Non-premixed Diffusion Flames Research Articles

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.

Study on the combustion, entrainment, and plume flow behaviors of annular pool fires.

Annular fire source is a common combustion form in fire accidents. Effects of Din/Dout (the ratio of inner to outer diameters of the floating-roof tank) on the flame morphology and plume entrainment mechanisms of annular pool fires were studied by numerical simulation. Results show, as Din/Dout increases, the area with low combustion intensity near the central axis of the pool surface gradually increases. Combined with the time-series HRR and the stoichiometric mixture fraction line of the fire plume, it reveals that the combustion of annular pool fire is dominated by non-premixed diffusion flame. The pressure near the pool outlet decreases with Din/Dout, while the plume turbulence presents an opposite trend. Based on the time-sequential plume flow and the gas-phase material distribution data, the flame merging mechanism of the annular pool fires is revealed. Furthermore, based on the similarity criterion, it verifies that the applicability of the above scaled simulation conclusions could also be extended to guide full-scale fires.

Generalizing progress variable definition in CFD simulation of combustion systems using tabulated chemistry models

In the Computational Fluid Dynamics (CFD) simulation of advanced combustion systems, the chemical kinetics must be examined in detail to predict the emissions and performance characteristics accurately. Nevertheless, the combustion simulation with detailed chemical kinetics is complicated because of the number of equations and a broad timescale spectrum. The Flamelet-Generated Manifold (FGM) is one of the examples of tabulation methods that has received much attention in recent years due to its fast and accurate prediction of combustion characteristics. The Progress Variable (PV) definition in FGM and other PV-based tabulated approaches is often selected randomly or depending on the user's experience. When complicated combustion systems are involved, such choices can become extremely difficult. In the current work, a generic approach for formulating a global PV is developed and tested in various operating conditions relevant to combustion engines. The method is based on a genetic algorithm optimization to maximize the monotonicity of PV, ensuring that for each value of PV, the dependent thermophysical properties have unique values. The FGM model's ability to reproduce the detailed kinetics evolution of the essential combustion and emission parameters of a non-premixed diffusion flame in Spray A configuration is evaluated in both one-dimensional counterflow and CFD simulation. It is concluded that with the use of the current approach, important combustion characteristics can be predicted much better compared to non-optimized PV while eliminating the manual selection of PV definition by the user. Since the algorithm needs to be executed before the chemistry tabulation in the pre-processing step, it does not increase the runtime of the FGM simulation. The algorithm only needs a few minutes to be finished on a standard desktop. The improvement in the results and the distribution of the values of important species in the computational domain is examined.

Experimental measurements of soot formation in fuel-rich homogeneous mixtures using an optical rapid compression machine

Advanced combustion strategies are necessary for the use of more environmentally sustainable fuels than traditional diesel. Alcohol fuels and alcohol/gasoline blends are of particular interest as they are readily available in the marketplace. Heavy-duty engines typically use compression ignited, conventional diesel mixing controlled combustion. Mixing controlled combustion features a non-premixed diffusion flame with a wide range of local equivalence ratios, leading to potentially high rates of soot formation. This work studies the sooting behavior of iso-octane and ethanol as a function of equivalence ratio. Measurements are carried out in a rapid compression machine (RCM) and are reported for pre-ignition conditions of 10–30 bar and temperatures of 650–800 K. Theoretical equilibrium and bulk gas temperatures are calculated for both fuels. These data are used to identify the critical equivalence ratio, the lowest equivalence ratio where soot is detected with a single-pass laser extinction diagnostic. The critical equivalence ratio for iso-octane varies between 1.82 and 1.77 for compressed pressures of 10 and 20 bar, respectively. Ethanol, sometimes considered sootless, had a critical equivalence ratio between 2.37 and 2.12 for compressed pressures of 20 and 30 bar, respectively. When characterizing soot formation by oxygenated equivalence ratio, the critical equivalence ratios for ethanol approach those of iso-octane. This suggests the oxygenated nature of alcohol fuels reduces sooting tendency, but other factors such as fuel molecular structure and morphology may play a role. It was observed that ethanol will form soot at equivalence ratios only slightly higher than iso-octane, which could have implications in mixing controlled combustion. It was seen for both fuels that soot formation is pressure sensitive, with the critical equivalence ratio being inversely proportional to compressed pressure and the rate of soot formation. Future work will investigate the sooting behavior of gasoline/ethanol blends.

