Numerical Simulation of Turbulent Diffusion Flames of a Biogas Enriched with Hydrogen

Any biogas produced by the anaerobic fermentation of organic materials has the advantage of being an environmentally friendly biofuel.Nevertheless,the relatively low calorific value of such gases makes their effective utilization in practical applications relatively difficult.The present study considers the addition of hydrogen as a potential solution to mitigate this issue.In particular,the properties of turbulent diffusion jet flames and the related pollutant emissions are investigated numerically for different operating pressures.The related numerical simulations are conducted by solving the RANS equations in the frame of the Reynolds Stress Model in combination with the flamelet approach.Radiation effects are also taken into account and the combustion kinetics are described via the GRI-Mech 3.0 reaction model.The considered hydrogen fuel enrichment spans the range from 0%to 50%in terms of volume.Pressure varies between 1 and 10 atm.The results show that both hydrogen addition and pressure increase lead to an improvement in terms of mixing quality and have a significant effect on flame temperature and height.They also reduce CO_(2) emissions but increase NOx production.Prompt NO is shown to be the predominant NO formation mechanism.

CFD study of non-premixed swirling burners: Effect of turbulence models

This research investigates a numerical simulation of swirling turbulent non-premixed combustion.The effects on the combustion characteristics are examined with three turbulence models:namely as the Reynolds stress model,spectral turbulence analysis and Re-Normalization Group.In addition,the P-1 and discrete ordinate(DO)models are used to simulate the radiative heat transfer in this model.The governing equations associated with the required boundary conditions are solved using the numerical model.The accuracy of this model is validated with the published experimental data and the comparison elucidates that there is a reasonable agreement between the obtained values from this model and the corresponding experimental quantities.Among different models proposed in this research,the Reynolds stress model with the Probability Density Function(PDF)approach is more accurate(nearly up to 50%)than other turbulent models for a swirling flow field.Regarding the effect of radiative heat transfer model,it is observed that the discrete ordinate model is more precise than the P-1 model in anticipating the experimental behavior.This model is able to simulate the subcritical nature of the isothermal flow as well as the size and shape of the internal recirculation induced by the swirl due to combustion.

Application of WSGSA Model in Predicting Temperature and Soot in C_(2)H_(4)/Air Turbulent Diffusion Flame

Soot,a product of insufficient combustion,is usually in the form of aggregate. The multi-scattering of soot fractal aggregates has been proved to play an important role in studying the soot radiative properties,which is rarely considered in predicting the radiative heat transfer in combustion flame. In the present study,based on the weighted sum of gray soot fractal aggregate(WSGSA) model,which is used to predict the temperature field and soot aggregates in turbulent diffusion flame,the flame temperature distribution and soot volume fraction distribution under the conditions of the model without considering radiation,the default radiation model in Fluent software and the WSGSA model are calculated respectively. The results show that the flame temperature will be seriously overestimated without considering radiation and the maximum relative discrepancy of flame centerline temperature is about 64.5%. The accuracy will be improved by the default radiation model in the Fluent software,but the flame temperature is still overestimated and the maximum relative discrepancy of flame centerline temperature is about 42.1%. However,more satisfactory results can be obtained by the WSGSA model,and the maximum relative discrepancy of flame centerline temperature is no more than 15.3%. Similar conclusions can also be obtained in studying the temperature distribution along different flame heights. Moreover,the soot volume fraction can be predicted more accurately with the application of the WSGSA model. Both without considering radiation and using the default radiation model in the Fluent software will result in the underestimating of soot volume fraction. All the results reveal that the WSGSA model can be used to predict the temperature and soot aggregates in the CH/air turbulent diffusion flame.

