Turbulent Premixed Flames Research Articles

Simultaneous visualization of instantaneous unburnt and preheating zones in turbulent premixed flames under transverse acoustic excitations

Instantaneous unburnt and preheating zones of bluff-body stabilized turbulent premixed flames under transverse acoustic excitations were investigated using simultaneous planar laser-induced fluorescence (PLIF) of acetone and CH2O, as well as multi-point hot-wire measurements. The PLIF images show that the unburnt zone marked by acetone images, the preheating zone marked by CH2O images, and the pixel-by-pixel product of acetone/CH2O have an increasing distribution area when slowly enlarging the sound pressure level (SPL). Wrinkled and bent edges of the unburnt and preheating zone can be seen at conditions away from the flame blow-off in the presence of the transverse acoustic excitations, and their sizes and areas increase as the flame blow-off is approached. At conditions near the flame blow-off with enlarging SPL to 123 dB, the flame turns from side to side over time and a large scale of the acetone/CH2O regions can be observed to deflect inside the center product zone, implying that the cold reactants can enter the product zone from the unburnt/preheating zones. The unburnt/preheating mainstream presents strong wrinkles and partial fractures. Such a phenomenon indicates that the local extinction of the shear layer flame can also be facilitated due to the turbulent fluctuation enhanced by the transverse acoustic wave. For a low flow velocity, increasing variations of the unburnt and preheating zones in the presence of the transverse acoustic wave can be revealed. The curvature of the acetone PLIF shows that the unburnt zones are more likely to be wrinkled with an increasing SPL and flow velocity. The root-mean-squared velocity measurements stress that the transverse acoustic wave mainly affects the turbulent premixed flame by enhancing the turbulent fluctuations.

Planar laser-induced imaging of CH3 for high resolution single-shot reaction-zone visualization in premixed methane/air flames over broad stoichiometric ratios

We report a novel approach for single-shot planar imaging of CH3 radicals in premixed methane and air flames. A 213 nm beam from the 5th harmonic of an Nd:YAG laser was resonantly absorbed by the CH3 radicals, which were excited to the highly pre-dissociative upper level and dissociated to H2 and CH (X), as the main dissociation channel. The CH radicals were consequently excited by a 388 nm beam from an alexandrite laser, and the fluorescence from the excited CH radicals was collected off-resonant at 431 nm. Using this Photo-Fragmentation Planar Laser-Induced Fluorescence (PF-PLIF) technique, instantaneous flame front structures, represented by CH3 radicals, can be visualized with high spatial resolution over a broad range of stoichiometric ratios. Signal-to-noise ratios up to 50 were observed for premixed methane/air flame with stoichiometric ratio as low as 0.26. The CH radicals naturally presented in flame front are more than 400 times lower in concentration than the CH3 radicals in premixed methane/air flames even at the conditions close to stoichiometric or slightly fuel rich cases where the highest CH concentrations exist, and the CH3/CH concentration ratios increase dramatically moving towards fuel lean conditions. By adopting a structured illumination of the 213 nm pump beam, the naturally presented CH radicals were visualized simultaneously with CH3 at slightly fuel rich laminar flames, where the CH signal intensity was 5 times lower than that from CH3. The results indicate that the CH3 PF-PLIF technique can provide much stronger signal than the CH PLIF and presented a much promising potential for applications in fuel-lean flames. Finally, the CH3 PF-PLIF was performed in premixed turbulent flames to demonstrate the feasibility of this technique for flame front visualization in turbulent premixed flames.

Influence of flow structures and combustion on cross-scale turbulent kinetic energy transfer in premixed swirl flames.

