Validation of Overset Discontinuous Galerkin and Hybrid RANS/LES Method for Jet Noise Prediction

Abstract

Launch vehicles experience extreme acoustic loads during lift-off which are capable of damaging sensitive mechanical components of the vehicle and payload, and may lead to significant injury in passengers. The noise is driven by the exhaust plumes and their interaction with ground structures, and accurate prediction of this environment poses several unique challenges when compared to traditional jet noise modeling techniques. In particular, acoustic and Mach waves propagating from the exhaust plume are in the vicinity of obstructions and are generally in the nonlinear regime, which renders the classical acoustic analogy and linearized approaches invalid. This work continues the development of a recently introduced novel approach for the accurate prediction of lift-off acoustic environments. The method makes use of unstructured hybrid RANS/LES to predict the complex exhaust plume physics, and an overset high-order accurate discontinuous Galerkin solver to accurately propagate acoustic waves across large distances throughout the environment. The framework upon which the solvers are developed is described along with details outlining the overset grid assembly and interpolation methods. Detailed validation studies are presented to demonstrate the accuracy of the method for prediction of noise due to the impingement of a supersonic jet on a 45 degree inclined flat plate, and comparisons with experimental data show the method to be effective and accurate in terms of predicted overall sound pressure level spectra.