Turbulent Flames Research Articles

Seeing through fire with one pixel

Seeing through fire is a critical capability in various applications, such as fire rescue, industrial combustion monitoring, and scientific research. However, the intense electromagnetic radiation emitted by flames can saturate and blind conventional imaging systems, making it challenging to visualize objects or scenes obscured by fire. In this paper, we present a novel method for real-time, full-color through-flame imaging using structured illumination and single-pixel detection. By projecting a series of carefully designed light patterns onto the scene and measuring the backscattered light with a single-pixel detector, we can computationally reconstruct the obscured scene while effectively suppressing the flame's contribution to the image. The single-pixel detector's high dynamic range and sensitivity enable it to capture the weak backscattered signal without being overwhelmed by the flame's intense radiation. We demonstrate the method's effectiveness in several experiments, showcasing its ability to image static and dynamic scenes through both steady and turbulent flames at a frame rate of 15 Hz. Furthermore, we show that the proposed method can be extended to full-color imaging by using three single-pixel detectors with different color filters. The results highlight the potential of this approach for enhancing visibility in fire-related scenarios and other challenging imaging conditions. We believe that the integration of this technology into augmented reality systems could provide firefighters and other users with valuable real-time visual information, improving situational awareness and decision-making in critical situations. We also believe the method can be used for seeing through any self-luminous objects.

Local Burning Behavior of Wind-Driven Flames under the Influence of Mixed-Convective Turbulent Flow Conditions

ABSTRACT Experiments were conducted to investigate the burning behavior of wind-driven turbulent flames stabilized over a condensed fuel surface. A controlled turbulent crossflow environment was established to examine the impact of external flow velocity and turbulence intensity on the flame profile parameters and mass burning rates. To address the complexities of practical fire scenarios, a mixed-convection parameter ξ x in the form of G r x 1 / ψ x 2 n was defined, encapsulating the combined effects of momentum, buoyancy, and freestream turbulence. Furthermore, the property of fuel (in terms of mass transfer number B) was integrated into the mixed-convection parameter ( ξ x ) to establish a fuel-dependency factor within the formulation . The interdependence of flame characteristics and the mass burning process was observed using ξ x , that yielded several meaningful correlations. In this regard, a power-law trend was obtained between flame standoff distance and the dimensionless parameter ξ x . Finally, the local mass burning rates were estimated and plotted against mixed-convective parameter ( ξ x ), resulting in a unified local mass burning rate correlation. This correlation incorporates the fuel-dependency aspect in its formulation and is applicable in both laminar and turbulent crossflow environments.

Soot modeling of oxygen-depleted and highly sooty turbulent buoyant flames using the In Situ Adaptive Tabulation (ISAT) method

In this paper, a newly developed LSP-based soot (LSPTwoEq) modeling is systematically assessed in the Large Eddy Simulation (LES) of various turbulent buoyant flames against a previously developed LSP-based soot (LSPOneEq) model. Two turbulence-soot interaction (TSI) closure schemes are formulated based on the Eddy Dissipation Concept (EDC). Accordingly, two soot modeling frameworks consisting of the mentioned soot and their corresponding TSI models are applied in the simulations. An in situ Adaptive Tabulation (ISAT) implementation is also proposed to speed up the simulation of the TSI model formulated for the LSPTwoEq model. The results indicate that the proposed soot modeling framework is significantly more accurate for ethylene flames in reduced oxygen concentrations and a highly sooty toluene flame. The accuracy of both frameworks is comparable for the ethylene flame with an air co-flow. The speed-up gained by using the ISAT scheme is found to be a factor of four for the ethylene flame with an air co-flow and the toluene flame. It is larger than nine for the ethylene flames with reduced oxygen co-flow.

