Comprehensive Modeling of the Cause-and-Effect Chain in Aero-Engine Combustor Simulations: from Primary Breakup to Soot Formation

Kurzfassung

Due to the significant environmental and health impacts, minimizing pollutant emissions, especially soot, is a critical challenge in developing next-generation aero-engines. While predictive soot models in computational fluid dynamics (CFD) are essential for reducing development time and cost, the full simulation of the entire process—from fuel injection and atomization to soot formation and evolution—remains challenging and often involves strong modeling assumptions. To address this challenge, this study combines smoothed particle hydrodynamics (SPH), used to predict liquid fuel atomization, with finite volume method (FVM) large eddy simulations (LES) with advanced combustion and soot models. This approach allows for consistent simulations from fuel breakup to soot formation and enables a detailed investigation of the complex interactions between spray dynamics and soot under engine-like conditions. To accurately capture the primary breakup, the fuel spray particle size distribution (PSD) is sampled from SPH simulations and used to initialize Lagrangian spray particles in the LES, where secondary breakup and evaporation are predicted. The objective of this work is to apply these methods to a single-sector aero-engine combustion chamber operated at elevated pressure and high preheating temperatures, with an aero-engine fuel injector geometry, and to investigate the influence of spray dynamics on soot formation. Comparison with experimental data demonstrates that the applied methods accurately capture the overall flow and combustion characteristics. Spray characteristics sampled from SPH simulations significantly improve the accuracy of mixing and soot formation predictions compared to conventional spray representation approaches. Furthermore, an extended analysis across various operating ranges demonstrates that spray initializations tailored to the respective conditions are essential for achieving accurate pollutant predictions.