Pseudospectral Methods for Computational Aeroacoustics
Abstract
Community noise prediction and aircraft certification are a couple of
reasons why there is a need to predict noise level in the acoustic
field. Recent work in Computational Aeroacoustics has involved direct
solution of the wave equation and Euler's equations in the far
field. There are
unique computational issues such as modeling quantities which
may become smaller than the computational error over large temporal
and spatial scales. Numerical methods need to be highly accurate
and virtually free of dispersion and dissipation errors in order
to address these issues.
This has led to the development of a method which
approximates solutions using
Chebyshev psuedospectral methods with grid transformations in the radial
direction and
and Fourier pseudospectral
methods in the angular directions. Results from the effect of grid
transformations on phase and amplitude errors will be shown. In
addition, solutions of the spherical wave problem using
this method have smaller errors than those previously found
with a multidomain Chebyshev psuedospectral method. Future work
will also be discussed.
For further information please contact:
mittelmann@asu.edu
