Spectral analysis and modelling of the spray liquid injection in a Lean Direct Injection (LDI) gas turbine combustor through Eulerian-Lagrangian Large Eddy Simulations

Abstract

The main challenge of next-generation aeronautical gas turbine engines lies in the increase of the efficiency of the cycle and the reduction of pollutant emissions below stringent restrictions. This has led to the design of new injection-combustion strategies working on more risky and problematic operating points such as those close to the lean extinction limit. In This context, the Lean Direct Injection (LDI) concept has emerged as a promising technology to reduce NOx for future aircraft power plants. The study of liquid fuel injection, atomization, evaporation and later interaction with air by means of a numerical approach is deemed to provide a detailed description of these phenomena affecting the overall engine cycle efficiency and emissions. In this context, the aim of this research is to use Large Eddy Simulation for the characterisation of the structure of a liquid non-reacting spray immersed in a strong swirling field in the CORIA Spray LDI burner. An Eulerian formulation is considered for the continuous phase and is coupled with a Lagrangian description for the dispersed phase. A precise description of the fuel droplet size distribution and size-classified velocity, as well as the characterisation of the instantaneous, mean and fluctuating air velocities is presented and compared to the available experimental data. The fuel spray model is shown to accurately reproduce the computed Sauter Mean Diameter (SMD) and the velocity of the droplets. Moreover, the main flow structures generated within the combustor (e.g., Precessing Vortex Core, Vortex Breakdown Bubble, recirculation zones, etc.), which play a crucial role in the fuel-air mixing process, are quantitatively characterised through advanced frequency analysis such as Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD). The characteristic swirling frequency of a single-branched PVC presenting two different phase-shifted POD modes with the same associated spectrum is detected. Finally, POD and DMD techniques are also applied to the numerical spray data to further investigate the spray-turbulence interactions inside the combustion chamber. In ths regard, DMD analysis has confirmed how the swirl-acoustic interactions led in the VBB and PVC oscillations play a crucial role in the way the fuel spray is internally forced by the PVC wavemaker travelling downstream the swirler and synchronised with its dominant frequency. This will contribute to a better understanding of the mixing quality and local equivalence ratios before the subsequent ignition process.