No abstract

Combustion oscillations that arise in gas turbines can lead to plant damage. One method used to predict these oscillations is to analyse the acoustics using a simple linear model. This model requires a transfer function to describe the response of the heat release to flow perturbations. A transfer function has been obtained for a swirled premixed combustion system using experiments under atmospheric conditions and CFD. These results have been compared with analytical models.The experimental and computational transfer functions both indicate a low frequency zero. A time-delay spread model gives a good representation of the computational transfer function. The experimental transfer function is described well by a model that combines a time-delay spread with a constant gain.

Flame-turbulence interactions are at the heart of modern combustion research as they have a major influence on efficiency, stability of operation and pollutant emissions. The problem remains a formidable challenge, and predictive modelling and the implementation of active control measures both rely on further fundamental measurements. Model burners with simple geometry offer an opportunity for the isolation and detailed study of phenomena that take place in real-world combustors, in an environment conducive to the application of advanced laser diagnostic tools. Lean premixed combustion conditions are currently of greatest interest since these are able to provide low NO (x) and improved increased fuel economy, which in turn leads to lower CO(2) emissions. This paper presents an experimental investigation of the response of a bluff-body-stabilised flame to periodic inlet fluctuations under lean premixed turbulent conditions. Inlet velocity fluctuations were imposed acoustically using loudspeakers. Spatially resolved heat release rate imaging measurements, using simultaneous planar laser-induced fluorescence (PLIF) of OH and CH(2)O, have been performed to explore the periodic heat release rate response to various acoustic forcing amplitudes and frequencies. For the first time we use this method to evaluate flame transfer functions and we compare these results with chemiluminescence measurements. Qualitative thermometry based on two-line OH PLIF was also used to compare the periodic temperature distribution around the flame with the periodic fluctuation of local heat release rate during acoustic forcing cycles

Gas turbines which are operated under lean, premixed, pre–vaporised (LPP) conditions are notoriously susceptible to self–excited oscillations. In the combustion chamber the unsteady heat released by combustion processes interacts with pressure fluctuations. The challenge is to develop a tool which can determine the frequency and stability characteristics of self–excited oscillations in realistic gas–turbine geometries. To this end, the flow through the gas turbine is described as far as possible by taking advantage of linearised theory and analytical models of the behaviour in the combustion chamber. First, a steady, mean flow solution for an idealised axi–symmetric combustor geometry is calculated using the inviscid Euler equations for continuity, momentum and energy with a specified distributed mean heat release. Superimposed on this is a linearised, three–dimensional perturbed flow in which the time and circumferential variation are described by a complex frequency and mode number respectively. Within this numerical model of the combustor a ‘flame model’ is used to describe the change in the rate of combustion due to inlet flow perturbations. The flame model may be given by an analytical expression—for example using a simple time lag with an expression proportional to the mean heat release in order to describe the unsteady heat release. An alternative approach would be to use a localised and detailed unsteady CFD calculation to determine the flow downstream of a generic premix duct geometry. If the flow is perturbed at the inlet a relationship between these fluctuations and the unsteady heat release may be obtained. In order to capture the response of the system to a wide frequency range an appropriately chosen broad–band forcing function may be used to perturb the flow. System identification techniques allow the transfer function to be extracted and a suitable flame model for the linearised Euler calculations may be constructed. Sample calculations of each aspect of the research will be presented to demonstrate the capabilities of each technique and the viability of combining the approaches towards the goal of aiding the design of gas–turbine combustors. Calculations using the linearised Euler methodology with analytical expressions for the flame model will demonstrate the capability of the approach to identify the frequencies of oscillation, mode shapes and zones of stability of particular combustor geometries.