Description: Installation effects arising from propulsion airframe interaction are known to produce substantial variations in the in-situ jet noise. A hybrid LES/RANS computational framework is proposed for prediction of noise from the engine and airframe, and interactions between airframe and propulsion systems. The basis of LES (large eddy simulation) is that the energy-bearing turbulent eddies in the dominant noise-generating region are directly captured in the simulation. Since LES must resolve the turbulent eddies it requires a grid which captures these motions; the number of grid points needed for LES is much larger than those for RANS and thus a brute-force LES of the entire noise producing region in a propulsion-airframe interaction problem is not feasible. However, the noise generation physics of these flows allows a logical assembly of a hybrid simulation tool where low-fidelity models (RANS) in one region of the flow are combined with turbulence-resolving models (LES) in other regions of the flow. Acoustic effects are another segment of propulsion-airframe interaction problem. Sound generated by various components of engine is altered by the presence of wing, fuselage, deployed flap etc. In the present proposal, alteration of sound due to the presence of airframe is added through application of Boundary Element Method (BEM) and an acoustic projection technique (FW-H surface method). To demonstrate the feasibility of using this framework, we focus on simulating flow configuration corresponding to a separate-flow nozzle of by pass ratio 5 with round fan and nozzle operating at the takeoff cycle point of with freestream Mach number of 0.28. Simulation results will be validated against experiments carried out in the Low Speed Aeroacoustics Wind Tunnel (LSAWT) at NASA Langley's Jet Noise Laboratory (JNL). The high-fidelity model developed and validated in Phase I will be extended to explore more complex engine/airframe configurations in Phase II.

Benefits: The proposed SBIR project and its successful completion would result in a strong predictive aeroacoustics toolbox for licensing to clients in electronic cooling (such as Valeo and Siemens), automotive, and transportation industry (such as Alstom Transportation) in addition to the aviation industry as a whole. It would also enable us to pursue consulting projects on a much larger scale. We envision that as part of this project, a complete tool box will be developed which includes but is not limited to: high fidelity compressible, and incompressible LES simulations for direct calculations of noise and boundary element methodology for far field noise calculations. This toolbox will all be packaged into one appropriate design tool to be used by design engineers.

The proposed hybrid LES/RANS framework is aimed to be a high fidelity simulation toolbox for physics-based aircraft noise prediction. Application of this tool will improve the understanding of noise generation and propagation mechanism in both component and system level. Such understanding is a vital step for discovering novel noise reduction concepts. Accordingly, the function of the final product aligns well with the objective of NASA's Quiet Aircraft Technology (QAT) program through both Airframe System Noise Reduction (ASNR) and Engine System Noise Reduction (ESNR). In addition, the absence of problem dependent tuning parameters and generality of this framework allows for application of this toolbox to "conventional" as well as "revolutionary" aircraft configurations.