Modelling of Turbulent Nonpremixed CH4/H2 Flame Using Second-Moment Turbulence Closure Models

Turbulent nonpremixed CH4/H2 flame has been simulated using several typical differential secondmoment turbulence closure (SMTC) models. To clarify the applicability of the various models, the LRR-IP model,JM model, SSG model as well as two modified LRR-IP models were tested. Some of above-mentioned SMTC models cannot provide the overall satisfactory predictions of this challenging case. It is confirmed again that the standard LRR-IP model considerably overpredict the centerline velocity decay rate, and therefore performs not well. Also it is interesting to observe that the JM model does not perform well in this challenging test case, although it has already been proved successful in other cases. The SSG model produces quite satisfactory prediction and performs equally well or better than the two modified LRR-IP models in the reacting case. It can be concluded that the modified LRR-IP models as well as the SSG model are superior to the other SMTC models in the turbulent nonpremixed CH4/H2 flame.

DNS Analysis on the Indirect Relationship between the Local Burning Velocity and the Flame Displacement Speed of Turbulent Premixed Flames

The local burning velocity and the flame displacement speed are the dominant properties in the mechanism of turbulent premixed combustion. The flame displacement speed and the local burning velocity have been investigated separately, because the flame displacement speed can be used for the discussion of flame-turbulence interactions and the local burning velocity can be used for the discussion of the inner structure of turbulent premixed flames. In this study, to establish the basis for the discussion on the effects of turbulence on the inner structure of turbulent premixed flames, the indirect relationship between the flame displacement speed and the local burning velocity was investigated by the flame stretch, the flame curvature, and the tangential strain rate using DNS database with different density ratios. It was found that for the local tangential strain rate and the local flame curvature, the local burning velocity and the flame displacement speed had the opposite correlations in each density ratio case. Therefore, it is considered that the local burning velocity and the flame displacement speed have a negative correlation.

Consideration on the Flow Velocity in the Experimental Analysis of the Flame Displacement Speed Using DNS Data of Turbulent Premixed Flames with Different Lewis Numbers

The flame displacement speed is one of the major characteristics in turbulent premixed flames. The flame displacement speed is experimentally obtained from the displacement normal to the flame surface, while it is numerically evaluated by the transport equation of the flame surface. The flame displacement speeds obtained both experimentally and numerically cannot be compared directly because their definitions are different. In this study, two kinds of experimental flame displacement speeds—involving the mean inflow velocity and the local flow velocity—were simulated using the DNS data with the different Lewis numbers, and were compared with the numerical flame displacement speed. The simulated experimental flame displacement speed involving the mean inflow velocity had no correlation with the numerical flame displacement speed, while the simulated displacement speed involving the local flow velocity had a clear correlation with the numerical displacement speed in the cases of higher Lewis number than unity. The correlation coefficient of the simulated displacement speed involving the local flow velocity with the numerical displacement speed had a maximum value on the isosurface of the reaction progress variable with the maximum temperature gradient where the dilation effect of the flame is strongest.

Evaluation of the Local Burning Velocity Using DNS Data of Turbulent Premixed Flames

The local burning velocity, which is based on the consumption rate of the unburned mixture, is one of the dominant parameters in turbulent premixed flames. In this study, the evaluating method of the local burning velocity was investigated using DNS data of turbulent premixed flames with different Lewis numbers. The local burning velocity was evaluated by integrating the chemical reaction rates along normal to the flame surface within three kinds of integration ranges that were defined as follows: the range which is defined by the half length of normal to the flame surface between its certain point and the other point crossing the flame surface (Range 1);the range which is defined by the reaction progress variable that the chemical reaction rate along normal to a planer flame surface takes a half of the maximum value (Range 2);the range which is defined by the length of normal to the flame surface between its certain point and the point which has the extreme value of the reaction progress variable (Range 3). As a result, Range 1 and Range 2 were affected by the flame shapes greatly, since the quantities of the integration ranges fluctuated widely dependent on the variations of turbulent premixed flames. Under the conditions of the turbulent combustion in this study, Range 3, which is hardly affected by a flame shape, is considered to be appropriate to the evaluation of the local burning velocity.

