no precedent in applying sublinear techniques to large-scale, physics based simulations. This project has definitively demonstrated their efficacy at mitigating high performance computing challenges and highlighted the rich potential for follow-on re- search opportunities in this space.
Direct numerical simulations of three-dimensional spatially-developing turbulent Bunsen flames were performed at three different turbulence intensities. The simulations were performed using a reduced methane-air chemical mechanism which was specifically tailored for the lean premixed conditions simulated here. A planar-jet turbulent Bunsen flame configuration was used in which turbulent preheated methane-air mixture at 0.7 equivalence ratio issued through a central jet and was surrounded by a hot laminar coflow of burned products. The turbulence characteristics at the jet inflow were selected such that combustion occured in the thin reaction zones (TRZ) regime. At the lowest turbulence intensity, the conditions fall on the boundary between the TRZ regime and the corrugated flamelet regime, and progressively moved further into the TRZ regime by increasing the turbulent intensity. The data from the three simulations was analyzed to understand the effect of turbulent stirring on the flame structure and thickness. Statistical analysis of the data showed that the thermal preheat layer of the flame was thickened due to the action of turbulence, but the reaction zone was not significantly affected. A global and local analysis of the burning velocity of the flame was performed to compare the different flames. Detailed statistical averages of the flame speed were also obtained to study the spatial dependence of displacement speed and its correlation to strain rate and curvature.
The one-dimensional turbulence (ODT) model is applied to a reactant-to-product counterflow configuration and results are compared with DNS data. The model employed herein solves conservation equations for momentum, energy, and species on a one dimensional (1D) domain corresponding to the line spanning the domain between nozzle orifice centers. The effects of turbulent mixing are modeled via a stochastic process, while the Kolmogorov and reactive length and time scales are explicitly resolved and a detailed chemical kinetic mechanism is used. Comparisons between model and DNS results for spatial mean and root-mean-square (RMS) velocity, temperature, and major and minor species profiles are shown. The ODT approach shows qualitatively and quantitatively reasonable agreement with the DNS data. Scatter plots and statistics conditioned on temperature are also compared for heat release rate and all species. ODT is able to capture the range of results depicted by DNS. However, conditional statistics show signs of underignition.
A turbulent lifted slot-jet flame is studied using direct numerical simulation (DNS). A one-step chemistry model is employed with a mixture-fraction-dependent activation energy which can reproduce qualitatively the dependence of the laminar burning rate on the equivalence ratio that is typical of hydrocarbon fuels. The basic structure of the flame base is first examined and discussed in the context of earlier experimental studies of lifted flames. Several features previously observed in experiments are noted and clarified. Some other unobserved features are also noted. Comparison with previous DNS modelling of hydrogen flames reveals significant structural differences. The statistics of flow and relative edge-flame propagation velocity components conditioned on the leading edge locations are then examined. The results show that, on average, the streamwise flame propagation and streamwise flow balance, thus demonstrating that edge-flame propagation is the basic stabilisation mechanism. Fluctuations of the edge locations and net edge velocities are, however, significant. It is demonstrated that the edges tend to move in an essentially two-dimensional (2D) elliptical pattern (laterally outwards towards the oxidiser, then upstream, then inwards towards the fuel, then downstream again). It is proposed that this is due to the passage of large eddies, as outlined in Su et al. (Combust. Flame, vol. 144 (3), 2006, pp. 494-512). However, the mechanism is not entirely 2D, and out-of-plane motion is needed to explain how flames escape the high-velocity inner region of the jet. Finally, the time-averaged structure is examined. A budget of terms in the transport equation for the product mass fraction is used to understand the stabilisation from a time-averaged perspective. The result of this analysis is found to be consistent with the instantaneous perspective. The budget reveals a fundamentally 2D structure, involving transport in both the streamwise and transverse directions, as opposed to possible mechanisms involving a dominance of either one direction of transport. It features upstream transport balanced by entrainment into richer conditions, while on the rich side, upstream turbulent transport and entrainment from leaner conditions balance the streamwise convection.
Application resilience is a key challenge that must be ad-dressed in order to realize the exascale vision. Previous work has shown that online recovery, even when done in a global manner (i.e., involving all processes), can dramatically re-duce the overhead of failures when compared to the more traditional approach of terminating the job and restarting it from the last stored checkpoint. In this paper we suggest going one step further, and explore how local recovery can be used for certain classes of applications to reduce the over-heads due to failures. Specifically we study the feasibility of local recovery for stencil-based parallel applications and we show how multiple independent failures can be masked to effectively reduce the impact on the total time to solution.
