Turbulent Premixed Combustion Model Research Articles

Modeling of non-homogeneous premixed hydrogen-air flame acceleration and deflagration to detonation transition in an obstructed channel

This paper uses Large Eddy Simulation (LES) technology to study the dynamic behavior of flame acceleration (FA) and deflagration-to-detonation transition (DDT) in a duct with different obstacle layouts in non-homogeneous concentration fields. A turbulent premixed combustion model with three additional source terms was established using the OpenFoam code library. It was found that the predicted results of the coupled model were in good agreement with the measured values of the experiment. Numerical results show that turbulent combustion and self-ignition effects are the main factors affecting flame acceleration. Still, self-ignition effects are a necessary and sufficient condition for inducing DDT phenomena. Detonation flames have around-flow and diffraction effects and do not evolve into jet and vortex flames, which differ from deflagration flames. Two initiation mechanisms were found under different obstacle layouts: hotspots were generated in the area swept by the Mach stem at the flame front, triggering detonation near the turbulent flame through the reactive gradient mechanism; focusing of sufficiently strong reflected waves deposited enough energy in a particular area near the flame brush and directly triggered detonation. In addition, the central obstacle layout is most conducive to FA and DDT, while the symmetrical layout is the least favorable.

An improved TRF mechanism for a new turbulent premixed combustion model with application to engine combustion CFD

In the present work, an existing TRF (toluene reference fuel) chemical kinetic mechanism has been improved. In the improvement, the existing TRF mechanism was found to have unphysically higher laminar flame speeds at temperatures greater than 750 K which are typical in engine compression-combustion operating conditions. Eight dominant reactions which mainly control the laminar flame speeds under higher temperatures ( T > 750 K) have been identified. Based on a recently published power-law laminar flame speed correlation which is able to provide good predictions under a wide range of engine conditions, correct reaction rate constants for the eight high-temperature dominant reactions have been determined for physical laminar flame speeds. The identification and improvement of the eight high-temperature dominant reactions are the first novelty of this work. In order to simulate ethanol and gasoline blends, a latest ethanol sub-mechanism has been implemented into the TRF mechanism. The improved TRF mechanism was validated using available experimental data in the literature. Based on the improved TRF mechanism, a new mechanism library has been generated, which can be used for an improved mechanism-dynamic-selection turbulent premixed combustion model in which turbulent diffusivity is constructed as a function of local turbulence/thermodynamics conditions. The dynamic turbulent diffusivity sub-model is the second novelty of this work. The improved TRF mechanism and its combination with the improved mechanism-dynamic-selection turbulent premixed combustion model were successfully applied to the prediction of combustion and emissions of GTDI (gasoline turbocharged direct injection) engines under both warm-up and transient cold start operating conditions, indicating that the improved models in this work are beneficial for predictive engine combustion CFD (computational fluid dynamics).

Multiple Mapping Conditioning Mixing Time Scales for Turbulent Premixed Flames

A novel multiple mapping conditioning (MMC) mixing time scale model for turbulent premixed combustion has been developed. It combines time scales for the flamelet and distributed flame regimes with the aid of a blending function. The blending function serves two purposes. Firstly, it helps to identify zones where the premixed flame resides and where the time scale associated with the premixed flame shall be used. Secondly, it uses the Karlovitz number to identify the turbulent premixed combustion regime and to reduce the weighting of the premixed flame time scale if Karlovitz numbers are high and deviations from the flamelet regime are expected. A series of three-dimensional direct numerical simulations (DNS) of statistically one dimensional, freely propagating turbulent methane-air flames provides a wide range of turbulent combustion regimes for the mixing model validation. The new mixing time scale provides correct predictions of the flame speed of freely propagating turbulent flames which could not be matched by most recognized mixing models. The turbulent flame structure predicted by the new model is in good agreement with DNS for all combustion regimes from flamelet to the thickened reaction zone.