Comparative Appraisal of Response Surface Methodology and Artificial Neural Network Method for Stabilized Turbulent Confined Jet Diffusion Flames Using Bluff-Body Burners

The present study was conducted to present the comparative modeling, predictive and generalization abilities of response surface methodology (RSM) and artificial neural network (ANN) for the thermal structure of stabilized confined jet diffusion flames in the presence of different geometries of bluff-body burners. Two stabilizer disc burners tapered at 30° and 60° and another frustum cone of 60°/30° inclination angle were employed all having the same diameter of 80 (mm) acting as flame holders. The measured radial mean temperature profiles of the developed stabilized flames at different normalized axial distances (x/dj) were considered as the model example of the physical process. The RSM and ANN methods analyze the effect of the two operating parameters namely (r), the radial distance from the center line of the flame, and (x/dj) on the measured temperature of the flames, to find the predicted maximum temperature and the corresponding process variables. A three-layered Feed Forward Neural Network in conjugation with the hyperbolic tangent sigmoid (tansig) as transfer function and the optimized topology of 2:10:1 (input neurons: hidden neurons: output neurons) was developed. Also the ANN method has been employed to illustrate such effects in the three and two dimensions and shows the location of the predicted maximum temperature. The results indicated the superiority of ANN in the prediction capability as the ranges of R2 and F Ratio are 0.868 - 0.947 and 231.7 - 864.1 for RSM method compared to 0.964 - 0.987 and 2878.8 7580.7 for ANN method beside lower values for error analysis terms.

Investigation of Detailed Kinetic Scheme Performance on Modelling of Turbulent Non-Premixed Sooting Flames

This paper presents the results of an application of a first-order conditional moment closure （CMC） approach coupled with a semi-empirical soot model to investigate the effect of various detailed combustion chemistry schemes on soot formation and destruction in turbulent non-premixed flames. A two-equation soot model repre- senting soot particle nucleation, growth, coagulation and oxidation, was incorporated into the CMC model. The turbulent flow-field of both flames is described using the Favre-averaged fluid-flow equations, applying a stan- dard k-c turbulence model. A number of five reaction kinetic mechanisms having 50 - 100 species and 200 - 1000 elementary reactions called ABF, Miller-Bowman, GRI-Mech3.0, Warnatz, and Qin were employed to study the effect of combustion chemistry schemes on soot predictions. The results showed that of various kinetic schemes being studied, each yields similar accuracy in temperature prediction when compared with experimental data. With respect to soot prediction, the kinetic scheme containing benzene elementary reactions tends to result in a better prediction on soot concentrations in comparison to those contain no benzene elementary reactions. Among five kinetic mechanisms being studied, the Qin combustion scheme mechanism turned to yield the best prediction on both flame temperature and soot levels.

Non-premixed turbulent combustion modeling based on the filtered turbulent flamelet equation

In turbulent combustion simulations, the flow structure at the unresolved scale level needs to be reasonably modeled. Following the idea of turbulent flamelet equation for the non-premixed flame case, which was derived based on the filtered governing equations(L. Wang, Combust. Flame 175, 259(2017)), the scalar dissipation term for tabulation can be directly computed from the resolved flowing quantities, instead of solving species transport equations. Therefore, the challenging source term closure for the scalar dissipation or any assumed probability density functions can be avoided;meanwhile the chemical sources are closed by scaling relations. The general principles are discussed in the context of large eddy simulation with case validation. The new model predictions of the bluff-body flame show sufficiently improved results, compared with these from the classic progress-variable approach.

