Spectral analysis of soot dynamics in an aero-engine model combustor

The dynamics of soot evolution in a swirl-stabilized model aero-engine combustor are studied using Large-Eddy Simulation (LES) and state-of-the-art combustion and soot models. The simulated combustor is a dual-swirl combustor with and without secondary oxidation air at a pressure of 3 bar. The comparison with experimental data shows that the simulation accurately captures the gas-phase statistics. Overall, soot statistics are well captured, particularly in the shear layers, although underpredicted in other locations. Three locations in the combustor: Shear Layer (SL), Inner Recirculation Zone (IRZ), and Outer Recirculation Zone (ORZ) are probed to obtain velocity, temperature, and soot volume fraction signals. Lomb–Scargle spectral analysis on the probed signals reveals that soot evolution in the SL is characterized by high-frequency dynamics, whereas in the IRZ and ORZ, it is characterized by low-frequency dynamics. Within the SL, couplings between the soot and flow dynamics are observed, with the soot volume fraction and velocity sharing a common dominant frequency. However, in the IRZ and ORZ, such couplings are not evident. Additionally, the wavelet transform is applied to the probed signals to study the temporal distribution of the frequencies. The analysis indicates that, in the SL, the dominant frequency of the velocity occurs continuously throughout the entire time series, while higher frequencies occur in short bursts. Conversely, for the soot volume fraction, both the dominant frequency and higher frequencies appear in short bursts. In the ORZ and IRZ, the soot volume fraction scalogram shows that low frequencies dominate and occur continuously throughout the time series. Finally, the phase space reconstructions show that the trajectory of the soot dynamics shows a circular pattern, indicative of periodic behavior. The center of attraction remains stationary over a relatively large time scale, suggesting stability in the dynamics as they evolve.