Thermoacoustic Stability of Annular Aero-Engine Combustors

Motivation

High amplitude combustion instabilities are frequently encountered during the development of combustion chambers. These instabilities can increase the amplitude of flame motion and hence damage the combustor due to large heat transfer rates or limit its operating conditions. To avoid costly corrections at later stages, predictive methods providing stability analysis at the design level are requested.

Figure 1: 1st longitudinal, 1st and 2nd circumferential modes in an annular aero-engine combustor

Especially low-NOx combustion systems working in the lean regime are more susceptible to these instabilities. This is owed to the fact that more air is needed to burn the amount of fuel injected and thus can lead to severe equivalence ratio fluctuations. However, this technology also yields considerably high savings in NOx-emissions and has urged the air industry to operate their combustors under leaner conditions.

The most critical instabilities appear in the low and medium frequency range. Fig. 1 exemplary shows the first three mode structures of an annular aero-engine combustor. Their length scales are typically of the order of the confining geometry. Hence, to correctly predict their frequencies, full annular combustor geometries have to be considered. These can be of considerable dimensions making Large-Eddy simulations a costly task for the near future. Their frequency and spatial character also essentially depend on the acoustic flame interaction and acoustic boundary conditions. A stability assessment methodology must be capable of taking all these parameters into account.

Project goals

The major objective of this research project is to provide methodologies to better understand combustion instabilities and to assess their unstable/stable character. Therefore, a hybrid approach is proposed: The flame response, when subjected to acoustic perturbations, has a high impact on the accuracy of stability prediction. This so called flame transfer function will be measured within experiments and will then be reformulated into a model. In a second step, this model can then be implemented in an acoustic code taking boundary conditions and the exact geometry of an industrial combustor into account. With this methodology predict stability and thermoacoustic modes of a combustion system.

Experimental investigations

Yet, the underlying mechanisms and in particular the acoustic flame coupling are not fully understood. Nevertheless, the flame behavior when subjected to acoustic forcing, can be accurately described through experiments. Here, the Chair of Thermodynamics has gathered a broad knowledge and experience in the recent years on the fields of premixed and partially premixed flames. This work will be extended to spray flames, which are typically utillized in aero-engine combustors. Measurements will be conducted at an atmospheric pressure rig at the Chair and shall deliver a methodology for spray flame measurements under more realistic conditions.

Numerical work

As the flame answer to acoustic perturbations is known from experiments it can be embedded via a model into an aero-acoustic code. This code will take the complex boundaries and the exact geometry of an industrial combustion system into account. To solve the acoustic field in a defined geometry, codes operating in time and frequency domain have delivered promising results. Linearized Euler and the wave equation are used to assess the stability of a given configuration.

Acknowledgements

The author greatfully acknowledge the financial support provided by the European Union in the 7th framework of the project KIAI (''Knowledge for Ignition, Acoustics and Instabilities'' ) Grant Agreement No. ACP8-GA-2009-234009.