Noise Prediction of Multi-Stream Internally Mixed Jets With External and Internal Pligs Using Large-Eddy Simulation

Results from ongoing research on jet noise prediction using wall-modeled large eddy simulations (WMLES) conducted within the Launch Ascent and Vehicle Aerodynamics (LAVA) computational framework are presented. The study primarily focuses on the aeroacoustic implications of multi-stream nozzle configurations with internal mixing generated by the addition of both internal and external plug geometries. While multi-stream nozzles with internally mixed jets have long been considered for reducing jet noise, the complexities of their impact on overall noise levels remain insufficiently understood. This research applies established best practices for WMLES to these complex nozzle configurations, aiming to assess their efficacy and identify any remaining accuracy limitations in predicting far-field noise characteristics. Through detailed comparisons with experimental data obtained from NASA's Glenn Research Center, initial findings underscore the challenges inherent in current simulation practices when confronted with intricate geometric and operational conditions. Responding to these challenges, the study explores innovative computational approaches to rectify discrepancies noted between experimental measurements and CFD predictions. One example is the introduction of a resonating sound source in the simulation environment to mimic potential model vibrations which were not measured in the experiment, but may exist. A second example is the strategic modifications to the geometry of external plugs to account for real-world deformations caused by heating and gravity. These novel strategies aim to enhance the accuracy and reliability of noise predictions from supersonic jets, advancing our understanding and capability to effectively reduce jet noise in commercial supersonic aircraft.