Time Domain Simulation of Combustion Instabilities in Annular Combustors

2003 ◽  
Vol 125 (3) ◽  
pp. 677-685 ◽  
Author(s):  
C. Pankiewitz ◽  
T. Sattelmayer

A novel method for the simulation of combustion instabilities in annular combustors is presented. It is based on the idea to solve the equations governing the acoustics in the time domain and couple them to a model for the heat release in the flames. The linear wave equation describing the temporal and spatial evolution of the pressure fluctuations is implemented in a finite element code. Providing high flexibility, this code in principle allows both the computational domain to be of arbitrary shape and the mean flow to be included. This yields applicability to realistic technical combustors. The fluctuating heat release acting as a volume source appears as a source term in the equation to be solved. Employing a time-lag model, the heat release rate at each individual burner is related to the velocity in the corresponding burner at an earlier time. As saturation also is considered, a nonlinearity is introduced into the system. Starting the simulation from a random initial perturbation with suitable values for the parameters of the heat release model, a self-excited instability is induced, leading to a finite-amplitude limit cycle oscillation. The feasibility of the approach is demonstrated with three-dimensional simulations of a simple model annular combustor. The effect of the model parameters and of axial mean flow on the stability and the shape of the excited modes is shown.


Author(s):  
Christian Pankiewitz ◽  
Thomas Sattelmayer

A novel method for the simulation of combustion instabilities in annular combustors is presented. It is based on the idea to solve the equations governing the acoustics in the time domain and couple them to a model for the heat release in the flames. The linear wave equation describing the temporal and spatial evolution of the pressure fluctuations is implemented in a finite element code. Providing high flexibility, this code in principle allows both the computational domain to be of arbitrary shape and the mean flow to be included. This yields applicability to realistic technical combustors. The fluctuating heat release acting as a volume source appears as a source term in the equation to be solved. Employing a time-lag model, the heat release rate at each individual burner is related to the velocity in the corresponding burner at an earlier time. As saturation also is considered, a non-linearity is introduced into the system. Starting the simulation from a random initial perturbation with suitable values for the parameters of the heat release model, a self-excited instability is induced, leading to a finite-amplitude limit cycle oscillation. The feasiblity of the approach is demonstrated with 3D-simulations of a simple model annular combustor. The effect of the model parameters and of axial mean flow on the stability and the shape of the excited modes is shown.



Author(s):  
Simon R. Stow ◽  
Ann P. Dowling

Lean premixed prevaporised (LPP) combustion can reduce NOx emissions from gas turbines, but often leads to combustion instability. Acoustic waves produce fluctuations in heat release, for instance by perturbing the fuel-air ratio. These heat fluctuations will in turn generate more acoustic waves and in some situations linear oscillations grow into large amplitude self-sustained oscillations. The resulting limit cycles can cause structural damage. Thermoacoustic oscillations will have a low amplitude initially. Thus linear models can describe the initial growth and hence give stability predictions. An unstable linear mode will grow in amplitude until nonlinear effects become sufficiently important to achieve a limit cycle. While the frequency of the linear mode can often provide a good approximation to that of the resulting limit cycle, linear theories give no prediction of its resulting amplitude. In previous work, we developed a low-order frequency-domain method to model thermoacoustic limit cycles in LPP combustors. This was based on a ‘describing function’ approach and is only applicable when there is a dominant mode and the main nonlinearity is in the combustion response to flow perturbations. In this paper that method is extended into the time domain. The main advantage of the time-domain approach is that limit-cycle stability, the influence of harmonics, and the interaction between different modes can be simulated. In LPP combustion, fluctuations in the inlet fuel-air ratio have been shown to be the dominant cause of unsteady combustion: these occur because velocity perturbations in the premix ducts cause a time-varying fuel-air ratio, which then convects downstream. If the velocity perturbation becomes comparable to the mean flow, there will be an amplitude-dependent effect on the equivalence ratio fluctuations entering the combustor and hence on the rate of heat release. Since the Mach number is low, the velocity perturbation can be comparable to the mean flow, with even reverse flow occurring, while the disturbances are still acoustically linear in that the pressure perturbation is still much smaller than the mean. Hence while the combustion response to flow velocity and equivalence ratio fluctuations must be modelled nonlinearly, the flow perturbations generated as a result of the unsteady combustion can be treated as linear. In developing a time-domain network model for nonlinear thermoacoustic oscillations an initial frequency-domain calculation is performed. The linear network model, LOTAN, is used to categorise the combustor geometry by finding the transfer function for the response of flow perturbations (at the fuel injectors, say) to heat-release oscillations. This transfer function is then converted into the time domain through an inverse Fourier transform to obtain the Green’s function, which thus relates unsteady flow to heat release at previous times. By combining this with a nonlinear flame model (relating heat release to unsteady flow at previous times) a complete time-domain solution can be found by stepping forward in time. If an unstable mode is present, its amplitude will initially grow exponentially (in accordance with linear theory) until saturation effects in the flame model become significant, and eventually a stable limit cycle will be attained. The time-domain approach enables determination of the limit-cycle. In addition, the influence of harmonics and the interaction and exchange of energy between different modes can be simulated. These effects are investigated for longitudinal and circumferential instabilities in an example combustor system and results are compared to frequency-domain limit-cycle predictions.



