Ethanol turbulent spray flame response to gas velocity modulation

A numerical investigation of the interaction between a spray flame and an acoustic forcing of the velocity field is presented in this paper. In combustion systems, a thermoacoustic instability is the result of a process of coupling between oscillations in heat released and acoustic waves. When liquid fuels are used, the atomisation and the evaporation process also undergo the effects of such instabilities, and the computational fluid dynamics of these complex phenomena becomes a challenging task. In this paper, an acoustic perturbation is applied to the mass flow of the gas phase at the inlet and its effect on the evaporating fuel spray and on the flame front is investigated with unsteady Reynolds averaged Navier-Stokes numerical simulations. Two flames are simulated: a partially premixed ethanol/air spray flame and a premixed pre-vaporised ethanol/air flame, with and without acoustic forcing. The frequencies used to perturb the flames are 200 and 2500 Hz, which are representative for two different regimes. Those regimes are classified based on the Strouhal number St = (D/U)f_f: at 200 Hz, St = 0.07, and at 2500 Hz, St = 0.8. The exposure of the flame to a 200 Hz signal results in a stretching of the flame which causes gas field fluctuations, a delay of the evaporation and an increase of the reaction rate. The coupling between the flame and the flow excitation is such that the flame breaks up periodically. At 2500 Hz, the evaporation rate increases but the response of the gas field is weak and the flame is more stable. The presence of droplets does not play a crucial role at 2500 Hz, as shown by a comparison of the discrete flame function in the case of spray and pre-vaporised flame. At low Strouhal number, the forced response of the pre-vaporised flame is much higher compared to that of the spray flame.     

Dynamic behavior of a freely-propagating turbulent premixed flame under global stretch-rate oscillations

Dynamic features of a freely propagating turbulent premixed flame under global stretch rate oscillations were investigated by utilizing a jet-type low-swirl burner equipped with a high-speed valve on the swirl jet line. The bulk flow velocity, equivalence ratio and the nominal mean swirl number were 5 m/s, 0.80 and 1.23, respectively. Seven velocity forcing amplitudes, from 0.09 to 0.55, were examined with a single forcing frequency of 50 Hz. Three kinds of optical measurements, OH-PLIF, OH^* chemiluminescence and PIV, were conducted. All the data were measured or post-processed in a phase-locked manner to obtain phase-resolved information. The global transverse stretch rate showed in-phase oscillations centering around 60 (1/s). The oscillation amplitude of the stretch rate grew with the increment of the forcing amplitude. The turbulent flame structure in the core flow region varied largely in axial direction in response to the flowfield oscillations. The flame brush thickness and the flame surface area oscillated with a phase shift to the stretch rate oscillations. These two properties showed a maximum and minimum values in the increasing and decreasing stretch periods, respectively, for all the forcing amplitudes. Despite large variations in flame brush thickness at different phase angles, the normalized profiles collapse onto a consistent curve. This suggests that the self-similarity sustains in this dynamic flame. The global OH^* fluctuation response (i.e. response of global heat-release rate fluctuation) showed a linear dependency to the forcing velocity oscillation amplitudes. The flame surface area fluctuation response showed a linear tendency as well with a slope similar to that of the global OH^* fluctuation. This indicated that the flame surface area variations play a critical role in the global flame response.     

Response of non-premixed flames to bulk flow perturbations

This paper describes the dynamics of non-premixed flames responding to bulk velocity fluctuations, and compares the dynamics of the flame sheet position and spatially integrated heat release to that of a premixed flame. The space–time dynamics of the non-premixed flame sheet in the fast chemistry limit is described by the stoichiometric mixture fraction surface, extracted from the solution of the