This paper experimentally analyzes the simultaneous influence of combustion and large-scale vortex structures on the transfer of kinetic energy across scales around the laminar flame thickness δL0 in a turbulent premixed swirl flame and a non-reacting swirl flow. High-resolution tomographic particle image velocimetry and formaldehyde planar laser induced fluorescence measurements are used to obtain 3D velocity fields and estimates of the progress variable and density fields. The kinetic energy transfer across a filter scale of Δ=1.5δL0 was then quantified using physical space analysis. Coherent flow structures were identified using the swirling strength and proper orthogonal decomposition was used to identify the dominant periodic flow structure. While non-reacting regions of the flow show mean down-scale energy transfer (forward-scatter), mean back-scatter is observed internal to the flame. Importantly, the back-scatter magnitude in the flame increases in regions undergoing flame/vortex interaction. That is, the mean back-scatter magnitude at locations simultaneously inside the flame and a large-scale coherent vortex is higher than regions in the flame and not in a vortex, with the mean back-scatter magnitude increasing with the swirling strength. This increased back-scatter may be due to locally higher heat release rates in locations of flame/vortex interaction. Overall, the results demonstrate the importance of kinetic energy back-scatter at scales around the flame thickness and the complicated relationship between back-scatter and local flame/flow structure; these results should be considered when modeling turbulence in flames.

Active control of combustion oscillation with active disturbance rejection control (ADRC) method

This research aims to develop a novel method for the active control of combustion oscillation based on active disturbance rejection control (ADRC). The effectiveness of parameter tuning and the controller in suppressing combustion oscillation was studied by simulation and experiments. The control performance was verified using a Rijke tube and turbulent premixed swirling flame within a model combustor. For the former, the results show that the first-order ADRC can rapidly mitigate the limit cycle oscillation and maintain the effectiveness of the control when the oscillation frequency changes. For the latter, a system identification method based on an acoustic network was adopted. Combining the identification results with a thermoacoustic network, the formation process of the limit cycle oscillation and the effectiveness of the first-order ADRC were simulated by SIMULINK. The experimental results show that the first-order ADRC performs better than the proportional–integral–derivative (PID) controller. At the same time, the first-order ADRC and second-order ADRC achieve good outcomes when the equivalence ratio changes. However, the second-order ADRC exhibits a better pressure-suppression effect and smaller controller output voltage.

Inertial effects on the interaction of water droplets with turbulent premixed flames: A direct numerical simulation analysis

The effects of droplet inertia on the reaction zone structure, overall burning rate, and flame surface area during interaction of water droplets with a statistically planar turbulent premixed stoichiometric n-heptane-air flame based on three-dimensional carrier phase Direct Numerical Simulations have been analysed. Different initially mono-sized droplets are considered for this study in order to analyse the inertial effects for different droplet sizes. Droplet inertia has been demonstrated to have an important influence on the extent of evaporation of water droplets and water vapour concentration arising from evaporation within the flame. It has been found that the residence time of droplets within the flame increases due to droplet inertia and therefore the evaporation and the cooling effect associated with the extraction of latent heat are stronger for inertial droplets than in the case of hypothetical inertialess droplets. The stronger cooling effects for inertial droplets lead to smaller burned gas temperature and thicker flame than the corresponding cases with inertialess droplets. The aforementioned cooling effect induced by latent heat of evaporation of water droplets reduces the burning rate and the likelihood of obtaining high gradient magnitudes of reaction progress variable and temperature within the flame for the cases with water droplets in comparison to the corresponding purely gaseous turbulent premixed flames, and this tendency is stronger for inertial droplets. The stronger cooling effect, along with the reduction of reactant concentrations within the flame due to greater extent of evaporation for inertial droplets, gives rise to a smaller extent of flame area generation and overall burning rate than in the corresponding cases with inertialess droplets. It has been found that water droplet inertia acts to reduce the burning rate and the consumption rate of reactants per unit flame surface area during the interaction of water droplets with turbulent premixed flames within the parameter range analysed in this paper.

Modelling filtered reaction rate in turbulent premixed flames using feature importance analysis, gene expression programming and tiny artificial neural networks

Two different machine learning (ML) approaches, Symbolic regression, more precisely Gene Expression Programming (GEP) and Tiny Artificial Neural Networks (TANN), have been used for deducing a model for the filtered reaction rate in the context of Large Eddy Simulation (LES) of premixed combustion. The models have been trained on data generated from a large range of filter widths and for different turbulence intensities, using a simple chemistry Direct Numerical Simulation (DNS) database of statistically planar turbulent premixed flames. Feature importance analysis has been applied to identify the important input variables for modelling filtered reaction rate in order to enable an efficient optimization process as well as moderate inference times. The resulting models show excellent correlation and good quantitative agreement with the filtered reaction rates extracted from explicitly filtered DNS data. The genetic optimization results in a mathematical expression that can be easily implemented, analysed and documented. Also, the TANN has a small number of trainable parameters that can be shared with the research community. The present results demonstrate that ML algorithms can be successfully used for modelling turbulent premixed flames in the context of Large Eddy Simulation. Both approaches can fulfil the requirements of transparency and have inference times that are acceptable in comparison to algebraic models.