2021 ◽  
pp. 1-26
Author(s):  
L. Yuan ◽  
C. Shen

Abstract On the basis of air heater characteristics, a new computational model was developed in this paper, which was aimed at investigating acoustics and instabilities in air heaters. This model included the effects of mean flow, viscosity, entropy waves, non-linear acoustics and realistic boundary conditions. In addition, it was practical for air heaters with hundreds of injectors, complex configurations and geometries. Analytical solutions of acoustics in a closed rectangular cavity were used to verify and validate the computational model. It was shown that the predicted critical parameters of air heater agreed well with the experimental data or design values. This model predicted the self-excited spinning tangential modes without any preliminary assumptions about them. Traditional combustion response function assumes that combustion mainly takes place in a so-called rapid combustion zone, and this zone is usually modelled as a disc in the combustor near the injection head. However, in practice, the flame has a spatial distribution. This paper described the effect of flame spatial distribution on predictions of oscillation frequency and mode. It was found that frequency and mode shape of oscillations closely depended on the length of the heat release zone. Comparison of different heat release zones indicated that the increment of heat release length exhibited an increased tendency toward lower-order longitudinal modes, when the heat release zone was located near the faceplate where it was the pressure antinode of the longitudinal mode. Modulation of heat release length might cause bifurcations between standing and turning modes. A noticeable tendency was toward higher-order standing tangential mode with increasing heat release length. Finally, theoretical analysis of the modal behaviour, i.e. standing or spinning waves, was performed.



2020 ◽  
Vol 12 ◽  
pp. 175682772093049
Author(s):  
Sreenath M Gopinathan ◽  
Alessandra Bigongiari ◽  
Maria Heckl

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.



Author(s):  
Simon R. Stow ◽  
Ann P. Dowling

Lean premixed prevaporized (LPP) combustion can reduce NOx emissions from gas turbines but often leads to combustion instability. Acoustic waves produce fluctuations in heat release, for instance, by perturbing the fuel-air ratio. These heat fluctuations will in turn generate more acoustic waves and in some situations linear oscillations grow into large-amplitude self-sustained oscillations. The resulting limit cycles can cause structural damage. Thermoacoustic oscillations will have a low amplitude initially. Thus linear models can describe the initial growth and hence give stability predictions. An unstable linear mode will grow in amplitude until nonlinear effects become sufficiently important to achieve a limit cycle. While the frequency of the linear mode can often provide a good approximation to that of the resulting limit cycle, linear theories give no prediction of its resulting amplitude. In previous work, we developed a low-order frequency-domain method to model thermoacoustic limit cycles in LPP combustors. This was based on a “describing-function” approach and is only applicable when there is a dominant mode and the main nonlinearity is in the combustion response to flow perturbations. In this paper that method is extended into the time domain. The main advantage of the time-domain approach is that limit-cycle stability, the influence of harmonics, and the interaction between different modes can be simulated. In LPP combustion, fluctuations in the inlet fuel-air ratio have been shown to be the dominant cause of unsteady combustion: These occur because velocity perturbations in the premix ducts cause a time-varying fuel-air ratio, which then convects downstream. If the velocity perturbation becomes comparable to the mean flow, there will be an amplitude-dependent effect on the equivalence ratio fluctuations entering the combustor and hence on the rate of heat release. Since the Mach number is low, the velocity perturbation can be comparable to the mean flow, with even reverse flow occurring, while the disturbances are still acoustically linear in that the pressure perturbation is still much smaller than the mean. Hence while the combustion response to flow velocity and equivalence ratio fluctuations must be modeled nonlinearly, the flow perturbations generated as a result of the unsteady combustion can be treated as linear. In developing a time-domain network model for nonlinear thermoacoustic oscillations an initial frequency-domain calculation is performed. The linear network model, LOTAN, is used to categorize the combustor geometry by finding the transfer function for the response of flow perturbations (at the fuel injectors, say) to heat-release oscillations. This transfer function is then converted into the time domain through an inverse Fourier transform to obtain Green’s function, which thus relates unsteady flow to heat release at previous times. By combining this with a nonlinear flame model (relating heat release to unsteady flow at previous times) a complete time-domain solution can be found by stepping forward in time. If an unstable mode is present, its amplitude will initially grow exponentially (in accordance with linear theory) until saturation effects in the flame model become significant, and eventually a stable limit cycle will be attained. The time-domain approach enables determination of the limit cycle. In addition, the influence of harmonics and the interaction and exchange of energy between different modes can be simulated. These effects are investigated for longitudinal and circumferential instabilities in an example combustor system and the results are compared with frequency-domain limit-cycle predictions.