预混预蒸发正庚烷湍流火焰结构的实验研究

采用预混伴流射流燃烧器(PPJB)研究了初始温度400 K的预蒸发正庚烷/空气混合物在薄反应区的湍流火焰结构.采用热线风速仪标定冷态湍流流场,CH-PLIF测量湍流火焰反应区结构,共振OH-PLIF测量湍流火焰结构特征.结果表明:高湍流条件下,反应区厚度在不考虑三维效应时最大厚度为层流火焰厚度的1.6倍,其可支持薄反应区理论对预蒸发液体燃料湍流火焰的适用性.火焰面褶皱比和湍流燃烧速度的增加与湍流强度的提升不匹配.通过对火焰面曲率和平均火焰面密度的分析,发现高湍流条件下,褶皱尺度、大小和数量变化不大.一方面,湍流可能会减弱非均匀扩散造成的差异;另一方面,更可能是由于对褶皱结构的不完全捕捉和对3D效果的忽视而引发对火焰褶皱的低估.

Large Eddy Simulation Study on the Turbulence and Flame Characteristics under Analogical Integral Scale and Turbulence Intensity of Turbulent Premixed Flames

Bunsen burner is a typical geometry for investigating the turbulence-flame interaction.In most experimental studies,only turbulence intensity u′and integral scale l0 are used to characterize the turbulent flow field,regardless of the perforation geometry of perforated plates.However,since the geometry influences the developing process and vortex broken,the plate geometry has to be considered when discussing the flame-turbulence interaction.In order to investigate conditions at the same l0 and u′using different geometries,large eddy simulation of CH_(4)/air flames with dynamic TF combustion model was performed.The model validation shows good agreement between Large Eddy Simulation(LES)and experimental results.In the non-reacting flows,the Vortex Stretching of circular-perforated plate condition is always larger than that of slot-perforated plate condition,which comes from the stresses in the flow fields to stretch the vorticity vector.In reacting flows,at the root of the flame,the Vortex Stretching plays a major role,and the total vorticity here of circular-perforated plate condition is still larger(53.8%and 300%larger than that of the slot-perforated plate at x/D=0 and x/D=2.5,respectively).More small-scale vortex in circular-perforated plate condition can affect and wrinkle the flame front to increase the Probability Density Function(PDF)at large curvatures.The 3D curvature distributions of both cases bias to negative values.The negative trend of curvatures at the instant flame front results from the Dilatation term.Also,the value of the Vortex Stretching and the Dilatation at the flame front of circular-perforated plate condition is obviously larger.

Combining flamelet-generated manifold and machine learning models in simulation of a non-premixed diffusion flame

Flamelet Generated Manifold(FGM)is an example of a chemistry tabulation or a flamelet method that is under attention because of its accuracy and speed in predicting combustion characteristics.However,the main problem in applying the model is a large amount of memory required.One way to solve this problem is to apply machine learning(ML)to replace the stored tabulated data.Four different machine learning methods,including two Artificial Neural Networks(ANNs),a Random Forest(RF),and a Gradient Boosted Trees(GBT),are trained,validated,and compared in terms of various performance measures.The progress variable source term and transport properties are replaced with the ML models.Particular attention was paid to the progress variable source term due to its high gradient and wide range of its value in the control variables space.Data preprocessing is shown to play an essential role in improving the performance of the models.Two ensemble models,namely RF and GBT,exhibit high training efficiency and acceptable accuracy.On the other hand,the ANN models have lower training errors and take longer to train.The four models are then combined with a one-dimensional combustion code to simulate a counterflow non-premixed diffusion flame in engine-relevant conditions.The predictions of the ML-FGM models are compared with detailed chemical simulations and the original FGM model for key combustion properties and representative species profiles.

Deep Neural Network-Based Generation of Planar CH Distribution through Flame Chemiluminescence in Premixed Turbulent Flame