Turbulent flame image classification using Convolutional Neural Networks

Pockets of unburned material in turbulent premixed flames burning CH4, air, and CO2 were studied using OH Planar Laser-Induced Fluorescence (PLIF) images to improve current understanding. Such flames are ubiquitous in most natural gas air combustors running gas turbines with dry exhaust gas recirculation (EGR) for land-based power generation. Essential improvements continue in the characterization and understanding of turbulent flames with EGR particularly for transient events like ignition and extinction. Pockets and/or islands of unburned material within burned and unburned turbulent media are some of the features of these events. These features reduce the heat release rates and increase the carbon monoxide and hydrocarbons emissions. The present work involves Convolutional Neural Networks (CNN) based classification of PLIF images containing unburned pockets in three turbulent flames with 0%, 5%, and 10% CO2. The CNN model was constructed using three convolutional layers and two fully connected layers using dropout and weight decay. Accuracies of 94.2%, 92.3% and 89.2% were registered for the three flames, respectively. The present approach represents significant computational time savings with respect to conventional image processing methods.

Modeling subgrid-scale scalar dissipation rate in turbulent premixed flames using gene expression programming and deep artificial neural networks

In this present study, gene expression programing (GEP) has been used for training a model for the subgrid scale (SGS) scalar dissipation rate (SDR) for a large range of filter widths, using a database of statistically planar turbulent premixed flames, featuring different turbulence intensities and heat release parameters. GEP is based on the idea to iteratively improve a population of model candidates using the survival-of-the-fittest concept. The resulting model is a mathematical expression that can be easily implemented, shared with the community, and analyzed for physical consistency, as illustrated in this work. Efficient evaluation of the cost function and a smart choice of basis functions have been found to be essential for a successful optimization process. The GEP based model has been found to outperform an existing algebraic model from the literature. However, the optimization process was found to be quite intricate, and the SGS SDR closure turned out to be difficult. Some of these problems have been explained using the model-agnostic interpretation method, which requires the existence of a trained artificial neural network (ANN). ANNs are known for their ability to represent complex functional relationships and serve as an additional benchmark solution for the GEP based model.

Numerical analysis of turbulent flame enhancement by nanosecond repetitively pulsed plasma discharges

Experimental studies of plasma-assisted combustion have shown that Nanosecond Repetitively Pulsed (NRP) discharges are a very efficient way to improve flame stability in lean regimes. These discharges generate a non-equilibrium plasma that induces a local heating and an important production of active species sufficient to enhance the combustion. The aim of this article is to understand the physical and chemical mechanisms involved in plasma-assisted turbulent combustion. High-performance computations of a lean bluff-body turbulent premixed methane-air flame, experimented at EM2C laboratory, are conducted for that purpose. Simulations are performed by combining a semi-empirical plasma model with an analytical combustion mechanism. Without electrical discharges, only a weak reaction zone behind the bluff body is observed. However, when the flame is assisted with the plasma, the flame power and surface increase significantly. A chemical analysis performed at the flame basis shows that the heat and the radical O produced by NRP discharges induce the dissociation of main burnt gases species H2O and CO2 into radicals OH, H, CO, O and species H2, O2. The species with long lifespan OH, H2 and O2 are convected from the center of the bluff-body recirculation zone to the flame front where they are consumed increasing the chemical reactivity. This causes a local increase of heat release rate that attaches the turbulent flame front which can develop downstream.