This paper reports the results of a joint experimental and numerical study of the flow characteristics and flame structure of a hydrogen rich jet injected normal to a turbulent, vitiated crossflow of lean methane combustion products. Simultaneous high-speed stereoscopic PIV and OH PLIF measurements were obtained and analyzed alongside three-dimensional direct numerical simulations of inert and reacting JICF with detailed H2/CO chemistry. Both the experiment and the simulation reveal that, contrary to most previous studies of reacting JICF stabilized in low-to-moderate temperature air crossflow, the present conditions lead to a burner-attached flame that initiates uniformly around the burner edge. Significant asymmetry is observed, however, between the reaction zones located on the windward and leeward sides of the jet, due to the substantially different scalar dissipation rates. The windward reaction zone is much thinner in the near field, while also exhibiting significantly higher local and global heat release than the much broader reaction zone found on the leeward side of the jet. The unsteady dynamics of the windward shear layer, which largely control the important jet/crossflow mixing processes in that region, are explored in order to elucidate the important flow stability implications arising in the inert and reacting JICF. The paper concludes with an analysis of the ignition, flame characteristics, and global structure of the burner-attached flame. Chemical explosive mode analysis (CEMA) shows that the entire windward shear layer, and a large region on the leeward side of the jet, are highly explosive prior to ignition and are dominated by non-premixed flame structures after ignition. The predominantly mixing limited nature of the flow after ignition is examined by computing the Takeno flame index, which shows that ~70% of the heat release occurs in non-premixed regions.
Three-dimensional direct numerical simulation results of a transverse syngas fuel jet in turbulent cross-flow of air are analyzed to study the influence of varying volume fractions of CO relative to H2 in the fuel composition on the near field flame stabilization. The mean flame stabilizes at a similar location for CO-lean and CO-rich cases despite the trend suggested by their laminar flame speed, which is higher for the CO-lean condition. To identify local mixtures having favorable mixture conditions for flame stabilization, explosive zones are defined using a chemical explosive mode timescale. The explosive zones related to flame stabilization are located in relatively low velocity regions. The explosive zones are characterized by excess hydrogen transported solely by differential diffusion, in the absence of intense turbulent mixing or scalar dissipation rate. The conditional averages show that differential diffusion is negatively correlated with turbulent mixing. Moreover, the local turbulent Reynolds number is insufficient to estimate the magnitude of the differential diffusion effect. Alternatively, the Karlovitz number provides a better indicator of the importance of differential diffusion. A comparison of the variations of differential diffusion, turbulent mixing, heat release rate and probability of encountering explosive zones demonstrates that differential diffusion predominantly plays an important role for mixture preparation and initiation of chemical reactions, closely followed by intense chemical reactions sustained by sufficient downstream turbulent mixing. The mechanism by which differential diffusion contributes to mixture preparation is investigated using the Takeno Flame Index. The mean Flame Index, based on the combined fuel species, shows that the overall extent of premixing is not intense in the upstream regions. However, the Flame Index computed based on individual contribution of H2 or CO species reveals that hydrogen contributes significantly to premixing, particularly in explosive zones in the upstream leeward region, i.e. at the preferred flame stabilization location. Therefore, a small amount of H2 diffuses much faster than CO, creating relatively homogeneous mixture pockets depending on the competition with turbulent mixing. These pockets, together with high H2 reactivity, contribute to stabilizing the flame at a consistent location regardless of the CO concentration in the fuel for the present range of DNS conditions.
This paper presents the results of DNS of a partially premixed turbulent syngas/air flame at atmospheric pressure. The objective was to assess the importance and possible effects of molecular transport on flame behavior and structure. To this purpose DNS were performed at with two proprietary DNS codes and with three different molecular diffusion transport models: fully multi-component, mixture averaged, and imposing the Lewis number of all species to be unity. Results indicate that At the Reynolds numbers of the simulations (Returb = 600, Re = 8000) choice of molecular diffusion models affects significantly the temperature and concentration fields;Assuming Le = 1 for all species predicts temperatures up to 250 K higher than the physically realistic multi-component model;Faster molecular transport of lighter species changes the local concentration field and affects reaction pathways and chemical kinetics. A possible explanation for these observations is provided in terms of species diffusion velocity that is a strong function of gradients: thus, at sufficiently large Reynolds numbers, gradients and their effects tend to be large. The preliminary conclusion from these simulations seems to indicate molecular diffusion as the third important mechanism active in flames besides convective transport and kinetics. If confirmed by further DNS and measurements, molecular transport in high intensity turbulent flames will have to be realistically modeled to accurately predict emissions (gaseous and particulates) and other combustor performance metrics.