FlameFoam: An open source CFD solver for turbulent premixed combustion

Severe accidents in nuclear power plants may cause significant and long-lasting harm to the environment in the case of containment integrity loss. Numerical simulation of phenomena threatening containment integrity, e.g., hydrogen combustion and explosion, is essential for the effective severe accident management.International research projects and benchmarks are organized in order to promote improvement of simulation codes adequacy and accuracy in the severe accident relevant conditions. Results of the most recent benchmarks show that current state-of-the-art simulations were able to mostly reproduce the observable flame speed at the considered conditions, but there are still inadequacies in estimating velocity and its maximum value.Addressing a need for an accessible open source turbulent premixed combustion simulation tool, a new solver is being developed. An OpenFOAM version 7 based combustion solver called flameFoam has been built by implementing a progress variable approach and turbulent flame closure model for premixed combustion simulation. Turbulent flame speed is defined as a function of laminar flame speed and turbulence parameters, and is evaluated using one of three different correlations. Laminar flame speed can be set as a constant or calculated using a laminar flame speed correlation. The basis for the solver are standard compressible OpenFOAM solvers rhoPimpleFoam, buoyantPimpleFoam and chtMultiRegionFoam.The article presents initial public flameFoam version 0.8 and provides examples of its validation. Initial validation and verification of general solver features are represented by the simulations of shock tube and backward facing step cases. Solver capabilities in relation to the current turbulent premixed combustion modeling state-of-the-art are demonstrated with the simulations of ETSON-MITHYGENE benchmark experiments.It is shown that flameFoam version 0.8 is capable of adequate simulation of general flow features, turbulent compressible flow phenomena and turbulent premixed combustion of lean hydrogen-air mixtures at the considered conditions. Solver code is publicly hosted and being developed further.

Development of a mechanism-dynamic-selection turbulent premixed combustion model with application to gasoline engine combustion and emissions simulation

In order to overcome the drawback (tuning reaction rate coefficients) of using a single mechanism in the chemical-kinetics-based turbulent premixed combustion modelling approach, a mechanism-dynamic-selection turbulent premixed combustion model has been developed, validated, and successfully applied to the combustion and emissions simulation of spark ignition gasoline engines. In this new model, based on the local thermal and turbulent conditions of any preheat cell, a certain chemical kinetic mechanism which can achieve the target ratio of turbulent flame speed to laminar flame speed is selected. To realise this goal, based on the laminar flame speed sensitivity analysis of an improved eTRF (ethanol toluene reference fuel) base chemical kinetic mechanism, a derived mechanism library has been generated. Based on enough experimental and DNS (direct numerical simulation) data of S T/S L, a mechanism selection database (mechanism as a function of Borghi-Peters coordinates) has been constructed and implemented into a CFD (computational fluid dynamics) code, also a classification scheme for identifying unburned, preheat, flame front, and burned zones of a combustion domain has been developed and implemented into the CFD code. The developed mechanism-dynamic-selection turbulent premixed combustion model has been validated using the combustion in a spherical constant volume and applied to the prediction of combustion and emissions of a spark ignition gasoline engine under transient cold start process. Validation results show that, without tuning any model parameters, the base mechanism, the derived mechanism library, the mechanism selection function, the classification scheme, as well as a simulation setup strategy for modelling engine transient cold start process, have good prediction performances and are practical for engine combustion and emissions simulation.

The Effects of Turbulent Burning Velocity Models in a Swirl-Stabilized Lean Premixed Combustor

Abstract The effects of turbulent burning velocities in a turbulent premixed combustion simulation with a G-equation are investigated using the 3D LES technique. Two turbulent burning velocity models – Kobayashi model, which takes into account the burning velocity pressure effect, and the Pitsch model, which considers the flame regions on the premixed flame structure – are implemented. An LM6000 combustor is employed to validate the turbulent premixed combustion model. The results show that the flame structures in front of the injector have different shapes in each model because of the different turbulent burning velocities. These different flame structures induce changes in the entire combustor flow field, including in the recirculation zone. The dynamic mode decomposition (DMD) method and linear acoustic analysis provide the dominant acoustic mode.