Learning-Based Super-Resolution Imaging of Turbulent Flames in Both Time and 3D Space Using Double GAN Architectures

This article presents a spatiotemporal super-resolution (SR) reconstruction model for two common flame types, a swirling and then a jet flame, using double generative adversarial network (GAN) architectures. The approach develops two sets of generator and discriminator networks to learn topographic and temporal features and infer high spatiotemporal resolution turbulent flame structure from supplied low-resolution counterparts at two time points. In this work, numerically simulated 3D turbulent swirling and jet flame structures were used as training data to update the model parameters of the GAN networks. The effectiveness of our model was then thoroughly evaluated in comparison to other traditional interpolation methods. An upscaling factor of 2 in space, which corresponded to an 8-fold increase in the total voxel number and a double time frame acceleration, was used to verify the model’s ability on a swirling flame. The results demonstrate that the assessment metrics, peak signal-to-noise ratio (PSNR), overall error (ER), and structural similarity index (SSIM), with average values of 35.27 dB, 1.7%, and 0.985, respectively, in the spatiotemporal SR results, can reach acceptable accuracy. As a second verification to highlight the present model’s potential universal applicability to flame data of diverse types and shapes, we applied the model to a turbulent jet flame and had equal success. This work provides a different method for acquiring high-resolution 3D structure and further boosting repeat rate, demonstrating the potential of deep learning technology for combustion diagnosis.

Effects of combustion progress variable and Karlovitz number on the scalar dissipation rate of turbulent premixed hydrogen-enriched methane–air flames: An experimental study

The scalar dissipation rate plays a key role in the modeling of turbulent premixed flames. Despite the above, our knowledge concerning the effects of several parameters, such as the combustion progress variable, Karlovitz number, and fuel composition on the scalar dissipation rate requires to be developed. This is carried out experimentally here. Two mixtures, methane–air (Lewis number of unity) and hydrogen-enriched methane–air (Lewis number of 0.8), are examined. Three passive (perforated plates) and one active turbulence generators are used, leading to a wide range of turbulence intensities with the corresponding Karlovitz number changing from 0.1 to 34.5. The hot-wire anemometry and the Rayleigh scattering techniques are used separately to study the background non-reacting turbulent flow and the reacting scalar dissipation rate, respectively. Three models that are available in the literature for estimating the Favre-averaged scalar dissipation rate are experimentally assessed. Using experimental results obtained in this study, it is shown that moving from fresh reactants towards combustion products increases the above modeling parameters by 2–3 times, depending on the tested Lewis number. For each tested Lewis number, empirical formulations that include the effects of both the Karlovitz number and combustion progress variable on the modeling parameters are proposed. Care must be taken in extending the proposed models of the present study to test conditions that are not examined here.

Differential diffusion modelling for LES of premixed and partially premixed flames with presumed FDF

Large eddy simulations (LES) with flamelet and presumed filtered density function closure are used to simulate turbulent premixed and partially premixed hydrogen flames. Different approaches to model differential diffusion are investigated and compared. In particular, two existing models are extended to the LES framework to correct the resolved diffusive flux of the controlling variables due to differential diffusion. A lean premixed turbulent hydrogen flame in a slotted burner configuration is simulated first to compare the capability of the considered models in capturing local mixture fraction redistribution, super-adiabatic temperatures and thermo-diffusive instabilities. Results show that both models describe the formation of cellular burning structures. Next, a partially premixed lifted hydrogen flame in vitiated hot coflow is simulated to gain insight on the relevance of differential diffusion modelling at a higher turbulence level, a different combustion mode and in the presence of a complex stabilisation mechanism. Good predictions of the turbulent mixing and temperature fields are observed. Moreover, results show that the flame lift-off height has an appreciable sensitivity to the differential diffusion model. When differential diffusion is included only in the thermochemistry database, only mild effects on the predicted temperature fields, mixing and flame height are observed. On the contrary, a considerable shift of the flame base is observed when corrections are applied in the LES at the resolved level, depending on what controlling variables are considered. Further analyses reveal how the corrections of diffusive fluxes in the thermochemistry and at the LES level affect differently the flame burning mode, whose details are given throughout the paper.