-equation. This procedure has some analogies to premixed flames, where the premixed flame sheet location is extracted from the G = 0 surface of the solution of the G-equation. A key difference between the premixed and non-premixed flame dynamics, however, is the fact that the non-premixed flame sheet dynamics are a function of the disturbance field everywhere, and not just at the reaction sheet, as in the premixed flame problem. A second key difference is that the non-premixed flame does not propagate and so flame wrinkles are convected downstream at the axial flow velocity, while wrinkles in premixed flames convect downstream at a vector sum of the flame speed and axial velocity. With the exception of the flame wrinkle propagation speed, however, we show that that the solutions for the space–time dynamics of the premixed and non-premixed reaction sheets in high velocity axial flows are quite similar. In contrast, there are important differences in their spatially integrated unsteady heat release dynamics. Premixed flame heat release fluctuations are dominated by area fluctuations, while non-premixed flames are dominated by mass burning rate fluctuations. At low Strouhal numbers, the resultant sensitivity of both flames to flow disturbances is the same, but the non-premixed flame response rolls off slower with frequency. Hence, this analysis suggests that non-premixed flames are more sensitive to flow perturbations than premixed flames at O(1) Strouhal numbers.     

Impact of symmetry breaking on the Flame Transfer Function of a laminar premixed flame

This work presents a numerical study of the acoustic response of a laminar flame with tunable asymmetry. A V-shaped premixed flame is stabilised in the wake of a cylindrical flame holder that can be rotated. The configuration is symmetric when the flame holder is fixed but increasing its rotation rate breaks the symmetry of the flow. This configuration is submitted to acoustic forcing to measure the effect of rotation of the flame holder on the Flame Transfer Functions. It appears that the asymmetry of the two flame branches changes their respective time delays, resulting in interference in the global unsteady heat release rate fluctuations. Consequently, the Flame Transfer Function exhibits dips and bumps, which are studied via laminar Direct Numerical Simulation. Potential applications for the control of combustion instabilities are discussed.     

Experimental and numerical study of transcritical oxygen-hydrogen rocket flame response to transverse acoustic excitation

The response of a transcritical oxygen-hydrogen flame to transverse acoustic velocity was investigated using a combination of experimental analyses and numerical modelling. The experiment was conducted on a rectangular rocket combustor with shear coaxial injectors and continuously forced transverse acoustic field. Simultaneous high-speed shadowgraph and filtered OH* radiation images were collected and reduced using dynamic mode decomposition in order to characterise the flame response to the acoustic disturbance. CFD modelling of a representative single injector under forcing conditions was carried out to gain insights into the three-dimensional features of the reacting flow field. Invisible in the 2D projection, the model reveals that the excited LOX jet develops into a flattened and widened structure normal to the imposed acoustic velocity. The comparison of co-located structures allowed features in the imaging to be attributed to the deformation and transverse displacement of lower density oxygen surrounding the denser liquid oxygen core by the transverse acoustic velocity.     

Premixed flame response to equivalence ratio perturbations

This paper studies the heat-release oscillation response of premixed flames to oscillations in reactant stream fuel/air ratio. Prior analyses have studied this problem in the linear regime and have shown that heat release dynamics are controlled by the superposition of three processes: flame speed, heat of reaction, and flame surface area oscillations. Each contribution has somewhat different dynamics, leading to complex frequency and mean fuel/air ratio dependencies. The present work extends these analyses to include stretch and non quasi-steady effects on the linear flame dynamics, as well as analysis of nonlinearities in flame response characteristics. Because the flame response is controlled by a superposition of multiple processes, each with a highly nonlinear dependence upon fuel/air ratio, the results are quite rich and the key nonlinearity mechanism varies with mean fuel/air ratio, frequency, and amplitude of excitation. In the quasi-steady framework, two key mechanisms leading to heat-release saturation have been identified. The first of these is the flame-kinematic mechanism, previously studied in the context of premixed flame response to flow oscillations and recently highlighted by Birbaud et al. (Combustion and Flame 154 (2008), 356–367). This mechanism arises due to fluctuations in flame position associated with the oscillations in flame speed. The second mechanism is due to the intrinsically nonlinear dependence of flame speed and mixture heat of reaction upon fuel/air ratio oscillations. This second mechanism is particularly dominant at perturbation amplitudes that cause the instantaneous stoichiometry to oscillate between lean and rich values, thereby causing non-monotonic variation of local flame speed and heat of reaction with equivalence ratio.     

