MGBK is a computer program that predicts the mixing noise and the shock noise generated by a subsonic or low-supersonic jet, given input data on the mean flow and turbulence fields as predicted by a suitable state-of-the-art computational fluid dynamics (CFD) program. More specifically, MGBK can be regarded as a post-processing program for use with a CFD program that computes a Reynolds-averaged Navier-Stokes solution with turbulence represented by a mathematical model of a type known in the art as "k — e" [in which k denotes the time-averaged kinetic-energy density associated with the local fluctuating (turbulent) component of flow, while e denotes the time-averaged rate of dissipation of this turbulent-kinetic-energy density]. The predictions generated by MGBK are in the forms of sound-pressure levels and frequency spectra on arcs or sidelines.
MGBK is an updated version of a noise-prediction program, called "MGB," developed during the 1970s. MGBK predicts only the noise generated in the part of a jet plume external to the exit plane of the nozzle of a jet engine; it does not predict mixing or turbomachinery noise generated within the engine. Still, in cases in which external mixing is the dominant cause of noise, MGBK could prove valuable for understanding noise and reducing noise through improvements in the designs of nozzles and/or mixers.
The approach taken in the development of MGB and MGBK has been one of mathematical modeling of noise sources associated with turbulence. In order to predict noise by use of MGBK, one must use an H-type structured computational grid with slices perpendicular to the direction of mean flow. The input to MGBK comprises data on the CFD-predicted flow field on this grid, plus several parameters. Although the prediction of noise applies to the external part of the jet plume only, boundary conditions in CFD predictions are specified well upstream of the nozzle exit and from 20 to 30 diameters downstream of the exit plane.
The input data are processed by means of semiempirical mathematical models of noise-source strength and acoustic propagation. The source model includes both self and shear noise-source terms. Anisotropy of turbulence is incorporated into the source model via an axisymmetric turbulence submodel. Inasmuch as noise is a byproduct of turbulence, the noise predictions of MGBK are unavoidably sensitive to the details of both the CFD input and the noise models and, hence, some degree of empiricism unavoidably enters via the models.
This program was written by Abbas Khavaran of Dynacs Engineering Co. for Glenn Research Center. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp under the Software category.
Inquiries concerning rights for the commercial use of this invention should be addressed to
NASA Glenn Research Center
Commercial Technology Office
Attn: Steve Fedor
Mail Stop 4 —8
21000 Brookpark Road
Refer to LEW-17062.