Interactions Between Flame Topology and Turbulent Transport in High-Pressure Premixed Combustion

Direct numerical simulations of a turbulent premixed stoichiometric methane-oxygen flame were conducted. The chosen combustion pressure was 20 bar, to resemble conditions encountered in modern rocket combustors. The chemical reactions followed finite rate detailed mechanism integrated into the EBI-DNS solver within the OpenFOAM framework. Flame geometry was thoroughly investigated to assess its interaction with the transport of turbulent properties. The resulting flame front was remarkably thin, with high density gradients and moderate Karlovitz and Damköhler numbers. At mid-flame positions, the variable-density transport mechanisms dominated, leading to the generation of both vorticity and turbulence. A reversion of this trend towards the products was observed. For intermediate combustion progress, vorticity transport is essentially a competition between the baroclinic torque and vortex dilatation. The growth of turbulent kinetic energy is strongly correlated to this process. A geometrical analysis reveals that the generation of enstrophy and turbulence is restricted to specific topologies. Convergent and divergent flame propagation promote turbulence creation due to pressure fluctuation gradients through different physical processes. The possibility of modeling turbulence transport based on curvature is discussed along with the inherent challenges.

Local effects in vortex-flame interactions: Implications for turbulent premixed flame scaling and observables

A computational investigation was carried out to assess the effects of highly energetic vortices on the local structure of laminar premixed flames. The study involved a lean atmospheric methane-air flame interacting with vortex sizes ranging from five times to half of the laminar flame thickness. The characteristic velocities of the vortices were chosen to correspond to extreme turbulence intensities. Vortices smaller than the flame thickness were found to have a minimal effect on the flame structure due to the high rate of dissipation. Vortices with sizes greater than or equal to the flame thickness were found to induce flame siphoning and/or product-side local flame annihilation and as a result transport considerable amounts of formaldehyde and thermal energy to the unburned side thereby modifying the state of the reactants. The results of the present study for high activation energy reacting flows raise questions regarding the general validity of: (1) conjectured mechanisms of the so-called flame broadening under intense turbulence conditions; and (2) qualitative arguments regarding sub-flame-thickness vortex effects on the preheat zone in the context of the Borghi–Peters regime diagram, as originally postulated by Damköhler.

Energy integral equation for premixed flame-wall interaction in turbulent boundary layers and its application to turbulent burning velocity and wall flux evaluations

An integral form of the energy conservation equation has been derived from the first principle for low Mach number conditions for statistically steady premixed flame-wall interaction within turbulent boundary layers. The validity of this equation has been demonstrated based on three-dimensional Direct Numerical Simulation data of statistically stationary oblique quenching of a turbulent premixed V-shaped flame in a channel flow configuration as a result of its interaction with an inert isothermal wall. It has been found that the wall heat flux and the integral of chemical heat release in the wall normal direction within the turbulent thermal boundary layer are the major contributors in the energy integral equation, and their difference is accounted for by the advection contribution. The magnitudes of the wall heat flux increase, and integral of heat release rate across the thermal boundary layer decrease with increasing distance from the leading edge of the boundary layer as a result of flame quenching. The integral form of the energy conservation equation has been utilised to demonstrate that the Nusselt number (or Stanton number) for wall heat transfer is intrinsically related to the turbulent burning velocity in the case of flame-wall interaction within turbulent boundary layers. A Flame Surface Density based reaction rate closure, modified to account for the near-wall behaviour, has been utilised to estimate the mean Nusselt number in the case of flame-wall interaction within turbulent boundary layers, which revealed that the modelling limitations of the mean reaction rate closure may give rise to inaccuracies in the estimation of the mean Nusselt number. By contrast, the measurements of mean velocity, temperature, and wall heat flux can be utilised to estimate the turbulent burning velocity within the turbulent boundary layer using the newly derived energy integral equation.

Experimental Study Of The Interaction Between A Turbulent Flame And A Cooling Air Film

The current experimental study is focused on the analysis of the interactions between a V-shaped lean turbulent premixed methane-air flame and a parietal cooling air film generated by means of a splash-cooling system. The flame structure, the reactive aerodynamics and the wall temperature are simultaneously studied by implementing planar laser-induced fluorescence of the OH radical (OH-PLIF), reactive particle image velocimetry (PIV) and surface intensity-ratio phosphor thermometry (PT). Results show that the mean flow field and the mean flame topology significantly differs according to the level of blowing ratio. Further joint correlations of experimental measurements highlight various FCAI regimes. For blowing ratios below unity, a common flame-wall interaction process occurs and the air film has a limited influence on the flame dynamics. Increasing the blowing ratio enables to establish a cooling air film at the wall. Though it effectively increases the level of thermal protection, it does not have much influence on the flame topology. Eventually, for larger blowing ratios, the flame dynamics is piloted by the shear layer dynamics, associated with strong flow strain. The flame is pushed away from the wall and is located in the outer region of the shear layer. The cooling effectiveness is not improved anymore and shows no more dependencies with the blowing ratio.