Influence of asymmetric flame transfer functions on can-to-can acoustic interactions between two lean-premixed combustors

In can-annular gas turbines, low-frequency thermoacoustic instabilities can arise from cross-talk interactions between neighboring combustors upstream of the first-stage turbine nozzles. In this experimental study, we investigate the influence of non-identical flame transfer functions (FTFs) between adjacent combustors on the development of self-excited thermoacoustic oscillations. To create different FTFs, we use five different swirl nozzles, one with high swirl (HS) and four with low swirl (LS), all with different porosities. We find that, compared with the LS FTFs, the HS FTF exhibits a smaller and flatter gain as well as a smaller phase difference. We attribute this behavior to differences in the flame structure and the stabilization mechanisms, namely an inner shear layer-stabilized diverging front in the HS case versus a detached reaction zone in the presence of a central jet with an outer swirl flow in the LS cases. Using two tunable lean-premixed combustors connected via a cross-talk section, we show that (i) symmetric FTF combinations (HS + HS or LS + LS) produce in-phase interactions, leading to push-push modes, but that (ii) asymmetric FTF combinations (HS + LS) produce out-of-phase interactions, leading to push-pull modes. Phase-resolved visualization of the asymmetric cases reveals that the inner shear layer-stabilized HS flame exhibits large angle fluctuations, whereas the aerodynamically stabilized LS flame is characterized by the periodic emergence of a bow-shaped front and an oval structure. For all the conditions tested, we find that asymmetry in the FTFs leads to either (i) a completely stable state with negligible amplitude or (ii) a mildly unstable state with an amplitude lower than that of the equivalent symmetric cases. These findings highlight the potential of using asymmetric FTFs for passive control of cross-talk-driven thermoacoustic instabilities in can-annular combustors.     

Unsteady effects on flame extinction limits during gaseous and two-phase flame/vortex interactions

In highly fluctuating flows, it happens that high values of the strain-rate do not induce extinction of the flame front. Unsteady effects minimize the flame response to rapidly varying strain fields. In the present study, the effects of time-dependent flows on non-premixed flames are investigated during flame/vortex interactions. Gaseous flames and spray flames in the external sheath combustion regime are considered. To analyse the flame/vortex interaction process, the velocity field and the flame geometry are simultaneously determined using particle imaging velocimetry and laser-induced fluorescence of the CH radical. The influence of vortex flows on the extinction limits for different vortex parameters and for different gaseous and two-phase flames is examined. If the external perturbation is applied over an extended period of time, the extinction strain-rate is that corresponding to the steady-state flame, and this critical value mainly depends on the fuel and oxidizer compositions and the injection temperature. If the external perturbation is applied during a short period of time, extinction occurs at strain-rates above the steady-state extinction strain-rate. This deviation appears for flow fluctuation timescales below steady flame diffusion timescales. This behaviour is induced by diffusive processes, limiting the ability of the flame to respond to highly fluctuating flows. With respect to unsteady effects, the spray flames investigated in this article behave essentially like gaseous flames, because evaporation takes place in a thin layer before the flame front. Extinction limits are only slightly modified by the spray, controlling process being the competition between aerodynamic and diffusive timescales.     

Autoignition delay modulation by high-frequency thermoacoustic oscillations in reheat flames

