Nonlinear combustion instability analysis of a bluff body burner based on the flame describing function

Lean premixed combustion is a common form of combustion organization in power equipment and propulsion systems. In order to understand the dynamic characteristics of lean premixed flame and predict and control its combustion instability, it is necessary to obtain its flame describing function (FDF). Based on the open source CFD toolbox, OpenFOAM, the dynamic K-equation model, and the finite rate Partially Stirred Reactor (PaSR) model were used to perform large eddy simulations (LES) of lean premixed combustion, and the response of the unsteady heat release rate to single-frequency harmonic disturbances was studied. The response of the unsteady heat release rate was characterized by the FDF, and the response of the unsteady heat release rate to the two-frequency harmonic disturbance was studied. The results show that the quantitative heat release rate response and flame dynamics have very proper accuracy. In the single-frequency harmonic disturbance, as the forcing frequency increases, the curling behavior of the flame surface and the instantaneous vortex structure change; the nonlinear kinematics effect is manifested by the entrainment of the vortex. At lower forcing frequencies, the heat release response changes linearly with the increase of forcing amplitude; at intermediate frequencies, the heat release response exhibits obvious nonlinear behavior; at high frequencies, the heat release response to amplitude changes decreases. The introduction of the second harmonic disturbance will significantly reduce the response range of the total heat release rate and make the combustion more stable.

Acoustic Energy Balance During the Onset, Growth and Saturation of Thermoacoustic Instabilities

Thermoacoustic instabilities can occur in a wide range of combustors and are prejudicial since they can lead to increased mechanical fatigue or even catastrophic failure. A well-established formalism to predict the onset, growth and saturation of such instabilities is based on acoustic network models. This approach has been successfully employed to predict the frequency and amplitude of limit cycle oscillations in a variety of combustors. However, it does not provide any physical insight in terms of the acoustic energy balance of the system. On the other hand, Rayleigh's criterion may be used to quantify the losses, sources and transfers of acoustic energy within and at the boundaries of a combustor. However, this approach is cumbersome for most applications because it requires computing volume and surface integrals and averaging over an oscillation cycle. In this work, a new methodology for studying the acoustic energy balance of a combustor during the onset, growth and saturation of thermoacoustic instabilities is proposed. The two cornerstones of this new framework are the acoustic absorption coefficient Delta and the cycle-to-cycle acoustic energy ratio lambda, both of which do not require computing integrals. Used along with a suitable acoustic network model, where the flame frequency response is described using the weakly nonlinear Flame Describing Function (FDF) formalism, these two dimensionless numbers are shown to characterize: 1) the variation of acoustic energy stored within the combustor between two consecutive cycles (rest of the abstract in the article).

Acoustic Energy Balance During the Onset, Growth and Saturation of Thermoacoustic Instabilities

Thermoacoustic instabilities can occur in a wide range of combustors and are prejudicial since they can lead to increased mechanical fatigue or even catastrophic failure. A well-established formalism to predict the onset, growth and saturation of such instabilities is based on acoustic network models. This approach has been successfully employed to predict the frequency and amplitude of limit cycle oscillations in a variety of combustors. However, it does not provide any physical insight in terms of the acoustic energy balance of the system. On the other hand, Rayleigh’s criterion may be used to quantify the losses, sources and transfers of acoustic energy within and at the boundaries of a combustor. However, this approach is cumbersome for most applications because it requires computing volume and surface integrals and averaging over an oscillation cycle. In this work, a new methodology for studying the acoustic energy balance of a combustor during the onset, growth and saturation of thermoacoustic instabilities is proposed. The two cornerstones of this new framework are the acoustic absorption coefficient Δ and the cycle-to-cycle acoustic energy ratio λ, both of which do not require computing integrals. Used along with a suitable acoustic network model, where the flame frequency response is described using the weakly nonlinear Flame Describing Function (FDF) formalism, these two dimensionless numbers are shown to characterize: 1) the variation of acoustic energy stored within the combustor between two consecutive cycles, 2) the acoustic energy transfers occurring at the combustor’s boundaries and 3) the sources and sinks of acoustic energy located within the combustor. The acoustic energy balance of the well-documented Palies burner is then analyzed during the onset, growth and saturation of thermoacoustic instabilities using this new methodology. It is demonstrated that this new approach allows a deeper understanding of the physical mechanisms at play. For instance, it is possible to determine when the flame acts as an acoustic energy source or sink, where acoustic damping is generated, and if acoustic energy is transmitted through the boundaries of the burner.