Multiple mapping conditioning coupled with an artificially thickened flame model for turbulent premixed combustion

A hybrid Euler/Lagrange approach is introduced for the simulation of turbulent stratified flames. Large eddy simulations (LES) are used for the simulation of the flow field while artificial thickening of the flame provides sufficient resolution for the computation of the evolution of the filtered reaction progress variable. This model is complemented by a sparse Lagrangian particle method that provides instantaneous and local solutions of the species composition and can account for deviations from the flamelet-structure due to turbulence. The combined approach provides a model applicable to different premixed flame regimes including the corrugated and thickened flame regimes. The particle mixing model is based on a multiple mapping conditioning (MMC) approach that conditions mixing on a reference field (the reaction progress variable). Thus, the model ensures localness of mixing in composition space and prevents unphysical mixing of unburnt fluid with burnt fluid across the flame front. The MMC-LES results show good agreement with experimental data, and flamelet-like structures as well as deviations thereof can be predicted. The results are rather insensitive towards the MMC specific modelling parameters but the modelling of the mixing time scale needs to be adapted to achieve consistency between the flame propagation speed predicted by the artificially thickened flame model and the flame dynamics predicted by MMC.

Numerical simulation of premixed combustion using the modified dynamic thickened flame model coupled with multi-step reaction mechanism

Thickened flame (TF) model is one of the effective methods to resolve the flame front in turbulent premixed combustion modeling. The multi-step reaction mechanism is becoming increasingly important for combustion simulations such as pollutant formation, ignition and extinction. The effect of TF model on flame structures when coupling with multi-step reaction mechanism was investigated. The simulation results show that, no matter in laminar or turbulent condition, the global TF model coupling with multi-step reaction mechanism results in an incomplete combustion, which is mainly due to the enhanced species diffusion. Although Durand and Polifke’s dynamic thickened flame (DTF) sensor performs well for predicting laminar flame structure when coupling with multi-step reaction mechanism, it underestimates the effective thickening factor. In turbulent premixed flame simulation, the underestimated thickening factor leads to a faster local fuel consumption speed because of the over-predicted sub-grid flame wrinkling factor. A modified DTF sensor suitable for multi-step reaction mechanism is proposed. This sensor using the hyperbolic tangent function of progress variable to calculate thickening factor dynamically. It ensures that both the preheated and reaction zones are thickened effectively. The sub-grid wrinkling factor is hence estimated corresponding to the calculated flame thickness. Results of 1D laminar and 3D turbulent flame show that this method performs well for predicting both burned gas temperature and species concentration in burnt gas, which is important for predicting emissions.

Numerical Implementation and validation of turbulent premixed combustion model for lean mixtures

The present paper discusses the numerical investigation of turbulent premixed flames under lean conditions. Lean premixed combustion, a low NOx emission technique but are prone to instabilities, extinction and blow out. Such flames are influenced by preferential diffusion due to different mass diffusivities of reactants and difference between heat and mass diffusivities in the reaction zone. In this numerical study, we estimate non-reacting flow characteristics with implementation of an Algebraic Flame Surface Wrinkling Model (AFSW) in the open source CFD code OpenFOAM. In these flows, the mean velocity fields and recirculation zones were captured reasonably well by the RANS standard k-epsilon turbulence model. The simulated turbulent velocity is in good agreement with experiments in the shear-generated turbulence layer. The reacting flow study was done at three equivalence ratios of 0.43, 0.5 and 0.56 to gauge the ability of numerical model to predict combustion quantities. At equivalence ratios 0.5 and 0.56 the simulations showed numerical oscillations and non-convergence of the turbulent quantities. This leads to a detailed parametric variation study where, the pre-constant of AFSW model is varied with values 0.3, 0.35 and 0.4. However the study revealed the weak dependence of pre-constant value on the equivalence ratio. Hence the pre-constant value is fit for specific equivalence ratio based on the parametric variation study. The tuned AFSW model with fitted pre-constant specific to given equivalence ratio predicted are compared with experiments and discussed. The tuned AFSW model produced turbulent flame speed values which are good agreement with experiments.

