Global Flame Response Research Articles

The Effect of Rectangular Confinement Aspect Ratio on the Flame Transfer Function of a Turbulent Swirling Flame

Abstract The flame transfer function (FTF) has been experimentally investigated for a premixed, turbulent and highly swirled flame in rectangular enclosures with varying aspect ratio. A range of equivalence ratios and inlet bulk velocities are considered to vary the length of the flame. Increasing confinement is also shown to increase flame length. More confined flames become increasingly asymmetric, with significant changes to the mean flame shape. Characteristic peaks and dips are observed in the FTF gain for several configurations, caused by constructive and destructive interference between different sources of flow perturbations. Changes in the geometrical confinement distinctly affects these interactions, with FTFs in the most confined chamber containing more pronounced peaks and dips. Analysis of the phase-averaged dynamics has been conducted for two limiting operational conditions in the least and most confined enclosures: a short flame at a low bulk velocity and high equivalence ratio condition; and a long flame at a high velocity and low equivalence ratio condition. The analysis of the short flames shows similar behavior in both enclosures, both in terms of the global response and the local structure of the heat release rate oscillations. Small differences in local response symmetry in the flame due to the close confinement do not affect the global flame response. By comparison, close confinement significantly affects the symmetry of the flame dynamics in the long flame. However, the changes in symmetry do not significantly modify the response, and more important is the change in flame length, which significantly alters the cutoff frequency, reducing the gain at high frequencies.

The effect of hydrogen enrichment, flame-flame interaction, confinement, and asymmetry on the acoustic response of a model can combustor

To maximise power density practical gas turbine combustion systems have several injectors which can lead to complex interactions between flames. However, our knowledge about the effect of flame-flame interactions on the flame response, the essential element to predict the stability of a combustor, is still limited. The present study investigates the effect of hydrogen enrichment, flame-flame interaction, confinement, and asymmetries on the linear and non-linear acoustic response of three premixed flames in a simple can combustor. A parametric study of the linear response, characterised by the flame transfer function (FTF), is performed for swirling and non-swirling flames. Flame-flame interactions were achieved by changing the injector spacing and the level of hydrogen enrichment by power from 10 to 50%. It was found that the latter had the most significant effect on the flame response. Asymmetry effects were investigated by changing one of the flames by using a different bluff-body to alter both the flame shape and flow field. The global flame response showed that the asymmetric cases can be reconstructed using a superposition of the two symmetric cases where all three bluff-bodies and flames are the same. Overall, the linear response characterised by the flame transfer function (FTF) showed that the effect of increasing the level of hydrogen enrichment is more pronounced than the effect of the injector spacing. Increasing hydrogen enrichment results in more compact flames which minimises flame-flame interactions. More compact flames increase the cut-off frequency which can lead to self-excited modes at higher frequencies. Finally, the non-linear response was characterised by measuring the flame describing function (FDF) at a frequency close to a self-excited mode of the combustor for different injector spacings and levels of hydrogen enrichment. It is shown that increasing the hydrogen enrichment leads to higher saturation amplitude whereas the effect of injector spacing has a comparably smaller effect.

Decomposition of swirling flame transfer function in the complex space

In this work, we propose an analytical decomposition of the flame transfer function (FTF) in the complex space. The decomposition provides new insight into the relationship between the local heat release oscillator and the global flame response of a periodically oscillating system in the linear regime. The FTF of a premixed methane-air swirling flame is decomposed using two-region and pixel-by-pixel division of the chemiluminescence images. The two-region decomposition uses horizontal or vertical dividing lines to examine the effects of dividing positions and orientations. The complex curves of the two-region division are more sensitive to the dividing line's position than its orientation. The pixel-by-pixel decomposition provides detailed distributions of the weighting factor, weighted gain, and phase of all local oscillators. The weighted gain distribution highlights the inner fluctuating region (IFR) and outer fluctuating region (OFR), as well as the node region. These regions are mainly controlled by the radial velocity and its fluctuations induced by the vortices in the inner and outer shear layers. The unwrapped phase, which gradually increases as the downstream distance from the injector increases, highlights the convective characteristics of the oscillating heat release rate. Angular wavenumbers and phase velocities are deduced from the phase distribution. The phase velocities agree well with the mean velocities measured with particle image velocimetry. Sorting all the local oscillators in the ascending phase allows constructing complex curves that favorably represent the FTF. The algebraic sum of weighted gain and the efficiency of phase interference are calculated to examine the phase interference. Phase interference plays a large role in suppressing the global fluctuating amplitude of the heat release rate. The local extrema at low and intermediate frequencies are attributed to in- or out-of-phase interference, while the reduced local response also explains why small FTF gains are observed at high frequencies.

