Data-Driven Multiobjective Optimization of Wave-Packets for Near-Field Subsonic Jet Noise

Abstract

Aeroacoustics of innovative aircraft cannot disregard the development of low-order models for the jet noise source, which are essential to assess the propulsion–airframe interactions from the conceptual design stage. The main scope of this work is to provide a noise source model that can be coupled with relatively low computational cost methods for aeroacoustic scattering. To this end, this paper presents for the first time a multiobjective optimization of the zeroth-mode wave-packet in the jet near field. The importance of calibrating the model with near-field pressure data stems from the fact that in new aircraft, the engine nacelles are typically positioned at a few diameters from the wing or fuselage. In this work, the near field of a high subsonic jet at a Mach number of 0.9 is represented as a cylindrical surface radiating the pressure disturbances of a wave-packet source, which is optimized using large-eddy simulation data from three lines at different radial distances. The optimized model has been tested at Strouhal numbers between 0.25 and one, and the optimized solutions have been chosen using a Pareto-ranking criterion considering the wavepacket prediction over an extra line. A good agreement is achieved between the reference data and model predictions for multiple near-field radial distances.