Combustion Dynamics in a Single-Element Lean Direct Injection Gas Turbine Combustor

ABSTRACT

A concurrent computational and experimental study of self-excited combustion dynamics in a model configuration of a lean direct injection (LDI) gas turbine combustor is described. Both acoustically open and closed configurations are considered in the computations and analyzed using dynamic mode decomposition (DMD) to identify the frequency couplings. In the acoustically open case, simulations are carried out with unperturbed and perturbed inlet mass flow rates at the dominant frequencies observed in the acoustically closed geometry. In the unperturbed case, a vortex breakdown bubble (VBB) is shown to be the dominant flow structure, while the perturbed simulations indicate the presence of swirling hydrodynamic modes and the possibility of coupling between hydrodynamics and acoustics as a mechanism for sustaining thermo-acoustic instabilities. Detailed analysis of the acoustically closed combustor simulation reveals the presence of another important hydrodynamic mode, the precessing vortex core (PVC). The possibility of nonlinear coupling between the acoustics and PVC modes is also revealed. The thermo-acoustic coupling is further analyzed by the Rayleigh index frequency spectrum.

KEYWORDS:

• Combustion instability
• lean direct injection
• swirl stabilized flames

Acknowledgments

The authors acknowledge the support of the NASA Glenn Research Center under NASA Research Announcement (NRA) grant number NNX11AI62A, with Technical Monitor Mr. Kevin Breisacher and Program Manager Ms Julie Fowler. The first author would also like to acknowledge the financial support from John Zink Company. We would like to give special thanks to Dr. Charles Merkle who helped set the original direction of the study, Dr. Phil Lee of Woodward for providing the fuel nozzle used in the experiment, and Prof. Hukam Mongia of Purdue and Dr. Clarence Chang of NASA Glenn Research Center for their technical advice.

Additional information

Funding

This work was supported by the Glenn Research Center [NNX11AI62A];