Application of the Computational Singular Perturbation Method to a Turbulent Diffusion CH4/H2/N2 Flame Using OpenFOAM

This work highlights the ability of the computational singular perturbation (CSP) method to calculate the significant indices of the modes on evolution of species and the degree of participation of reactions. The exploitation of these indices allows us to deduce the reduced models of detailed mechanisms having the same physicochemical properties. The mechanism used is 16 species and 41 reversible reactions. A reduction of these 41 reactions to 22 reactions is made. A constant pressure application of the detailed and reduced mechanism is made in OpenFOAM free and open source code. Following the Reynolds-averaged Navier–Stokes simulation scheme, standard k–ε and partial stirred reactor are used as turbulence and combustion models, respectively. To validate the reduced mechanism, comparison of numerical results (temperature and mass fractions of the species) was done between the detailed mechanism and the simplified model. This was done using the DVODE integrator in perfectly stirred reactor. After simulation in the computational fluid code dynamic (CFD) OpenFOAM, other comparisons were made. These comparisons were between the experimental data of a turbulent nonpremixed diffusion flame of type “DLR-A flame,” the reduced mechanism, and the detailed mechanism. The calculation time using the simplified model is considerably reduced compared to that using the detailed mechanism. An excellent agreement has been observed between these two mechanisms, indicating that the reduced mechanism can reproduce very well the same result as the detailed mechanism. The accordance with experimental results is also good.

On the influences of key modelling constants of large eddy simulations for large-scale compartment fires predictions

ABSTRACTAn in-house large eddy simulation (LES) based fire field model has been developed for large-scale compartment fire simulations. The model incorporates four major components, including subgrid-scale turbulence, combustion, soot and radiation models which are fully coupled. It is designed to simulate the temporal and fluid dynamical effects of turbulent reaction flow for non-premixed diffusion flame. Parametric studies were performed based on a large-scale fire experiment carried out in a 39-m long test hall facility. Several turbulent Prandtl and Schmidt numbers ranging from 0.2 to 0.5, and Smagorinsky constants ranging from 0.18 to 0.23 were investigated. It was found that the temperature and flow field predictions were most accurate with turbulent Prandtl and Schmidt numbers of 0.3, respectively, and a Smagorinsky constant of 0.2 applied. In addition, by utilising a set of numerically verified key modelling parameters, the smoke filling process was successfully captured by the present LES model.

Effects of Substitution on Counterflow Ignition and Extinction of C3 and C4 Alcohols

Dwindling reserves and inherent uncertainty in the price of conventional fuels necessitates a search for alternative fuels. Alcohols represent a potential source of energy for the future. The structural features of an alcohol fuel have a direct impact on combustion properties. In particular, substitution in alcohols can alter the global combustion reactivity. In this study, experiments and numerical simulations were conducted to investigate the critical conditions of extinction and autoignition of n-propanol, 1-butanol, iso-propanol, and iso-butanol in nonpremixed diffusion flames. Experiments were carried out in the counterflow configuration, while simulations were conducted using a skeletal chemical kinetic model for the C3 and C4 alcohols. The fuel stream consists of the prevaporized fuel diluted with nitrogen, while the oxidizer stream is air. The experimental results show that autoignition temperatures of the tested alcohols increase in the following order: iso-propanol > iso-butanol > 1-butanol ≈ n-propanol. The simulated results for the branched alcohols agree with the experiments, while the autoignition temperature of 1-butanol is slightly higher than that of n-propanol. For extinction, the experiments show that the extinction limits of the tested fuels increase in the following order: n-propanol ≈ 1-butanol > iso-butanol > iso-propanol. The model suggests that the extinction limits of 1-butanol are slightly higher than n-propanol with extinction strain rate of iso-butanol and iso-propanol maintaining the experimentally observed trend. The transport weighted enthalpy (TWE) and radical index (Ri) concepts were utilized to rationalize the observed reactivity trends for these fuels.