Isolating effects of large and small scale turbulence on thermodiffusively unstable premixed hydrogen flames

Lean turbulent premixed hydrogen/air flames have substantially increased flame speeds, commonly attributed to differential diffusion effects. In this work, the effect of turbulence on lean hydrogen combustion is studied through Direct Numerical Simulation using detailed chemistry and detailed transport. Simulations are conducted at six Karlovitz numbers and four integral length scales. A general expression for the burning efficiency is proposed which depends on the conditional mean chemical source term and gradient of a progress variable, and the amount of superadiabatic burning. At a fixed Karlovitz number, the normalized turbulent flame speed and area both increase almost linearly with the integral length scale ratio. The effect on the mean source term profile is minimal, indicating that the increase in flame speed can solely be attributed to the increase in flame area. At a fixed integral length scale, both the flame speed and area first increase with Karlovitz number before decreasing. The qualitative observations and trends do not change when Soret effects are neglected. Specifically, neglecting Soret diffusion is shown to reduce the flame speed, area, and burning efficiency. At higher Karlovitz numbers, the diffusivity is enhanced due to penetration of turbulence into the reaction zone, significantly dampening differential diffusion effects.Novelty and SignificanceLean premixed hydrogen flames are subject to thermodiffusive instabilities, which can lead to system level instabilities such as flashback. A comprehensive study of the thermodiffusively unstable turbulent flames with detailed chemistry and detailed transport (including Soret diffusion) was conducted across a wide range of turbulent intensities and integral length scales. Varying these two parameters independently was necessary to isolate the effects of large- and small-scale turbulence. Using these results, we propose a general expression for the burning efficiency to explain the relationship between the turbulence intensity, flame speed, and flame area. This work is an important step in developing predictive models which can aid in the design of practical combustion devices.

Experimental research on the flow field and flame geometry of free buoyant diffusion flames with low dimensionless heat release rates

The flame geometry of turbulent diffusion flames is dominated by the turbulent mixing between the gaseous fuel and the entrained air. The significant variations of the power law for the mean flame height regarding the dimensionless heat release rate (Q˙*) in different Q˙* ranges implied that the turbulent mixing characteristics would change greatly among these Q˙* regimes. This paper presents a combined experimental and analytical study to reveal the turbulent transport properties and flame shape of free turbulent buoyant diffusion flames of relatively low Q˙* prevalent in large-scale fires. The instantaneous velocity fields of free buoyant methane diffusion flames (burner diameter: d = 0.30 m, Q˙*= 0.22–0.40) were measured by the stereo particle image velocimetry (SPIV) technique. The results indicate that the radial profiles of the time-averaged axial velocity, axial normal stress, radial normal stress, shear stress, and turbulent kinetic energy exhibited self-similarity at different heights under all the heat release rates. The turbulent flow in the buoyant flame was found to be anisotropic. Based on the gradient transport approximation, the mean turbulent viscosity within the mean flame height (νt=) was scaled by g1/2d3/2, where g is the gravitational acceleration. The semi-physical correlations for the mean flame height and width were derived based on the concepts of turbulent mixing and equal axial convection and radial diffusion times, respectively. The two correlations agreed reasonably with the experimental data of this work.

Modeling of Soot Emission in Turbulent Diffusion Flames Impinging on a Cold Surface

ABSTRACT A hydrocarbon flame impinging on a cold surface produces an enhanced soot emission compared to the identical free flame. Cooking with biomass fuels is one practical process where soot emissions due to flame impingement has important environmental and health consequences. This problem was examined in a simplified system in which a turbulent non-premixed ethylene flame impinged on a cold surface. Previous experiment on this configuration examined the influence of a number of parameters on soot emission. The present paper uses numerical modeling to investigate the mechanism of soot emission enhancement in this configuration. Variations in the height of the surface above the fuel jet inlet showed going through a maximum, replicating the data. Since soot behavior can be affected by both thermal quench and flow field disruption, we decoupled the processes by replacing the cold surface with an adiabatic surface. These results indicated that most of the emission enhancement was due to the flow field disruption and not due to thermal quench. The modeling suggests that the presence of the surface changes the scalar dissipation rate, which influences C2H2 concentration, a key species in the soot surface growth process. Specifically, the surface causes a spike in the scalar dissipation rate adjacent to the surface, which significantly influences C2H2 concentration and results in increased soot production. The surface also forces the flame to spread radially, allowing more O2 to penetrate the flame immediately, enhancing the soot oxidation process. The combination of the enhanced soot production and oxidation produces the observed behavior.