Modelling turbulent premixed flame–wall interactions including flame quenching and near-wall turbulence based on a level-set flamelet approach

A level-set flamelet model for turbulent premixed combustion that accounts for the main effects of flame–wall interaction, quenching and near-wall turbulence, is proposed based on the G-equation. The structure of laminar unquenched and quenched flames is analysed and a consistent G-equation valid for both flame types is derived. For the modelling of the turbulent quenching process, it is argued that the state of individual flamelets can be described as either quenched or unquenched. This binary mechanism leads to a kinematic description of turbulent flame–wall interactions in which the fraction of unquenched flames is described by the unquenched factor Q. It is shown that Q allows for a general and appropriate scaling of the turbulent burning velocity due to the fact that only unquenched flamelets contribute to the overall propagation speed. For the modelling of turbulent premixed combustion, a unified G-equation valid for unquenched and quenched flames is derived. Modelling closures accounting for near-wall turbulence as well as unsteady flame development effects are introduced. The modelling approach is analysed a priori as well as a posteriori using a turbulent channel flow. With regard to the turbulent burning velocity, it is found that the effect of quenching is dominant compared to the effect of wall-bounded turbulence. The latter becomes important for a proper estimation of the turbulent flame length. Moreover, the model is compared against available DNS data of flame–wall interaction in a turbulent channel flow. It is shown that the turbulent burning velocity, the turbulent flame thickness as well as the reactive flame surface density near the wall are correctly reproduced. The present modelling approach thus allows for a consistent modelling of flame–wall interactions, which can also be transferred to other combustion models that are based on turbulent flame speed correlations.

A posteriori tests of a dynamic thickened flame model for large eddy simulations of turbulent premixed combustion

Dynamic models where model parameters are automatically adjusted from known resolved fields are a very attractive formulation for large eddy simulations. Now widely used for unresolved momentum transport, this approach remains rather marginal to describe filtered reaction rates despite of very promising results. Global and local dynamic formulations for the flame wrinkling factor are combined with the Thickened Flame (TFLES) model to simulate the F3 pilot stabilized jet flame studied experimentally by Chen and coworkers. The influence of physical (flame wrinkling inner cut-off length scale) and numerical (test filter width, averaging procedure, updating frequency) characteristics of a flame wrinkling factor dynamic model for turbulent premixed combustion is investigated. Numerical results are discussed in terms of mean flow fields as well as dynamical behaviors. It is shown that the dynamic model is robust and relatively insensitive to the numerical input coefficients to be provided beforehand in the code. This finding indicates that the model parameter does not need to be adjusted any more. However, a model for the inner cut-off scale of flame wrinkles, lost in the filtering process, is required.

Special Issue on Recent Advances in Turbulent Premixed Combustion Modeling

The importance of fuel-lean combustion in premixed mode for modern gas turbine and automotive engines has been well recognized. This combustion mode is prone to combustion-induced oscillation, whic...

Simulations of gasoline engine combustion and emissions using a chemical-kinetics-based turbulent premixed combustion modeling approach

This work presents a turbulent premixed combustion modeling approach which is based on chemical kinetics. In this approach, the smallest length scales are of the order of 0.1–1.0 mm for typical engine simulations with a Reynolds-averaged Navier–Stokes turbulence model and, after adaptive mesh refinement technology is used to consider the magnitude of the subgrid field, the Reynolds-averaged Navier–Stokes turbulent flow field can be well resolved. For solution of the flame front, an artificially thickened laminar flame concept is introduced to balance the computational accuracy and the computational cost. Around the artificially thickened laminar flame front, a special grid resolution strategy is designed, i.e. using much finer resolution in the normal direction of the flame front and typical adaptive mesh refinement resolution in the other two perpendicular directions. Then, chemical kinetics can be applied to the chemistry process which occurs in the flame front. To use this chemical-kinetics-based turbulent premixed combustion modeling approach better, a good chemical kinetics mechanism is very important. For this reason a practical primary-reference-fuel chemical kinetics mechanism is improved and validated in present work. The newly improved mechanism resolves several issues in the existing mechanisms, including unrealistically fast autoignition reactions and limited laminar flame speed validation. After reoptimization of those laminar-flame-speed-related reactions, the new mechanism can correctly compute the laminar flame speeds for a wide range of Ford spark ignition engines and for various operating conditions. Using this combustion modeling approach together with the new mechanism, simulations of the combustion and the emissions of several spark ignition engines for typical operating conditions were carried out. The simulated in-cylinder pressures, the simulated burn rates, and the simulated emissions including the brake specific carbon monoxide emissions, the nitrogen oxide emissions, and the unburned hydrocarbon emissions are compared with the experimental data, and very good agreement is found without tuning any model constants.