Analytical approximations for heat release rate laws in the time- and frequency-domains

This paper focusses on the relationship between the heat release rate and the acoustic field, which is a crucial element in modelling thermoacoustic instabilities. The aim of the paper is twofold. The first aim is to develop a transformation tool, which makes it easy to switch between the time-domain representation (typically a heat release law involving time-lags) and the frequency-domain representation (typically a flame transfer function) of this relationship. Both representations are characterised by the same set of parameters n1, n2, …, nk. Their number is quite small, and they have a clear physical meaning: they are time-lag dependent coupling coefficients. They are closely linked to the impulse response of the flame in the linear regime in that they are proportional to the discretised (with respect to time) impulse response. In the nonlinear regime, the parameters n1, n2, …, nk become amplitude-dependent. Their interpretation as time-lag dependent coupling coefficients prevails; however, the link with the impulse response is lost. Nonlinear flames are commonly described in the frequency-domain by an amplitude-dependent flame transfer function, the so-called flame describing function. The time-domain equivalent of the flame describing function is sometimes mistaken for a ‘nonlinear impulse response’, but this is not correct. The second aim of this paper is to highlight this misconception and to provide the correct interpretation of the time-domain equivalent of the flame describing function.

Dynamic Response of Acoustically Forced Turbulent Premixed Biogas Flames

Increasing demand for energy and the need for diversification of fuels used in gas turbine power generation is continuing to drive forward the development of fuel-flexible combustion systems, with particular focus on biomass derived sustainable fuels. The technical challenges arising from burning sustainable fuels are largely associated with the change in the chemical, thermal and transport properties of these fuels due to the variation of the constituents and their impact on the performance of the combustor including emissions, static and dynamic stabilities. There is a lack of detailed understanding on the effect of fuel composition on the flame sensitivity to acoustic and flow perturbations. This paper describes an experimental study investigating the acoustic flame response of simulated biogas (methane/carbon dioxide/air mixtures) turbulent premixed flames. The effect of variation in carbondioxide, CO2, content on the flame response was quantified. Special emphasis was placed on understanding the dependence of this flame response on the amplitude of the acoustic forcing. The flame was subjected to strong velocity perturbations using loud speakers. It was observed that the addition of CO2 had considerable influence on the magnitude of heat release response. The magnitude and the phase of flame describing function indicated that the mechanism of saturation in these flames for all conditions tested were the same. The difference in magnitude could been attributed to dilution effect and hence further investigation were carried out with N2 and Ar to clarify the role of CO2. The results indicate that the thermal capacity of the diluent gases could be playing a significant role in nonlinear flame dynamics.

Flame Describing Functions of a Confined Premixed Swirled Combustor With Upstream and Downstream Forcing

The frequency response of a confined premixed swirled flame is explored experimentally through the use of describing functions that depend on both the forcing frequency and the forcing level. In these experiments, the flame is forced by a loudspeaker connected to the bottom of the burner in the fresh gas region or by a set of loudspeakers connected to the combustion chamber exhaust tube in the burnt gas region. The experimental setup is equipped with a hot-wire (HW) probe and a microphone, both of which located in front of each other below the swirler. The forcing level is varied between |v′0|/v¯0=0.10 and 0.72 RMS, where v¯0 and v′0 are, respectively, the mean and the fluctuating velocity at the HW probe. An additional microphone is placed on a water-cooled waveguide connected to the combustion chamber backplate. A photomultiplier equipped with an OH* filter is used to measure the heat release rate fluctuations. The describing functions between the photomultiplier signal and the different pressure and velocity reference signals are then analyzed in the case of upstream and downstream forcing. The describing function measured for a given reference signal is shown to vary depending on the type of forcing. The impedance of the injector at the HW location is also determined for both upstream and downstream forcing. For all describing functions investigated, it is found that their phase lags do not depend on the forcing level, whereas their gains strongly depend on |v′0|/v¯0 for certain frequency ranges. It is furthermore shown that the flame describing function (FDF) measured with respect to the HW signal can be retrieved from the specific impedance at the HW location and the describing function determined with respect to the signal of the microphone located in front of the HW. This relationship is not valid when the signal from the microphone located at the combustion chamber backplate is considered. It is then shown that a one-dimensional (1D) acoustic model allows to reproduce the describing function computed with respect to the microphone signal inside the injector from the microphone signal located at the combustion chamber backplate in the case of downstream forcing. This relation does not hold for upstream forcing because of the acoustic dissipation across the swirler which is much larger compared to downstream forcing for a given forcing level set at the HW location. This study sheds light on the differences between upstream and downstream acoustic forcing when measuring describing functions. It is also shown that the upstream and downstream forcing techniques are equivalent only if the reference signal used to determine the FDF is the acoustic velocity in the fresh gases just before the flame.