Flame front structure is one of the most fundamental characteristics and, hence, vital for understanding combustion processes. Measuring flame front structure in turbulent flames usually needs laser-based diagnostic techniques, mostly planar laser-induced fluorescence (PLIF). The equipment of PLIF, burdened with lasers, is often too sophisticated to be configured in harsh environments. Here, to shed the burden, we propose a deep neural network-based method to generate the structures of flame fronts using line-of-sight CH* chemiluminescence that can be obtained without the use of lasers. A conditional generative adversarial network (CGAN) was trained by simultaneously recording CH-PLIF and chemiluminescence images of turbulent premixed methane/air flames. Two distinct generators of the C-GAN, namely Resnet and U-net, were evaluated. The former net performs better in this study in terms of both generating snap-shot images and statistics over multiple images. For chemiluminescence imaging, the selection of the camera’s gate width produces a trade-off between the signal-to-noise (SNR) ratio and the temporal resolution. The trained C-GAN model can generate CH-PLIF images from the chemiluminescence images with an accuracy of over 91% at a Reynolds number of 5000, and the flame surface density at a higher Reynolds number of 10,000 can also be effectively estimated by the model. This new method has the potential to achieve the flame characteristics without the use of laser and significantly simplify the diagnosing system, also with the potential for high-speed flame diagnostics.

Temporally resolving premixed turbulent flame structures using self-supervised adversarial reconstruction of CH-PLIF

Understanding the turbulence-flame interaction is crucial to model the low-emission combustors developed for energy and propulsion applications. To this end, a novel frame interpolation (FI) method is proposed to better resolve the spatiotemporal evolution of premixed turbulent flame structures. The framework is completely selfsupervised, agnostic to optical flow, and driven by leveraging transferrable feature knowledge at lower speeds and adversarial learning to statistically map the flame dynamics across frames. The method is successfully applied on a 10 kHz CH planar laser-induced fluorescence (PLIF) dataset of highly wrinkled premixed flames with turbulent Reynolds numbers (ReT ) of 1100, 1400, and 7900, by down-sampling the image sequence to 5 kHz and restoring the sequence back to 10 kHz via FI. All reconstructions recovered important flame events and displayed excellent resemblance of the corrugated CH-layer geometries to that of the ground truths, with average intersection over union (IoU) and structural similarity index (SSIM) scores of 0.49 and 0.82, which are above the high-similarity baselines of 0.36 and 0.75, respectively. The wrinkling parameters (WP) of the flames also matched the ground truths, wherein R2 was roughly 0.95 for ReT = 1100 and 1400 and 0.85 for ReT = 7900 (lower due to the turbulence-induced uncertainties). The FI is further iteratively repeated to 40 kHz on the ReT = 7900 flames to facilitate pocket analysis by confidently linking their origin of formation, thus, enabling distinction from 3D tunnels, and improving statistical characterization of their consumption speeds. Given that the object features do not exhibit highly turbulent motions with regard to the initial time step, the proposed FI method is shown to be highly accurate and useful to analyzing finite-resolution experimental image sets including, but not restricted to, CH-PLIF.

Optical investigations on lean combustion improvement of natural gas engines via turbulence enhancement

In the global background of“Carbon Peak”and“Carbon Neutral”,natural gas engines show great advantages in energy-saving and pollution reduction.However,natural gas engines suffer from the issues of combustion instabilities when operating under lean burning conditions.In this paper,the role of turbulence enhancement in improving the lean combustion of natural gas was investigated in an optical SI engine with high compression ratios.Variable swirl control valves(SCV)were designed and intake tumble and swirl were combined to regulate turbulent motion and turbulent intensity.Particle image velocimetry was employed to measure in-cylinder turbulence,and transient pressure acquisition and high-speed photography were synchronously performed to quantify combustion evolutions.The results show that incylinder turbulent intensity is enhanced significantly through reducing SCV closing angles.Such that flame propagation speed and thermal efficiency are significantly improved with an increment of turbulent intensity,which indicated that mean effective pressures are not sensitive to spark timing.The analysis of flame images shows that the combined turbulence increases in the radial orientation from the spark plug to the cylinder wall,leading to an earlier flame kernel formation and a faster burning rate.Therefore,the combined turbulence has the potential in reducing the cyclic variations of lean combustion in natural gas engines.

