Abstract
Efficient methods are presented for digital simulation of a general
homogeneous process (multidimensional or multivariate or multivariate-multidimensional)
as a series of cosine functions with weighted amplitudes, almost
evenly spaced frequencies, and random phase angles. The approach
is also extended to the simulation of a general non-homogeneous oscillatory
process characterized by an evolutionary power spectrum. Generalized
forces involved in the modal analysis of linear or non-linear structures
can be efficiently simulated as a multivariate process using the
cross-spectral density matrix computed from the spectral density
function of the multidimensional excitation process. Possible applications
include simulation of (i) wind-induced ocean wave elevation, (ii)
spatial random variation of material properties, (iii) the fluctuating
part of atmospheric wind velocities and (iv) random surface roughness
of highways and airport runways.
Users
Please
log in to take part in the discussion (add own reviews or comments).