A NEW EFFICIENT ALGORITHM FOR COMPUTATIONAL AEROACOUSTICS ON PARALLELPROCESSORS

One of the great challenges in computational physics is the prediction of flow associated noise, where the quantities of interest, namely th e sound waves can be at high frequencies and are usually orders of mag nitude smaller in magnitude than the mean quantities. In order to nume rically resolve such small scales governed by the fluid dynamics equat ions, high resolution schemes are required. Thus solutions of flow noi se problems are computationally intensive, An efficient, hybrid, data parallel computational aeroacoustics algorithm has been developed for the prediction of noise radiation and scattering from three-dimensiona l geometries. The algorithm solves the Euler/Navier-Stokes equations i n the interior and nonreflecting boundary conditions on the outer boun daries. A moving surface Kirchhoff method is coupled to the flow solve r for far-field predictions. The algorithm uses standard time and spat ial discretization techniques but utilizes several new optimization st rategies that are high ly suitable for single zone solutions on data p arallel processors. One strategy, for example, enables simultaneous re sidual evaluations of the interior and far-field nonreflecting boundar y conditions equations, reducing the computational effort spent on the m by approximately 60% CPU time savings. The algorithms for the flow s olver and the Kirchhoff method and their coupling are described in thi s paper, and results for some example radiation and scattering problem s are presented. (C) 1996 Academic Press, Inc.