Implementation of the P1 Radiation Model in a Velocity-Turbulent Frequency-Composition Joint PDF Method

ABSTRACT This research employs a more detailed radiation treatment to improve the accuracy of the velocity-turbulent frequency-composition joint probability density function (PDF) methods in numerical solution of the turbulent non-premixed flames. The radiation heat transfer in the traditional velocity-frequency-composition joint PDF methods has been disregarded or treated simply using an emission-only radiation model, even though radiation is important in the combustion systems. In the present PDF method, radiation is modeled using the P1-approximation and the radiative properties are calculated by the weighted-sum-of-gray-gases (WSGG) model. The results of present method are first validated by solving a laboratory turbulent piloted jet flame and comparing them with the measured data. Then, two constructed large-scale flames are modeled and simulated. Results reveal that reabsorption of emitted radiation increases from 2.98% in a laboratory-scale flame to 56.8% in the scaled-up flame and has a strong effect on the temperature distribution in the scaled-up flame. It is shown that reabsorption of emitted radiation increases the peak temperature on the flame axis by 119 degrees. Furthermore, radiation becomes more significant with the increase of the flame scales, so that the radiation fraction increases from 4.15% in the small-scale flame to 16.6% in the large-scale flame.

Nonpremixed and Premixed FPV Modeling of a Turbulent CH4-H2 Bluff-Body Flame

ABSTRACT In this study, modeling of nonpremixed CH4/H2 bluff-body (HM1) flame was performed using nonpremixed and premixed-based flamelet progress variable (FPV) models. The models were developed in an OpenFOAM flow solver, and the performance of the different variants of the FPV model was evaluated in the prediction of flame structure. The premixed FPV tables were constructed by employing two different interpolation techniques, i.e. linear and cubic spline, to retrieve the reactive scalars outside the flammability range. A generalized progress variable equation was introduced in the flow solver, which can be used for both tabulation approaches. Inside the flammable range, the profiles of the reactive scalars obtained from premixed FPV were found to be similar to those of the nonpremixed FPV. However, there exists disparity in the profiles outside the flammable range, particularly in the fuel-rich side between the nonpremixed and premixed approaches. A modified k-ε model was found to be suitable for modeling the turbulence compared to standard k-ε and k-ω shear stress transport models. The predictions of mean reactive scalars by the premixed FPV model, when combined with the linear interpolation, agree very well with experimental data and with the nonpremixed FPV predictions. However, deviations were observed in the predictions of the mean reactive scalar while using the cubic spline method in premixed FPV. All three approaches exhibit almost similar variations in the predictions of minor species. The results obtained from the present approach depict that the developed solver can be extended for modeling turbulent-chemistry interactions arising in partially premixed turbulent flames.

Development of a multiple laser-sheet imaging technique for the analysis of three-dimensional turbulent explosion flame structures

The present work describes the development of a multiple laser-sheet imaging technique for the study of turbulent, premixed flame surface structures at high Karlovitz stretch factor values. Experiments were conducted using CH4 air mixtures at 365 K and 0.5 MPa and up to a root mean square turbulence velocity of 1.5 m/s. A high-speed Nd:YAG laser capable of pulsing up to 60 kHz in conjunction with a high-speed camera and a rotating mirror was used to reconstruct time-resolved three-dimensional turbulent flames. This has, for the first time, enabled the direct measurement of wrinkled flame surface areas, along with reaction progress variable and flame brush thickness. These are important parameters for the characterization of turbulent burning rates and provide more insight into the dynamic nature of the flames and their structures. In addition, the current data aid toward direct comparison with results from combustion simulation studies.

A comparative study of the effect of cavity and obstacle on premixed methane–air flame evolution

To compare the effect of the cavity and obstacle on the flame evolution, a series of experimental studies are carried out to investigate the premixed methane-air flame propagation in the combustion chamber featured by cavity or obstacle. The flame behavior is obtained by analyzing the flame-front dynamics and pressure variation. The flame stretches into an inclined shape due to the flow contraction with the presence of an obstacle. In contrast, the flame forms a vortex by following the unburned gas vortex owing to the flow expansion in a cavity. Furthermore, the flame downstream of the obstacle is more turbulent than in cavity cases. Moreover, the flow contraction caused by obstacles also contributes to a higher flame-tip velocity than cavity cases. When the change in cross-section caused by obstacle and cavity is equivalent, the maximum pressure in the combustion chamber with an obstacle is generally larger than that in cavity cases ascribed to the combined effect of the changes in fuel amount and heat loss. In addition, the maximum rate of pressure rise in the combustion chamber featured by the obstacle is larger than that in the combustion chamber featured by a cavity, corresponding to the highly turbulent flame downstream of the obstacle.