This paper investigates the sensitivity of the autoignition delay in reheat flames to acoustic pulsations associated with high-frequency transverse thermoacoustic oscillations. A reduced order model for the response of purely autoignition-stabilised flames to acoustic disturbances is compared with experimental observations. The experiments identified periodic flame motion associated with high-amplitude transverse limit-cycle oscillations in an atmospheric pressure reheat combustor. This flame motion was assumed to be the result of a superposition of two flame-acoustic coupling mechanisms: autoignition delay modulation by the oscillating acoustic field and displacement and deformation of the flame by the acoustic velocity. The reduced order model coupled to reaction kinetics calculations reveals that a significant portion of the observed flame motion can be attributed to autoignition delay modulation. The ignition position responds instantaneously to the acoustic pressure at the time of ignition, as observed experimentally. The model also provides insight into the importance of the history of acoustic disturbances experienced by the fuel-air mixture prior to ignition. Due to the high-frequency nature of the instability, a fluid particle can experience multiple oscillation cycles before ignition. The ignition delay responds in-phase with the net-acoustic perturbation experienced by a fluid particle between injection and ignition. These findings shed light on the underlying mechanisms of the flame motion observed in experiments and provide useful insight into the importance of autoignition delay modulation as a driving mechanism of high-frequency thermoacoustic instabilities in reheat flames.     

On nonlinear model equations for the response of premixed flames to acoustic like accelerations

The response of a dynamical flame model to imposed acoustic accelerations is studied analytically and numerically. Through linear stability analyses, two analytical approximations for the primary and the parametric stability boundaries are found. The approximation for the primary instability boundary is accurate for any periodic accelerations, in the limit of large acoustic frequencies. The critical acoustic amplitude u _a for Landau–Darrieus instability suppression is identified and found to depend only on the density contrast and the shape of the periodic acoustic stimuli. The proposed model evolution equation is next integrated numerically with various imposed acoustic accelerations; the primary and parametric flame responses are identified. It is shown analytically and numerically that in the presence of a fully developed, yet weakened by acoustics, Landau–Darrieus (or primary) instability the wrinkle amplitude and the mean flame speed oscillate at the same frequency as the acoustic stimuli; the threshold for suppression of primary instability by acoustic forcing is determined exactly. Increasing the acoustic amplitude allows the flame to respond parametrically to the acoustics. This response is characterised by troughs and crests interchanging their roles while the mean flame speed again oscillates with the same frequency as the acoustic stimuli and at twice that of wrinkle amplitude oscillations.     

Stabilization mechanism of lifted flame edge in the near field of coflow jets for diluted methane

The stabilization mechanism of lifted flames in the near field of coflow jets has been investigated experimentally and numerically for methane fuel diluted with nitrogen. The lifted flames were observed only in the near field of coflow jets until blowout occurred in the normal gravity condition. To elucidate the stabilization mechanism for the stationary lifted flames of methane having the Schmidt number smaller than unity, the behavior of the flame in the buoyancy-free condition, and unsteady propagation characteristics after ignition were investigated numerically at various conditions of jet velocity. It has been found that buoyancy plays an important role for flame stabilization of lifted flames under normal gravity, such that the flame becomes attached to the nozzle in microgravity. The stabilization mechanism is found to be due to the variation of the propagation speed of the lifted flame edge with axial distance from the nozzle in the near field of the coflow as compared to the local flow velocity variation at the edge.     

Physics-informed recurrent neural networks for linear and nonlinear flame dynamics

This paper demonstrates the ability of recurrent neural networks (RNNs) to predict the linear and the nonlinear response of a premixed laminar flame to incoming velocity perturbations. We develop data-driven models, which require the velocity and heat release rate fluctuations as input data. Both time series are obtained from Direct Numerical Simulations (DNS) of a laminar flame. The length of the signals, and, hence, the cost of the simulation, is comparable to those used in the linear framework of System Identification. A more robust type of RNNs, namely long short term memory (LSTM), is employed to reduce the dependency on large datasets. The LSTM framework is modeled as a time series regression problem and four models are trained with decreasing data set lengths. All purely data-driven models accurately predict the unsteady time series of the heat release rate and, hence, the Flame Transfer Functions (FTFs). We further improve the model accuracy by incorporating a physical constraint, namely the low-frequency limit for perfectly-premixed flames, into the LSTM model. This step reduces the required data length compared to the purely data-driven approach. The proposed model, called PI-LSTM, is able to reproduce the linear and the nonlinear FTFs for amplitudes up to 50% of the laminar flame based on one numerical simulation, where the length of the time series is 100 ms.     