In this paper, two- and three-dimensional direct numerical simulations (DNS) of autoignition phenomena in stratified dimethyl-ether (DME)/air turbulent mixtures are performed. A reduced DME oxidation mechanism, which was obtained using rigorous mathematical reduction and stiffness removal procedure from a detailed DME mechanism with 55 species, is used in the present DNS. The reduced DME mechanism consists of 30 chemical species. This study investigates the fundamental aspects of turbulence-mixing-autoignition interaction occurring in homogeneous charge compression ignition (HCCI) engine environments. A homogeneous isotropic turbulence spectrum is used to initialize the velocity field in the domain. The computational configuration corresponds to a constant volume combustion vessel with inert mass source terms added to the governing equations to mimic the pressure rise due to piston motion, as present in practical engines. DME autoignition is found to be a complex three-staged process; each stage corresponds to a distinct chemical kinetic pathway. The distinct role of turbulence and reaction in generating scalar gradients and hence promoting molecular transport processes are investigated. By applying numerical diagnostic techniques, the different heat release modes present in the igniting mixture are identified. In particular, the contribution of homogeneous autoignition, spontaneous ignition front propagation, and premixed deflagration towards the total heat release are quantified.
Results are presented here from a three-dimensional direct numerical simulation of a temporally-evolving planar slot jet flame and experimental measurements within a spatially-evolving axisymmetric jet flame operating with DME (dimethyl ether, CH3OCH3) as the fuel. Both simulation and experiment are conducted at a Reynolds number of 13050. The Damköhler number, stoichiometric mixture fraction and fuel and oxidizer compositions also are matched between simulation and experiment. Simultaneous OH/CH2O PLIF imaging is performed experimentally to characterize the spatial structure of the turbulent DME flames. The simulation shows a fully burning flame initially, which undergoes partial extinction and subsequently, reignition. The scalar dissipation rate (χ) increases to a value much greater than that calculated from near-extinction strained laminar flames, leading to the observed local extinction. As the turbulence decays, the local values of χ decrease and the flame reignites. The reignition process appears to be strongly dependent on the local χ value, which is consistent with previous results for simpler fuels. Statistics of OH and CH2O are compared between simulation and experiment and found to agree. The applicability of OH/CH2O (formaldehyde) product imaging as a surrogate for peak heat release rate is investigated. The concentration product is found to predict peak heat release rate extremely well in the simulation data. When this product imaging is applied to the experimental data, a similar extinction/reignition pattern also is observed in the experiments as a function of axial position. A new 30-species reduced chemical mechanism for DME was also developed as part of this work.
This paper reports the results of a joint experimental and numerical study of the ow characteristics and flame stabilization of a hydrogen rich jet injected normal to a turbulent, vitiated cross ow of lean methane combustion products. Simultaneous high-speed stereoscopic PIV and OH PLIF measurements were obtained and analyzed alongside three-dimensional direct numerical simulations of inert and reacting JICF with detailed H2/CO chemistry. Both the experiment and the simulation reveal that, contrary to most previous studies of reacting JICF stabilized in low-to-moderate temperature air cross ow, the present conditions lead to an autoigniting, burner-attached flame that initiates uniformly around the burner edge. Significant asymmetry is observed, however, between the reaction zones located on the windward and leeward sides of the jet, due to the substantially different scalar dissipation rates. The windward reaction zone is much thinner in the near field, while also exhibiting significantly higher local and global heat release than the much broader reaction zone found on the leeward side of the jet. The unsteady dynamics of the windward shear layer, which largely control the important jet/cross flow mixing processes in that region, are explored in order to elucidate the important flow stability implications arising in the reacting JICF. Vorticity spectra extracted from the windward shear layer reveal that the reacting jet is globally unstable and features two high frequency peaks, including a fundamental mode whose Strouhal number of ~0.7 agrees well with previous non-reacting JICF stability studies. The paper concludes with an analysis of the ignition, ame stabilization, and global structure of the burner-attached flame. Chemical explosive mode analysis (CEMA) shows that the entire windward shear layer, and a large region on the leeward side of the jet, are highly explosive prior to ignition and are dominated by non-premixed flame structures after ignition. The predominantly mixing limited nature of the flow after ignition is confirmed by computing the Takeno flame index, which shows that ~70% of the heat release occurs in non-premixed regions.