Numerical Simulation of H2 Addition Effect to CH4 Premixed Turbulent Flames for Gas Turbine Burner

Numerical Simulation of H2 Addition Effect to CH4 Premixed Turbulent Flames for Gas Turbine Burner

Improved combustion stability of biogas at different CO2 concentrations using inhomogeneous partially premixed stratified flames

Biogas is a renewable and clean energy source that plays an important role in the current environment of low- carbon transition. If high-content CO2 in biogas can be separated, transformed, and utilized, it not only realizes high-value utilization of biogas but also promotes carbon reduction in the biogas field. To improve the combustion stability of biogas, an inhomogeneous, partially premixed stratified (IPPS) combustion model was adopted in this study. The thermal flame structure and stability were investigated for a wide range of mixture inhomogeneities, turbulence levels, CO2 concentrations, air-to-fuel velocity ratios, and combustion energies in a concentric flow slot burner (CFSB). A fine-wire thermocouple is used to resolve the thermal flame structure. The flame size was reduced by increasing the CO2 concentration and the flames became lighter blue. The flame temperature also decreased with increase in CO2 concentration. Flame stability was reduced by increasing the CO2 concentration. However, at a certain level of mixture inhomogeneity, the concentration of CO2 in the IPPS mode did not affect the stability. Accordingly, the IPPS mode of combustion should be suitable for the combustion and stabilization of biogas. This should support the design of highly stabilized biogas turbulent flames independent of CO2 concentration. The data show that the lower stability conditions are partially due to the change in fuel combustion energy, which is characterized by the Wobbe index (WI). In addition, at a certain level of mixture inhomogeneity, the effect of the WI on flame stability becomes dominant.

Modeling of Micro Aluminum Particle Flames Using Particle Burning Time

ABSTRACT This work presents a Lagrangian model, referred to as the integral model in this study, for aluminum-dust combustion, with the intent to reduce the computational burden and numerical stiffness. A comparison with available experimental data and a detailed model explicitly resolving chemical kinetics is presented for the canonical case of a planar flame. The influence of particle diameter, dust concentration and fresh-gases temperature on the flame speed and flame temperature is presented. Most results are within experimental uncertainties, unfortunately significant for aluminum-dust flames. Paths for improvements are proposed when needed. Regarding computational burden, it is shown that the integral model is faster and more robust than its counterpart using chemical kinetics. Insights on polydisperse flows and validation with constant volume combustion cases are provided. This study paves the way for a use in more complex cases such as turbulent flames or afterburning in energetic materials.

Investigation of a new method for direct chemistry integration in Conditional Source-term Estimation

This paper examines a new Conditional Source-term Estimation (CSE) implementation in which an integral inversion is performed for species mass fractions without any chemistry tabulation. To tackle the numerical stiffness due to chemistry, a constant-pressure chemical reactor is considered in conditional space. The objective of this work is to investigate this new method in CSE by comparing its numerical performance (accuracy of predictions and computational efficiency) with previous CSE implementations using tabulated chemistry. For the first time, the constraints for mass and mixture fraction conservation in conditional space are included in the inversion process. The investigation is performed in a Reynolds Averaged Navier-Stokes (RANS) framework using a well-documented turbulent flame with detailed measurements in conditional and physical spaces for many reacting scalars. A chemical mechanism including 16 species and 35 reactions is incorporated. CSE predictions of conditional and Favre averaged temperature and different species mass fractions are analysed. CSE with full chemistry performs as well as CSE with tabulated chemistry, if not slightly better. Numerical errors are discussed. The new CSE implementation is made possible by the proposed numerical treatment of stiffness due to chemistry and first-order Tikhonov regularisation technique with additional physical constraints. The new approach is found to be more CPU intensive than CSE with a pre-tabulation technique, but as a preliminary analysis, the computational cost of CSE with direct integration of a chemical mechanism appears to be much smaller than the run time associated with CMC for a similar set-up. Further investigation is required to refine this initial conclusion. These results open new opportunities towards CSE simulations that involve fuel blends and more complex operating conditions for example with spray and multi-stream inlets with non-uniform compositions for which accurate chemistry tabulations are difficult to generate.