Effect of wall boundary conditions on the nonlinear response of turbulent premixed flames

Fully compressible Large Eddy Simulations (LES) are performed to investigate the effect of wall boundary conditions on the nonlinear response of turbulent premixed flames. For the unforced flames, the flame length and flame angle from the present LES results in the isothermal case are in better agreement with experimental results than those in the adiabatic case. For the forced flames, both amplitudes and phases of nonlinear flame response at moderate frequency are well reproduced in the isothermal case. While in the adiabatic case, the amplitudes are well reproduced, the prediction of phases is slightly underestimated. At high frequency, LES results from the isothermal case provide reasonable agreement with the experimental results in general, while the adiabatic case gives unreasonable prediction of nonlinear flame response. Results show that wall boundary conditions affect the spatial distributions of heat release fluctuations by changing the temperature field, which is affected by the evolutions of the flame structure and flow field in the nonlinear oscillation cycle. At moderate frequency, heat release fluctuations in different parts of the combustor are in phase, and wall boundary conditions have limited influence on the global flame response, while at high frequency, wall boundary conditions have significant influence on the global flame response since the heat release fluctuations in different parts of combustor are out of phase. Accurate prediction of flame response at high frequency needs accurate calculation of the wall temperature.

Premixed Flame Dynamics in Response to Two-Dimensional Acoustic Forcing

ABSTRACTThermoacoustic instability in annular combustors involves feedback interaction between individual (acoustically compact) flames and – predominantly – azimuthal acoustic modes of the annular combustion chamber. During such an interaction, the flames experience acoustic pressure fluctuations in the combustor and transverse acoustic velocity fluctuations simultaneously. To investigate the dynamics of premixed swirl flames under the influence of transverse acoustics, experiments were performed on a single-burner atmospheric combustion rig with controlled simultaneous axial and transverse acoustic forcing, where the axial acoustics simulates the effect of acoustic pressure fluctuations in the combustor. In this paper, we present results on the descriptive analysis of spatially resolved flame surface fluctuations, acquired via OH-PLIF (OH* planar laser-induced florescence) diagnostics of the forced flame and establish the role of the interaction among coherent structures generated by axial and transverse velocity components in generating global flame response changes with respect to the phase difference between axial and transverse acoustic forcing as reported recently (Saurabh and Paschereit, Combustion Flame 182, 2017).

Experimental Study of the Global and Local Flame Responses to Acoustic Perturbation

In this study, the global and local flame responses to the acoustic perturbation were investigated to understand the thermoacoustic features of a low swirl flame. The effects of the Strouhal number and the acoustic perturbation level were examined. The global flame response was investigated with flame chemiluminescence. With the images captured with the planar laser-induced fluorescence of hydroxyl radicals method, the phase-averaged flame surface density was calculated to examine the flame dynamics and the local mean heat release oscillation. The results showed that the Strouhal number was useful in characterizing the flame responses: especially the oscillation amplitude. The flame showed a significant shape deformation when intense flame rollup happened; otherwise, the flame shape deformed weakly. The local analysis showed that large flame deformations caused strong local fundamental oscillations. However, the contribution of the local fundamental oscillation to the global fundamental fluctuation was affected by the local phase distribution, which was associated with the shape deformation characterized by the Strouhal number. The global first harmonic oscillation was generally much weaker than the global fundamental oscillation. The global first harmonic and fundamental oscillations only had comparable amplitudes when the intense local fundamental oscillations tended to cancel each other out.

Premixed flame response to helical disturbances: Mean flame non-axisymmetry effects

A key component of the thermoacoustic instability feedback mechanism is the excitation of hydrodynamic flow disturbances by narrowband acoustic fluctuations. In earlier work, we considered the response of time-averaged axisymmetric flames to helical disturbances. That study showed that although all helical modes cause local flame wrinkling, only the axisymmetric hydrodynamic mode, m = 0, contributes to the global heat release rate fluctuations. This paper extends this work to consider time-averaged flames that are non-axisymmetric. We show that distortions of the unforced flame shape changes its receptivity to helical disturbances, as it does not allow for the perfect destructive interference seen in axisymmetric flames. This results in a non-zero global flame response contribution from helical modes. Given that helical modes often have the largest amplitudes in jet flows, particularly those with swirl, these results show that the degree of non-axisymmetry of the flame has an important influence in determining which hydrodynamic modes, axisymmetric or helical, control the global heat release response of the flame. These points have been illustrated with example calculations that show how different control parameters and the degree of time-averaged non-axisymmetry influence the global flame response. An important implication of these results relates to scaling results from simplified geometries where a round jet is placed inside a round enclosure, to more realistic ones (such as where multiple jets are placed next to each other) where the flame shape will be distorted from being perfectly circular.