Enstrophy evolution during head-on wall interaction of premixed flames within turbulent boundary layers

The statistical behaviors of mean enstrophy and its evolution during head-on interaction of premixed flames propagating toward a chemically inert flat wall across the turbulent boundary layer have been analyzed using direct numerical simulations for a friction velocity-based Reynolds number of Reτ=110. The enstrophy dynamics have been analyzed for both isothermal and adiabatic thermal wall boundary conditions. The contributions of vortex-stretching and viscous dissipation are found to be leading order source and sink, respectively, to the mean enstrophy transport in both non-reacting and reacting flows irrespective of the wall boundary condition. However, the contributions due to dilatation rate and baroclinic torque play important roles in addition to the leading order contributions of the vortex-stretching and viscous dissipation terms in the enstrophy transport in turbulent premixed flames. The thermal boundary condition has been demonstrated to affect the near-wall behavior of the enstrophy transport contribution due to dilatation rate, which also affects the near-wall distribution of the enstrophy. The magnitudes of the leading order contributors to the enstrophy transport decrease with the progress of head-on interaction for both wall boundary conditions. Moreover, the overall sink contributions to the enstrophy transport dominate over the source contributions, giving rise to a drop in the mean enstrophy with the progress of head-on interaction. The enstrophy distribution changes significantly during flame-wall interaction, which gives rise to a modification of the relative proportion of the coherent structures in the reacting flow turbulent boundary layer compared to the corresponding non-reacting flow features.

Computations of a Bluff-Body Stabilised Premixed Flames Using ERN Method

Combustible carbon-based energy is still prevailing as the world’s leading energy due to its high energy density. However, the oxidation of these hydrocarbons disturbs the natural carbon cycle greatly by increasing greenhouse gases. As emission legislation becomes more rigorous, lean premixed combustion becomes promising because it can reduce nitrogen oxides (NOx) and Carbon Monoxide (CO) emissions without compromising efficiency. However, utilising lean premixed flames in industrial combustors is not easy because of its thermo-acoustic instabilities associated with pressure fluctuations and the non-linearity in the mean reaction rate. Therefore, reliable predictive combustion models are required to predict emissions with sensible computational costs to use the mode efficiently in designing environmentally friendly combustion systems. Along with the promising methodologies capable of modelling turbulent premixed flames with low computational costs is the ERN-RANS framework. Thus, this study aims to compute a bluff-body stabilised premixed flames close to blow-Off using the ERN-RANS framework. As a result, a satisfactory agreement is reached between the predicted and measured values.

Development of a turbulence dissipation based reaction rate model for progress variable in turbulent premixed flames

This study presents an algebraic combustion closure for Large eddy simulation (LES) exhibiting attributes of simplicity and simultaneous accuracy under realistic combustion conditions. The model makes use of the interlink between the reaction and dissipation rates in premixed turbulent combustion but relaxes the thin flame assumption by considering finite-rate chemistry effects in the small-scale turbulence structure. The core idea of the approach is to approximate the reaction progress in the unresolved spectrum of wave lengths and to use it within a filtered reaction rate expression. The model is implemented in OpenFOAM 4.0 and is tested on a turbulent, premixed flame behind a bluff-body, applying an LES approach for turbulence modelling. The cross comparison of velocity, temperature and composition data with experiments and a well-investigated combustion model in literature reveals competitive performance of the new model. Especially in the near-field of the bluff body flame, corresponding to thin and moderately thickened flame regions, its ability to capture the flame structure is highly promising. The chosen, partly explicit approach to recover the temperature from the transported sensible enthalpy, involving a strong coupling between filtered reaction and heat release rate, also shows advantages over obtaining the temperature from presumed probability density functions.