Modelling of combustion performances and emission characteristics of coal gases in a model gas turbine combustor

SUMMARY In recent years, gas mixtures are being used as alternative fuels in combustors. These gas mixtures are obtained by different methods. For instance, coal gasification and carbonization as coal have the largest reserves among fossil fuels. Gas mixtures obtained via coal gasification and carbonization are called water gas, generator gas, town gas and coke oven gas. These fuels contain various gases. As a result of this, heating values of fuels are also different. Therefore, combustion performances and emission characteristics of these fuels need to be investigated. In this study, combustion performances and emissions including CO, CO2 and NOX of water gas, generator gas, town gases, coke oven gas and methane were numerically investigated in a model gas turbine combustor. The numerical modelling of turbulent nonpremixed diffusion flames has been performed in this combustor. Mathematical models used in this study involved the k–e model of turbulent flow, the PDF/mixture fraction model of nonpremixed combustion and P-1 radiation model. A CFD code ANSYS Fluent was used for all numerical investigations. Temperature distributions of axial and radial directions were determined. A NOX post-processor was used for the prediction of NOX emissions from the gas turbine combustor. Modelling was performed for 60 kW thermal power and different equivalance ratios (i.e. Ф = 0.91, Ф = 0.77 and Ф = 0.67). The studied type 1 model gas turbine combustor was modelled for Ф = 0.91 equivalance ratio. Then, Other equivalance ratios were analysed for type 2 model gas turbine combustor. The effect of dilution air on combustion performances and emission characteristics was also investigated. It is concluded that the coke oven gas, the town gas I, town gas II and the water gas are appropriate for usage as alternative fuel, whereas the generator gas is not suitable for gas turbine combustors. Copyright © 2013 John Wiley & Sons, Ltd.

일반 공기 및 순산소 연소 조건에서 Fuel-NOx생성 특성의 비교

Nitric oxide () formation characteristics in non-premixed diffusion flames of methane fuels have been investigated experimentally and numerically by adding 10% ammonia to the fuel stream, according to the variation of the oxygen ratio in the oxidizer with oxygen/carbon dioxide and oxygen/nitrogen mixtures. In an experiment of coflow jet flames, in the case of an oxidizer with oxygen/carbon dioxide, the emission increased slightly as the oxygen ratio increased. On the other hand, in case of an oxygen/nitrogen oxidizer, the emission was the maximum at an oxygen ratio of 0.7, and it exhibited non-monotonic behavior according to the oxygen ratio. Consequently, the emission in the condition of oxyfuel combustion was overestimated as compared to that in the condition of conventional air combustion. To elucidate the characteristics of formation for various oxidizer compositions, 1D and 2D numerical simulations have been conducted by adopting one kinetic mechanism. The result of 2D simulation for an oxidizer with oxygen/nitrogen well predicted the trend of experimentally measured emissions.

Modeling and simulation of titanium dioxide nanoparticle synthesis with finite-rate sintering in planar jets

Numerical simulations of titanium dioxide nanoparticle synthesis in planar, non-premixed diffusion flames are performed. Titania is produced by the oxidation of titanium tetrachloride using a methane–air flame. The flow field is obtained using the two-dimensional Navier–Stokes equations. The methane–air flame and oxidation of titanium tetrachloride are modeled via one-step reactions. Evolution of the particle field is obtained via a nodal method which accounts for nucleation, condensation, coagulation, and coalescence with finite-rate sintering. The modeling of finite-rate sintering is accomplished via the use of uniform primary-particle size distribution. Simulations are performed at two different jet-to-co-flow velocity ratios as well as with finite-rate and instantaneous sintering models. In doing so we elucidate the effect of fluid mixing and finite-rate sintering on the particle field. Results show that highly agglomerated particles are found on the periphery of the eddies, where the collisions leading to nanoparticle coagulation occur faster than nanoparticle coalescence.