Joint PDF Closure of Turbulent Premixed Flames

In this paper, a novel model for turbulent premixed combustion in the corrugated flamelet regime is presented, which is based on transporting a joint probability density function (PDF) of velocity, turbulence frequency and a scalar vector. Due to the high dimensionality of the corresponding sample space, the PDF equation is solved with a Monte-Carlo method, where individual fluid elements are represented by computational particles. Unlike in most other PDF methods, the source term not only describes reaction rates, but accounts for “ignition” of reactive unburnt fluid elements due to propagating embedded quasi laminar flames within a turbulent flame brush. Unperturbed embedded flame structures and a constant laminar flame speed (as expected in the corrugated flamelet regime) are assumed. The probability for an individual particle to “ignite” during a time step is calculated based on an estimate of the mean flame surface density (FSD), latter gets transported by the PDF method. Whereas this model concept has recently been published [21], here, a new model to account for local production and dissipation of the FSD is proposed. The following particle properties are introduced: a flag indicating whether a particle represents the unburnt mixture; a flame residence time, which allows to resolve the embedded quasi laminar flame structure; and a flag indicating whether the flame residence time lies within a specified range. Latter is used to transport the FSD, but to account for flame stretching, curvature effects, collapse and cusp formation, a mixing model for the residence time is employed. The same mixing model also accounts for molecular mixing of the products with a co-flow. To validate the proposed PDF model, simulation results of three piloted methane-air Bunsen flames are compared with experimental data and very good agreement is observed.

Systematic method of applying ANN for chemical kinetics reduction in turbulent premixed combustion modeling

A novel method to apply artificial neural network (ANN) for both chemical kinetics reduction and source term evaluation is introduced and tested in direct numerical simulation (DNS) and large eddy simulation (LES) of reactive flows. To gather turbulence affected flame data for ANN training, a new computation-economical method, called 1D pseudo-velocity disturbed flame (PVDF), is developed and used to generate thermo-chemical states independent of the modeled flame. Then a back-propagation ANN is trained using scaled conjugate gradient algorithm to memorize the sample states with reduced orders. The new method is employed in DNS and LES modeling of H2/air and C3H8/air premixed flames experiencing various levels of turbulence. The test result shows that compared to traditional computation with full mechanism and direct integration, this method can obtain quite large speed-ups with adequate prediction accuracy.

An EDC-based turbulent premixed combustion model

In the present study, a new turbulent premixed combustion model is proposed by integrating the Coherent Flame Model with the modified eddy dissipation concept, and relating the fine structure mass fraction to the flame surface density. First, experimental results of turbulent flame speed available from literature are compared with the predicted results at different turbulence intensities to validate the flame surface density model. It is observed that the model is able to predict the turbulent burning speeds accurately. Then, a comprehensive validation is carried out utilizing data on a turbulent lifted methane flame issuing into a vitiated co-flow. Detailed comparison of temperature and species concentrations between experiment and simulation is performed at different heights of the flame. Overall, the model is found to predict both the spatial variation and peak values of the scalars at various heights satisfactorily.

Topological analysis of methane lean premixed flame in turbulent jet flow by direct numerical simulation