Effect of pressure and H2 content on global consumption-based turbulent flame speed at gas turbine relevant conditions

This study focuses on the effect of pressure level and H2 content in the CH4–H2 fuel mixture on the global consumption-based turbulent flame speed (ST). The data considered in the analysis are collected for a variety of CH4–H2 fuel compositions (from 50% to 100% H2 vol.), pressure levels ranging from 0.5 to 1.5 MPa, and preheating temperatures up to 623.15 K. Different trends of ST/SL0 (SL0 is the adiabatic unstretched laminar flame speed) are observed as a function of the normalized turbulence intensity (u’/SL0), depending on the H2 content in CH4–H2 fuel blends. Yet, for all the CH4–H2 fuel mixtures considered, ST/SL0 consistently increases with H2 content. Rising pressure levels are responsible for larger ST/SL0, in direct proportion with the H2 content. These trends are mostly caused by the decrease in SL0, which causes changes to the characteristics of the flamelet that becomes more sensitive to hydrodynamic wrinkling through the high molecular diffusivity of H2.

Linear and nonlinear flame response prediction of turbulent flames using neural network models

Modelling the flame response of turbulent flames via data-driven approaches is challenging due, among others, to the presence of combustion noise. Neural network methods have shown good potential to infer laminar flames’ linear and nonlinear flame response when externally forced with broadband signals. The present work extends those studies and analyses the ability of neural network models to evaluate the linear and nonlinear flame response of turbulent flames. In the first part of this work, the neural network is trained to evaluate and interpolate the linear flame response model when presented with data obtained at various thermal conditions. In the second part, the neural network is trained to infer the nonlinear flame response model when presented with time series exhibiting sufficient large amplitudes. In both cases, the data is obtained from a large eddy simulation of an academic combustor when acoustically forced by broadband signals.

Effect of knock on piston thermal load of a high compression ratio natural gas engine based on stepwise decoupling calculation

High compression ratio natural gas engine (NGE) is prone to knock at high load, which aggravates the thermal load of piston. Therefore, accurately evaluating the piston thermal load in the practical design is essential to enhance engine reliability and improve performance. The typical piston thermal load simulation adopts uniform wall boundary conditions, making it difficult to reveal the influence of turbulent flame and knock on the piston thermal load. In this paper, a stepwise decoupling method is applied to calculate the piston thermal load. The advancement of this method lies in its integration of chemical reaction mechanism with computational fluid dynamics (CFD) to calculate the spatial variations in heat transfer coefficient (HTC) and near-wall gas temperature at the piston top. Moreover, the conjugate heat transfer (CHT) method is adopted to solve the piston thermal load by coupling the thermal boundary conditions (TBC) obtained in the first step with the cooling oil boundary conditions. Based on this method, the effect of knock on piston thermal load of a high compression ratio NGE under different knock intensities is explored in this paper. It is found that turbulent combustion causes uneven distribution of thermal load at piston top. The decrease in knock intensity significantly reduces the gas temperature, heat flux and thermal load at piston top. Particularly under severe knock, the average temperature of piston is 22.9 K and 32.4 K higher than that of light knock and normal combustion, respectively. Furthermore, the main factor affecting the piston temperature field distribution is the gas temperature rather than the HTC. The gas temperature at piston top is overall the highest on the side near the inlet valve pit, which leads to a higher temperature on this side. Compared to light knock and normal combustion, when severe knock occurs, the thermal load on the exhaust valve pit increases significantly. This study provides an effective method for analyzing the heat transfer of the piston under knocking combustion, which is of guiding significance for controlling the thermal load of the high-temperature components in the combustion chamber.