A Priori Analysis on Deep Learning of Filtered Reaction Rate

A filtered reaction rate model driven by deep learning is proposed and analyzed a priori in the context of large eddy simulation (LES). A deep artificial neural network (ANN) is trained on the explicitly filtered reaction rate source term extracted from a database comprised of turbulent premixed planar flame direct numerical simulations (DNSes) employing single-step chemistry. The filtered DNS database to be used for the training of the ANN covers a wide range of turbulence intensities and LES filter widths. An interpretation technique of deep learning is employed to search the principal input parameters in the high dimensional database to alleviate the model complexity. The deep learning filtered reaction rate model is then tested on the unseen filtered planar flames featuring untrained turbulence intensities and LES filter widths, in conjunction with another canonical type of flame configuration that it has not been trained on. The deep learning filtered reaction rate model achieves good agreement with the filtered DNS results and also provides a quantitatively accurate surrogate model when compared to existing algebraic models and other combustion models from the literature.

Alignment statistics of pressure Hessian with strain rate tensor and reactive scalar gradient in turbulent premixed flames

The relative alignment of the eigenvectors of pressure Hessian with reactive scalar gradient and strain rate eigenvectors in turbulent premixed flames have been analyzed for Karlovitz number values ranging from 0.75 to 126 using a detailed chemistry three-dimensional direct numerical simulations database of H2–air premixed flames. The reactive scalar gradient preferentially aligns with the most extensive strain rate eigendirection for large Damköhler number and small Karlovitz number values, whereas a preferential collinear alignment between the reactive scalar gradient with the most compressive strain rate eigendirection is observed in flames with small Damköhler number and large Karlovitz number. By contrast, the eigenvectors of pressure Hessian do not perfectly align with the reactive scalar gradient, and the net effect of the pressure Hessian on the evolution of the normal strain rate contribution to the scalar dissipation rate transport acts to reduce the scalar gradient in the zone of high dilatation rate. The eigenvectors of pressure Hessian and the strain rate are aligned in such a manner that the contribution of pressure Hessian to the evolution of principal strain rates tends to augment the most extensive principal strain rate for small and moderate values of Karlovitz numbers, whereas this contribution plays an important role for the evolution of the intermediate principal strain rate for large values of Karlovitz number. As the reactive scalar gradient does not align with the intermediate strain rate eigenvector, the influence of pressure Hessian contributions to the scalar–turbulence interaction remains weak for large values of Karlovitz number.

Turbulence-induced bias in time-averaged laser absorption tomography of correlated concentration and temperature fields with a first-order correction

The influence of correlated scalar fluctuations on time-averaged laser absorption tomography measurements of temperature and species in a piloted turbulent premixed flame was examined using a coupled spectroscopic and fluid-dynamic analysis. To understand bias associated with turbulence, spatio-temporally resolved temperature and species mole fraction profiles predicted by large eddy simulations (LES) were used to synthetically generate time-resolved line-of-sight absorption measurements at short time scales (microsecond) to reflect the unsteady nature of a canonical jet burner across various transverse measurement planes. Inversion methods were employed on the time-averaged line-of-sight data to produce radially-resolved temperature and mole fraction profiles, analogous to those produced by laser absorption tomography performed on a time-averaged axisymmetric flowfield. It is shown that bias in the measurements compared to true time-averaged scalar fields is a function primarily of temperature dependence in absorptivity and non-zero correlation between temperature and species concentration scalars. A first-order correction to tomography measurements is proposed to account for the bias based on estimated correlations and the known spectroscopic parameters of the probed absorption transitions.

Self-wrinkling induced by Darrieus-Landau instability in turbulent premixed Bunsen flames from low to moderately high Reynolds numbers

In this experimental work we analyze various turbulent premixed flames under the effects (or lack thereof) of the Darrieus-Landau hydrodynamic instability. In particular, the effects of an increasing level of turbulence are considered by means of different observables. On the one hand we obtain results which can improve the understanding of fundamental mechanisms ruling the interaction between flame fronts, either unstable and stable, and turbulence. On the other hand, conclusions we draw may lead to new definitions or improvement of closure models for numerical simulation of turbulent premixed flames under the effects of hydrodynamic instability.