Flame structure and global flame response to the equivalence ratios of interacting partially premixed methane and hydrogen flames

The interacting partially premixed methane and hydrogen flames established in a one-dimensional counterflow field were investigated numerically with the OPPDIF code and GRI-v3.0 was used to consider both fuels. The flame structure and response of the maximum flame temperature, heat-release rate, and flame speed to the equivalence ratios (Φ) and global strain rate (ag) were investigated. The maximum temperature decreased with increasing ag. The maximum temperature for cases with a stoichiometric hydrogen-side flame was higher than for other cases with the same ag.The hydrogen-side flame played a key role in determining the maximum temperature. The maximum heat-release rates (MHRRs) for all cases show different trends. The MHRR of the methane-side flame was affected considerably by the interacting flame structure and hydrogen-side flame condition. However, the MHRRs of the hydrogen were independent of methane-side flame condition. For the cases where Φ of the methane-side flame was varied while the hydrogen-side flame was kept stoichiometric (Var-S), the MHRR and flame speed of the hydrogen-side flame were independent of the methane-side flame conditions. However, the methane-side flames had a negative flame speed except near-stoichiometric conditions. On the other hand, in the cases where Φ of the hydrogen-side flame was varied while the methane-side flame was kept stoichiometric (S-Var), the hydrogen-side flames had the MHRR and flame speed similar to those of an unstretched partially premixed hydrogen flame.

Numerical study of the direct pressure effect of acoustic waves in planar premixed flames

Recently the unsteady response of 1-D premixed flames to acoustic pressure waves for the range of frequencies below and above the inverse of the flame transit time was investigated experimentally using OH chemiluminescence Wangher (2008) [1]. They compared the frequency dependence of the measured response to the prediction of an analytical model proposed by Clavin et al. (1990) [2], derived from the standard flame model (one-step Arrhenius kinetics) and to a similar model proposed by McIntosh (1991) [3]. Discrepancies between the experimental results and the model led to the conclusion that the standard model does not provide an adequate description of the unsteady response of real flames and that it is necessary to investigate more realistic chemical models. Here we follow exactly this suggestion and perform numerical studies of the response of lean methane flames using different reaction mechanisms. We find that the global flame response obtained with both detailed chemistry (GRI3.0 [4]) and a reduced multi-step model by Peters (1996) [5] lies slightly above the predictions of the analytical model, but is close to experimental results. We additionally used an irreversible one-step Arrhenius reaction model and show the effect of the pressure dependence of the global reaction rate in the flame response. Our results suggest first that the current models have to be extended to capture the amplitude and phase results of the detailed mechanisms, and second that the correlation between the heat release and the measured OH ∗ chemiluminescence should be studied deeper.

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.

Role of transport properties in the transient response of premixed methane/air flames

We compute the detailed interaction of a premixed methane/air flame with a two-dimensional counter-rotatingvortex pair using mixture-averaged transport properties for stoichiometric and rich N 2 -diluted reactants. We present comparisons of the observed flame structure and transient flame response with results using temperature-tabulated properties, where dependence on mixture composition is neglected. We find that the significance of using the more accurate transport model depends on the particular aspect of transient flame response under consideration. While the effect on global flame response is generally small, we do observe quantitative differences of as much as 30%–45% in the extent of transient accumulation of some species on the reactants side of the flame. On the other hand, no significant qualitative differences are observed in the transient flame response. We examine and discuss these observations with regard to available experimental data in a premixed methane/air V-flame under the same mixture conditions as the computations and with similar flow-flame evolution. We find that the earlier-reported disagreements between computed and experimentally observed transient OH and CH responses are not much improved with the use of the more accurate mixture-averaged transport. This suggests that the root of the disagreements is not the transport model but the chemical mechanism and/or rate constants therein. Further work is needed to identify necessary improvements in the mechanism to achieve better agreement with the experiments.