A parallel adaptive mesh refinement algorithm for predicting turbulent non-premixed combusting flows

A parallel adaptive mesh refinement (AMR) algorithm is proposed for predicting turbulent non-premixed combusting flows characteristic of gas turbine engine combustors. The Favre-averaged Navier–Stokes equations governing mixture and species transport for a reactive mixture of thermally perfect gases in two dimensions, the two transport equations of the k–ω turbulence model, and the time-averaged species transport equations, are all solved using a fully coupled finite-volume formulation. A flexible block-based hierarchical data structure is used to maintain the connectivity of the solution blocks in the multi-block mesh and facilitate automatic solution-directed mesh adaptation according to physics-based refinement criteria. This AMR approach allows for anisotropic mesh refinement and the block-based data structure readily permits efficient and scalable implementations of the algorithm on multi-processor architectures. Numerical results for turbulent non-premixed diffusion flames for a bluff-body burner are described and compared to available experimental data. The numerical results demonstrate the validity and potential of the parallel AMR approach for predicting complex non-premixed turbulent combusting flows.

Diffusion flame instabilities in a 0.75 mm non-premixed microburner

We examine the cellular instabilities of laminar non-premixed diffusion flames that arise in a polycrystalline alumina microburner with a channel wall gap of dimension 0.75 mm. Changes in the flame structure are observed as a function of the fuel type (H 2, CH 4, and C 3H 8) and diluent. The oxidizer is O 2/inert. In contrast to previous observations on laminar diffusion flame instabilities, the current instabilities occur in the direction of flow above the splitter plate, and only occur for the heavier fuel types. They are not observed in a H 2–O 2 mixture, which will only support a continuous laminar flame inside our burner, regardless of the initial mixture strength and whether or not the flame is in near-quenching conditions. The only exception is when helium is added to the H 2–O 2 mixture, raising the effective Lewis numbers of both components.

Numerical Simulation of Non-Premixed Diffusion Flame and Reaction Product Aerosol Behavior in Liquid Metal Pool Combustion

In the present study, a numerical methodology has been developed to solve a non-premixed diffusion flame under natural convection. The methodology has been applied to the calculation of the liquid sodium pool combustion experiment at various pool temperatures and oxygen molar fractions. The authors have proposed expressions of aerosol dynamics, radiation heat transfer and evaporation of liquid sodium that are applicable to sodium combustion phenomena. The computations reproduce the experimental observations concerning burning rate, flame temperature and flame height, consistently. The aerosol release fractions are also in good agreement with the measurement. Dominant mechanisms of the mass and heat transfer are identified through the numerical simulation. An intrinsic feature found in the present study is that the liquid sodium pool combustion is self-limited and a negative feedback mechanism is at work. Interaction among the thermal-hydraulics, chemical reaction and aerosol dynamic behavior plays an important role in the phenomena and it has been successfully analyzed by the numerical simulation. The present method can be used to understand sodium combustion phenomena and applied to the modeling of sodium pool combustion for safety analyses of liquid metal fast reactors. The numerical simulation is a useful tool because it can easily employ various conditions by changing parameters.

Passive and active control of NOx in industrial burners

Industrial burners that use either fuel oil or natural gas are considered as candidates for implementing strategies for control of NO x . Two burners, at 20 and 800 kW burning fuel oil are installed in a tunnel furnace and subjected to active control by way of introducing pulses into the air supply. Using optical diagnostics to identify the CH radical, the imaging reveals the evolution of the vortical structures within flames: a non-premixed diffusion flame appearing in the braids of the vortex and a premixed core region. Prompt NO x mechanisms are explored, noting that thermal NO x in this case is not the predominant contributor in the fuel burner. In the case of natural gas burners, thermal NO x control is achieved passively. The burners achieve significant reductions in NO x by radically altering the fuel-air jet distribution from the normal air-surrounding fuel ports approach: the air and fuel jets are arranged on a base circle, each low momentum fuel port alternating with high momentum air port and each type pointing away from the burner axis at a different, but constant, angle. The NO x reduction mechanism is speculated to be due to flue gas recirculation with concomitant reduction in flame temperature. The burner has been tested in both single and multiple modes in a unique multi-burner furnace. Results obtained using both active and passive control and manipulation indicate that significant enhancement of combustion can be achieved by applying modern tools and new ideas.