The response of turbulent premixed flames to normal acoustic excitation

To model the thermo-acoustic excitation of flames in practical combustion systems, it is necessary to know how a turbulent flame front responds to an incident acoustic wave. This will depend partly on the way in which the burning velocity responds to the wave. In this investigation, the response of CH₄/air and CH₄/H₂/air mixtures has been observed in a novel flame stabilisation configuration, in which the premixture of fuel and air is made to decelerate under controlled conditions in a wide-angle diffuser. Control is provided by an annular wall-jet of air and by turbulence generators at the inlet. Ignition from the outlet of the diffuser allows an approximately flat flame to propagate downwards and stabilise at a height that depends on the turbulent burning velocity. When the flow is excited acoustically, the ensemble-averaged height oscillates. The fluctuations in flow velocity and flame height are monitored by phase-locked particle image velocimetry and OH-planar laser induced fluorescence, respectively. The flame stabilised against a lower incident velocity as the acoustic amplitude increased. In addition, at the lowest frequency of 52 Hz, the fluctuations in turbulent burning velocity (as represented by the displacement speed) were out-of-phase with the acoustic velocity. Thus, the rate of displacement of the flame front relative to the flow slowed as the flow accelerated, and so the flame movement was bigger than it would have been if the burning velocity had not responded to the acoustic fluctuation. With an increase in frequency to 119 Hz, the relative flame movement became even larger, although the phase-difference was reduced, so the effect on burning velocity was less dramatic. The addition of hydrogen to the methane, so as to maintain the laminar burning velocity at a lower equivalence ratio, suppressed the response at low amplitude, but at a higher amplitude, the effect was reversed.     

Lift-off characteristics of triple flame with concentration gradient

The concentration gradient and uniform mean velocity of a triple flame in a mixing layer were studied using a multi-slot burner, which can stabilize the lift-off flame especially at a very small concentration gradient. Flame stabilization conditions were examined, and the lift-off heights of the triple flame were measured for methane and propane flames. A hot-wire anemometer was used to measure the velocity distributions. Mass spectroscopy (for methane) and Rayleigh scattering (for propane) were used to measure the concentration gradients. OH radical distribution was measured by laser-induced fluorescence (LIF), and in-stream velocity variation was measured with particle-image velocimetry (PIV). Maximum in-stream temperatures were measured using the coherent anti-Stokes Raman scattering (CARS) technique. Lift-off heights of triple flames have minimum values during the increase of concentration gradient, and the propagation velocity of triple flames reaches its maximum at a critical concentration gradient. This is caused by three factors: velocity distribution upstream, flammable limit of premixed gas, and reaction of diffusion flame. The critical concentration gradient, which maximizes the propagation velocity is suggested as a new criterion of transition from a premixed flame to a triple flame.     

Effect of fuel mixture fraction and velocity perturbations on the flame transfer function of swirl stabilized flames

A novel methodology is developed to decompose the classic Flame Transfer Function (FTF) used in the thermo-acoustic stability analysis of lean premix combustors into contributions of different types. The approach is applied, in the context of Large Eddy Simulation (LES), to partially-premixed and fully-premixed flames, which are stabilized via a central recirculation zone as a result of the vortex breakdown phenomenon. The first type of decomposition is into contributions driven by fuel mixture fraction and dynamic velocity fluctuations. Each of these two contributions is further split into the components of turbulent flame speed and flame surface area. The flame surface area component, driven by the pure dynamic velocity fluctuation, which is shown to be a dominant contribution to the overall FTF, is also additionally decomposed over the coherent flow structures using proper orthogonal decomposition. Using a simplified model for the dynamic response of premixed flames, it is shown that the distribution of the FTF, as obtained from LES, is closely related to the characteristics of the velocity field frequency response to the inlet perturbation. Initially, the proposed method is tested and validated with a well characterized laboratory burner geometry. Subsequently, the method is applied to an industrial gas turbine burner.     