Analysis of Turbulence and Surface Growth Models on the Estimation of Soot Level in Ethylene Non-Premixed Flames

Soot prediction in a combustion system has become a subject of attention, as many factors influence its accuracy. An accurate temperature prediction will likely yield better soot predictions, since the inception, growth and de- struction of the soot are affected by the temperature. This paper reported the study on the influences of turbulence closure and surface growth models on the prediction of soot levels in turbulent flames. The results demonstrated that a substantial distinction was observed in terms of temperature predictions derived using the k-c and the Rey- nolds stress models, for the two ethylene flames studied here amongst the four types of surface growth rate model investigated, the assumption of the soot surface growth rate proportional to the particle number density, but inde- pendent on the surface area of soot particles,f（As） = pNs, yields in closest agreement with the radial data. Without any adjustment to the constants in the surface growth term, other approaches where the surface growth directly proportional to the surface area and square root of surface area, f （As） = As and f （A,） = √As, result in an un- der-prediction of soot volume fraction. These results suggest that predictions of soot volume fraction are sensitive to the modelling of surface growth.

IMPLEMENTATION OF THE CONDITIONAL MOMENT CLOSURE MODEL TO A NONPREMIXED METHANOLAIR FLAME STABILIZED ON A BLUFF-BODY

A turbulent nonpremixed flame of methanol air stabilized on a bluff body is simulated by the conditional moment closure(CMC) model. An elliptic, conservative formulation of the CMC is employed with the turbulent diffusion term modeled by the turbulent eddy viscosity. Results of flow, conditional temperature and species concentration profiles with respect to mixture fraction, and unconditional temperature, OH and NO fields are obtained. Reasonable agreement between experiments and predictions to show that the CMC model is capable to predicting the species concentration and temperature in turbulent nonpremixed combustion characterized by a finite reaction rate. Discrepancy on the rich side may partly be due to limitation of the singly conditional moment closure, substitution of the unconditional mean variables for the conditional ones, or lack of adequate chemical kinetics on the rich side of methanol air reaction.

A numerical investigation of CO2 dilution on the thermochemical characteristics of a swirl stabilized diffusion flame

The turbulent combustion flow modeling is performed to study the effects of CO_2 addition to the fuel and oxidizer streams on the thermochemical characteristics of a swirl stabilized diffusion flame. A flamelet approach along with three well-known turbulence models is utilized to model the turbulent combustion flow field. The k-ω shear stress transport(SST) model shows the best agreement with the experimental measurements compared with other models. Therefore, the k-ω SST model is used to study the effects of CO_2 dilution on the flame structure and strength, temperature distribution, and CO concentration. To determine the chemical effects of CO_2 dilution, a fictitious species is replaced with the regular CO_2 in both the fuel stream and the oxidizer stream. The results indicate that the flame temperature decreases when CO_2 is added to either the fuel or the oxidizer stream. The flame length reduction is observed at all levels of CO_2 dilution. The H radical concentration indicating the flame strength decreases, following by the thermochemical effects of CO_2 dilution processes. In comparison with the fictitious species dilution, the chemical effects of CO_2 addition enhance the CO mass fraction. The numerical simulations show that when the dilution level is higher, the rate of the flame length reduction is more significant at low swirl numbers.

Numerical Investigation of an Impinging Diffusion Flames-Effects of Fuel Variability

In our study, we investigate the differences between the combustion of different hydrocarbon fuels CH4, C3H8, C4H10. A numerical simulation of an impinging jet diffusion flames is used. The jet injector has a 10 mm in diameter and the distance between the jet flame and the vertical wall is 2 time half diameter. The fuel jet velocity was fixed for 11.8 m/s, corresponding to a Reynolds number of 6881. The flame characteristics varied from hydrocarbon to another for the same Reynolds number. The combustion products of CO, CO2, NO, OH, are depending on the methane and propane and butane flames for the same conditions. The temperature of the flame was varied from hydrocarbon to another the same as for the chemical species production rate. The concentration of the thermal and prompt NO pollutant depends on the temperature flow field and on the thermochemical characteristics of the hydrocarbon fuels.