We present hierarchical streamline bundles, a new approach to simplifying and visualizing 2D flow fields. Our method first densely seeds a flow field and produces a large number of streamlines that capture important flow features such as critical points. Then, we group spatially neighboring and geometrically similar streamlines to construct a hierarchy from which we extract streamline bundles at different levels of detail. Streamline bundles highlight multiscale flow features and patterns through a clustered yet non-cluttered display. This selective visualization strategy effectively accentuates visual foci and therefore is able to convey the desired insight into the flow fields. The hierarchical streamline bundles we have introduced offer a new way to characterize and visualize the flow structure and patterns in multiscale fashion. Streamline bundles highlight critical points clearly and concisely. Exploring the hierarchy allows a complete visualization of important flow features. Thanks to selective streamline display and flexible LOD refinement, our multiresolution technique is scalable and is promising for viewing large and complex flow fields. In the future, we would like to seek a cost-effective way to generate streamlines without enforcing the dense seeding condition. We will also extend this approach to handle real-world 3D complex flow fields.
Combined experimental and numerical studies of the transient response of ignition to strained flows require a well-characterized ignition trigger. Laser deposition of a small radical pool provides a reliable method for initiating ignition of mixtures that are near the ignition limit. Two-dimensional direct numerical simulations are used to quantify the sensitivity of ignition kernel formation and subsequent edge-flame propagation to the oxidizer temperature and the initial width and amplitude of O-atom deposition used to trigger ignition in an axisymmetric counterflow of heated air versus ambient hydrogen/nitrogen. The ignition delay and super-equilibrium OH concentration in the nascent ignition kernel are highly sensitive to variations in these initial conditions. The ignition delay decreases as the amplitude of the initial O-atom deposition increases. The spatial distribution and the magnitude of the OH overshoot are governed by multi-dimensional effects. The degree of OH overshoot near the burner centerline increases as the diameter of the initial O-atom deposition region decreases. This result is attributed to preferential diffusion of hydrogen in the highly curved leading portion of the edge flame that is established following thermal runaway. The edge-flame speed and OH overshoot at the leading edge of the edge flame are relatively insensitive to variations in the initial conditions of the ignition. The steady edge-flame speed is approximately twice the corresponding laminar flame speed. The rate at which the edge flame approaches its steady state is insensitive to the initial conditions and depends solely on the diffusion time scale at the edge flame. The edge flame is curved toward the heated oxidizer stream as a result of differences in the chemical kinetics between the leading edge and the trailing diffusion flame. The structure of the highly diluted diffusion flame considered in this study corresponds to Liñán's 'premixed flame regime' in which only the oxidizer leaks through the reaction zone such that the flame is located at fuel lean rather than stoichiometric mixture fraction conditions.
The catalytic effect of nitric oxide (NO) on the dynamics of extinction and re-ignition of a vortex-perturbed non-premixed hydrogen-air flame is studied in a counterflow burner. A diffusion flame is established with counterflowing streams of nitrogen-diluted hydrogen at ambient temperature and air heated to a range of temperatures that brackets the auto-ignition temperature. Localized extinction is induced by impulsively driving a fuel-side toroidal vortex into the steady flame, and the recovery of the extinguished region is monitored by planar laser-induced fluorescence (PLIF) of the hydroxyl radical (OH). The dynamics of flame recovery depend on the air temperature and fuel concentration, and four different recovery modes are identified. These modes involve combinations of edge-flame propagation and the expansion of an auto-ignition kernel that forms within the extinguished region. The addition of a small amount of NO significantly alters the re-ignition process by shifting the balance between chain-termination and chain-propagation reactions to enhance auto-ignition. The ignition enhancement by this catalytic effect causes a shift in the conditions that govern the recovery modes. In addition, the effects of NO concentration and vortex strength on the flame recovery are examined. Direct numerical simulations of the flame-vortex interaction with and without NO doping show how the small amount of OH produced by NO-catalyzed reactions has a significant impact on the development of an auto-ignition kernel. This joint experimental and numerical study provides detailed insight into the interaction between transient flows and ignition processes.