1987 ◽  
Vol 52 (8) ◽  
pp. 1888-1904
Author(s):  
Miloslav Hošťálek ◽  
Ivan Fořt

A theoretical model is described of the mean two-dimensional flow of homogeneous charge in a flat-bottomed cylindrical tank with radial baffles and six-blade turbine disc impeller. The model starts from the concept of vorticity transport in the bulk of vortex liquid flow through the mechanism of eddy diffusion characterized by a constant value of turbulent (eddy) viscosity. The result of solution of the equation which is analogous to the Stokes simplification of equations of motion for creeping flow is the description of field of the stream function and of the axial and radial velocity components of mean flow in the whole charge. The results of modelling are compared with the experimental and theoretical data published by different authors, a good qualitative and quantitative agreement being stated. Advantage of the model proposed is a very simple schematization of the system volume necessary to introduce the boundary conditions (only the parts above the impeller plane of symmetry and below it are distinguished), the explicit character of the model with respect to the model parameters (model lucidity, low demands on the capacity of computer), and, in the end, the possibility to modify the given model by changing boundary conditions even for another agitating set-up with radially-axial character of flow.



Author(s):  
Xiaofeng Yang ◽  
Saurabh Gupta ◽  
Tang-Wei Kuo ◽  
Venkatesh Gopalakrishnan

A comparative cold flow analysis between Reynolds-averaged Navier–Stokes (RANS) and large eddy simulation (LES) cycle-averaged velocity and turbulence predictions is carried out for a single cylinder engine with a transparent combustion chamber (TCC) under motored conditions using high-speed particle image velocimetry (PIV) measurements as the reference data. Simulations are done using a commercial computationally fluid dynamics (CFD) code CONVERGE with the implementation of standard k-ε and RNG k-ε turbulent models for RANS and a one-equation eddy viscosity model for LES. The following aspects are analyzed in this study: The effects of computational domain geometry (with or without intake and exhaust plenums) on mean flow and turbulence predictions for both LES and RANS simulations. And comparison of LES versus RANS simulations in terms of their capability to predict mean flow and turbulence. Both RANS and LES full and partial geometry simulations are able to capture the overall mean flow trends qualitatively; but the intake jet structure, velocity magnitudes, turbulence magnitudes, and its distribution are more accurately predicted by LES full geometry simulations. The guideline therefore for CFD engineers is that RANS partial geometry simulations (computationally least expensive) with a RNG k-ε turbulent model and one cycle or more are good enough for capturing overall qualitative flow trends for the engineering applications. However, if one is interested in getting reasonably accurate estimates of velocity magnitudes, flow structures, turbulence magnitudes, and its distribution, they must resort to LES simulations. Furthermore, to get the most accurate turbulence distributions, one must consider running LES full geometry simulations.



2021 ◽  
pp. 1-52
Author(s):  
Gregory Tierney ◽  
Walter A. Robinson ◽  
Gary Lackmann ◽  
Rebecca Miller

AbstractHigh-impact events such as heat waves and droughts are often associated with persistent positive geopotential height anomalies (PAs). Understanding how PA activity will change in a future warmer climate is therefore fundamental to projecting associated changes in weather and climate extremes. This is a complex problem because the dynamics of PAs and their associated blocking activity are still poorly understood. Furthermore, climate-change influences on PA activity may be geographically dependent and encompass competing influences. To expose the salient impacts of climate change, we use an oceanic channel configuration of the Weather Research and Forecasting model (WRF) in a bivariate experiment focused on changes in environmental temperature, moisture, and baroclinicity. The 500-hPa wind speed and flow variability are found to increase with increasing temperature and baroclinicity, driven by increases in latent heat release and a stronger virtual temperature gradient. Changes to 500-hPa sinuosity are negligible. PAs are objectively identified at the 500-hPa level using an anomaly threshold method. When using a fixed threshold, PA trends indicate increased activity and strength with warming, but decreased activity and strength with Arctic amplification. Use of a climate-relative threshold hides these trends and highlights the importance of accurate characterization of the mean flow. Changes in PA activity mirror corresponding changes in 500-hPa flow variability and are found to be attributable to changes in three distinct dynamical mechanisms: baroclinic wave activity, virtual temperature effects, and latent heat release.



1976 ◽  
Vol 15 (2) ◽  
pp. 197-222
Author(s):  
R. J. Hartman

This paper uses the general solution of the linearized initial-value problem for an unbounded, exponentially-stratified, perfectly-conducting Couette flow in the presence of a uniform magnetic field to study the development of localized wave-type perturbations to the basic flow. The two-dimensional problem is shown to be stable for all hydrodynamic Richardson numbers JH, positive and negative, and wave packets in this flow are shown to approach, asymptotically, a level in the fluid (the ‘isolation level’) which is a smooth, continuous, function of JH that is well defined for JH < 0 as well as JH > 0. This system exhibits a rich complement of wave phenomena and a variety of mechanisms for the transport of mean flow kinetic and potential energy, via linear wave processes, between widely-separated regions of fluid; this in addition to the usual mechanisms for the absorption of the initial wave energy itself. The appropriate three-dimensional system is discussed, and the role of nonlinearities on the development of localized disturbances is considered.



Sign in / Sign up

Export Citation Format

Share Document