Direct numerical simulation is employed to investigate the premixed jet flame of methane in lean, combined with a detailed chemical kinetics including 17 species and 58 elemental steps and distinct Lewis numbers. Cold methane-air mixture at 0.55 equivalence ratio is injected into the coflow area with 9500 Reynolds number. The coflow ambient gas is set to be the burnt gas of the methane-air mixture in main jet and temperature is assigned to be the corresponding adiabatic flame temperature 1515 K. The whole simulation process is run based on paralleled calculation method, and chemical sources are acquired by dynamically calling CHEMKIN library function. The flame structure at 6.98 ms is shown to demonstrate the obvious dependence of flame structure and reaction rate on vortices motion. Then by tracing the interaction between different flame elements and vortices, the mechanism behind this dependence is found. The concepts including surface density function, mean curvature and stretch rate are used to describe the geometrical structure of vortex-affected flame. Those are all calculated according to simulation result and statistically associated with heat release rate that indicates reaction intensity. The statistical results show that surface density function and reaction rate own direct positive correlation; largely wrinkled flames with greater curvatures magnitude tend to become thick and weaken surface density function and reaction rate, appearing local extinction phenomena; large tangential stretch rates exerted on flame usually produce a local thinning effect in normal making reaction rate intensified. The DNS presented will supply beneficial references to better understanding of flame-vortex interaction and development of more accurate general model for turbulent premixed combustion.

Analysis of a strong mass-based flame stretch model for turbulent premixed combustion

In the present paper a theory describing effects of strong flame stretch on turbulent flame propagation [L. P. H. de Goey and J. H. M. ten Thije Boonkkamp, “A flamelet description of premixed laminar flames and the relation with flame stretch,” Combust. Flame 119, 253 (1999)] is extended to volume averaged quantities and validated with direct numerical simulation (DNS). The extended theory describes the fuel consumption rate in terms of subgrid scale contributions connected to propagation effects including strong flame stretch. In case there is no preferential diffusion present, it is predicted that the total consumption rate is not affected by local stretch at all. Then the total consumption is described by the unstretched mass burning rate multiplied with the flame surface density. DNSs of turbulent flame kernels have been carried out in order to support the results from the theory. The chemistry is described by application of the flamelet generated manifold technique. The strong stretch theory is shown to be valid up to realizations in the thin reaction zone regime by three independent methods. The local effects of stretch are described, evaluated, and interpreted. Locally the mass burning rate changes by fuel leakage tangential to the flame, but this has no integral effect. The method can be used for subgrid scale modeling of turbulent flame propagation.

Pollutant formation modelling in natural gas SI engines using a level set based flamelet model

The level set flamelet model for turbulent premixed combustion was recently extended to improve the modelling of the spark ignition process, to incorporate the influence of unsteady flame kernel development on turbulent flame propagation, to embrace flame—wall interaction, and to comprise pollutant prediction models. The formation of nitrogen oxides (NO x) is predicted using the extended Zeldovich mechanism. The concentrations of the involved intermediate species are determined using a chemical equilibrium assumption. The formation of engine-out unburnt hydrocarbons (UHCs) and carbon monoxides (COs) is directly linked to incomplete combustion due to flame quenching at the time the flame hits the wall. In this paper, the new model is used for the analysis of combustion in a homogeneous charged natural gas SI engine operating on lean conditions. The investigated engine operating (PAM) points differ in the applied equivalence ratio, ignition timing, and rate of exhaust gas recirculation. The influence of the load on the peak pressures, peak pressure locations, and pollutant formations is discussed in detail. A good agreement with experimental data was obtained among all engine operating points.

Embedding quasi laminar 1D flame profiles to model turbulent premixed combustion with a joint PDF method

AbstractA new model for turbulent premixed combustion is presented which is based on a joint velocity composition probability density function (JPDF) method. The key idea is a scale separation approach. The method combines the model by Bray, Moss and Libby [1] (BML) for premixed combustion with the flamelet approach for nonpremixed combustion. Here, a Lagrangian formulation of the BML model is considered. The progress variable used by the BML model becomes a computational particle property and its value is triggered by the arrival of the flame front at the particle's position. Similar as in the flamelet approach we assume that the smallest eddies are not small enough to disturb the reactive diffusive flame structure. To resolve the (embedded) quasi laminar flame structure, a flame residence time is introduced. With that residence time, the evolution of the particle composition, including enthalpy, can be determined from precomputed laminar 1D flames. The main challenge with this approach is to model the probability that an embedded flamefront arrives at the particle location, which is necessary to close the chemical source term. Numerical experiments of a turbulent premixed flame show good agreement with experimental data. (© 2008 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)