Direct numerical simulation of a three-dimensional spatially developing turbulent slot-burner Bunsen flame has been performed with a new reduced methane-air mechanism. The mechanism, derived from sequential application of directed relation graph theory, sensitivity analysis and computational singular perturbation over the GRI-1.2 detailed mechanism is non-stiff and tailored to the lean conditions of the DNS. The simulation is performed for three flow through times, long enough to achieve statistical stationarity. The turbulence parameters have been chosen such that the combustion occurs in the thin reaction zones regime of premixed combustion. The data is analyzed to study possible influences of turbulence on the structure of the preheat and reaction zones. The results show that the mean thickness of the turbulent flame, based on progress variable gradient, is greater than the corresponding laminar flame. The effects of flow straining and flame front curvature on the mean flame thickness are quantified through conditional means of the thickness and by examining the balance equation for the evolution of the flame thickness. Finally, conditional mean reaction rate of key species compared to the laminar reaction rate profiles show that there is no significant perturbation of the heat release layer.
Three-dimensional direct numerical simulation of a spatially developing slot-burner Bunsen flame has been performed. The simulation is aimed at better understanding the dynamics of turbulent premixed flames in the thin reaction zones regime. A reduced chemical model for methane-air chemistry consisting of 13 resolved species, 4 quasi-steady state species and 73 elementary reactions has been developed specifically for the current simulation. Using the new chemical model a lean premixed methane-air flame at preheated conditions and ambient pressure is simulated. The simulation is performed long enough to achieve statistical stationarity. The data is analyzed to study possible influences of turbulence on the flame thickness. The results show that the average flame thickness increases, in agreement with a few, although not unanimous, experimental results.
Power sources capable of supplying tens of watts are needed for a wide variety of applications including portable electronics, sensors, micro aerial vehicles, and mini-robotics systems. The utility of these devices is often limited by the energy and power density capabilities of batteries. A small combustion engine using liquid hydrocarbon fuel could potentially increase both power and energy density by an order of magnitude or more. This report describes initial development work on a meso-scale external combustion engine based on the Stirling cycle. Although other engine designs perform better at macro-scales, we believe the Stirling engine cycle is better suited to small-scale applications. The ideal Stirling cycle requires efficient heat transfer. Consequently, unlike other thermodynamic cycles, the high heat transfer rates that are inherent with miniature devices are an advantage for the Stirling cycle. Furthermore, since the Stirling engine uses external combustion, the combustor and engine can be scaled and optimized semi-independently. Continuous combustion minimizes issues with flame initiation and propagation. It also allows consideration of a variety of techniques to promote combustion that would be difficult in a miniature internal combustion engine. The project included design and fabrication of both the engine and the combustor. Two engine designs were developed. The first used a cylindrical piston design fabricated with conventional machining processes. The second design, based on the Wankel rotor geometry, was fabricated by through-mold electroforming of nickel in SU8 and LIGA micromolds. These technologies provided the requisite precision and tight tolerances needed for efficient micro-engine operation. Electroformed nickel is ideal for micro-engine applications because of its high strength and ductility. A rotary geometry was chosen because its planar geometry was more compatible with the fabrication process. SU8 lithography provided rapid prototypes to verify the design. A final high precision engine was created via LIGA. The micro-combustor was based on an excess enthalpy concept. Development of a micro-combustor included both modeling and experiments. We developed a suite of simulation tools both in support of the design of the prototype combustors, and to investigate more fundamental aspects of combustion at small scales. Issues of heat management and integration with the micro-scale Stirling engine were pursued using CFD simulations. We found that by choice of the operating conditions and channel dimensions energy conversion occurs by catalysis-dominated or catalysis-then-homogeneous phase combustion. The purpose of the experimental effort in micro-combustion was to study the feasibility and explore the design parameters of excess enthalpy combustors. The efforts were guided by the necessity for a practical device that could be implemented in a miniature power generator, or as a stand-alone device used for heat generation. Several devices were fabricated and successfully tested using methane as the fuel.