Nonlinear hydrodynamic and thermoacoustic oscillations of a bluff-body stabilised turbulent premixed flame

Turbulent premixed flames often experience thermoacoustic instabilities when the combustion heat release rate is in phase with acoustic pressure fluctuations. Linear methods often assume a priori that oscillations are periodic and occur at a dominant frequency with a fixed amplitude. Such assumptions are not made when using nonlinear analysis. When an oscillation is fully saturated, nonlinear analysis can serve as a useful avenue to reveal flame behaviour far more elaborate than period-one limit cycles, including quasi-periodicity and chaos in hydrodynamically or thermoacoustically self-excited system. In this paper, the behaviour of a bluff-body stabilised turbulent premixed propane/air flame in a model jet-engine afterburner configuration is investigated using computational fluid dynamics. For the frequencies of interest in this investigation, an unsteady Reynolds-averaged Navier–Stokes approach is found to be appropriate. Combustion is represented using a modified laminar flamelet approach with an algebraic closure for the flame surface density. The results are validated by comparison with existing experimental data and with large eddy simulation, and the observed self-excited oscillations in pressure and heat release are studied using methods derived from dynamical systems theory. A systematic analysis is carried out by increasing the equivalence ratio of the reactant stream supplied to the premixed flame. A strong variation in the global flame structure is observed. The flame exhibits a self-excited hydrodynamic oscillation at low equivalence ratios, becomes steady as the equivalence ratio is increased to intermediate values, and again exhibits a self-excited thermoacoustic oscillation at higher equivalence ratios. Rich nonlinear behaviour is observed and the investigation demonstrates that turbulent premixed flames can exhibit complex dynamical behaviour including quasiperiodicity, limit cycles and period-two limit cycles due to the interactions of various physical mechanisms. This has implications in selecting the operating conditions for such flames and for devising proper control strategies for the avoidance of thermoacoustic instability.     

Experimental comparison of opposed and concurrent flame spread in a forced convective microgravity environment

Flame spread experiments in both concurrent and opposed flow have been carried out in a 5.18-s drop tower with a thin cellulose fuel. Flame spread rate and flame length have been measured over a range of 0–30 cm/s forced flow (in both directions), 3.6–14.7 psia, and oxygen mole fractions 0.24–0.85 in nitrogen. Results are presented for each of the three variables independently to elucidate their individual effects, with special emphasis on pressure/oxygen combinations that result in earth-equivalent oxygen partial pressures (normoxic conditions). Correlations using all three variables combined into a single parameter to predict flame spread rate are presented. The correlations are used to demonstrate that opposed flow flames in typical spacecraft ventilation flows (5–20 cm/s) spread faster than concurrent flow flames under otherwise similar conditions (pressure, oxygen concentration) in nearly all spacecraft atmospheres. This indicates that in the event of an actual fire aboard a spacecraft, the fire is likely to grow most quickly in the opposed mode as the upstream flame spreads faster and the downstream flame is inhibited by the vitiated atmosphere produced by the upstream flame. Additionally, an interesting phenomenon was observed at intermediate values of concurrent forced flow velocity where flow/flame interactions produced a recirculation downstream of the flame, which allowed an opposed flow leading edge to form there.     

Measurements of three-dimensional mean flame surface area ratio in turbulent premixed Bunsen flames

A data processing scheme with particular emphasis on proper flame contour smoothing is developed and applied to measure the three-dimensional mean flame surface area ratio in turbulent premixed flames. The scheme is based on the two-sheet imaging technique such that the mean flame surface area ratio is an average within a window covering a finite section of the turbulent flame brush. This is in contrast to the crossed-plane tomograph technique which applies only to a line. Two sets of Bunsen flames have been investigated in this work with the turbulent Reynolds number up to 4000 and the Damköhler number ranging from less than unity to close to 10. The results show that three-dimensional effects are substantial. The measured three-dimensional mean flame surface area ratio correlates well with a formula similar to the Zimont model for turbulent burning velocity but with different model constants. Also, the mean flame surface area ratio displays a weak dependency on turbulence intensity but a strong positive dependency on the turbulence integral length scale.