The large amplitude surge-sway-yaw 'slow-drift' motions of a floating
body constrained by weak restoring forces in random waves are consider
ed. A multiple time scales approximation is employed to separate the f
ast time scale associated with the linear motions from the slowly vary
ing motions. The ideal fluid free surface flow is approximated by a pe
rturbation series expansion for small slow-drift velocities and wave-s
teepness, and is solved around the instantaneous position of the body.
The linear zero-speed and forward-speed velocity potentials are solve
d for arrays of vertical cylinders, using exact interaction theory. Th
e horizontal mean drift forces and the wave-drift damping are obtained
, and results for realistic configurations are compared with well-esta
blished methods. The surge-sway-yaw equations of the slow-drift motion
s are solved numerically in the time domain under the influence of sho
rt-crested, random waves, including viscous forces. The random wave-si
gnal is generated by the filtering of white Gaussian noise. The slowly
-varying forces are obtained using the Newman approximation and effici
ent summations of time series. The results are compared with full QTF-
matrix (Quadratic Transfer Function-matrix) computations of the exciti
ng force. The use of a robust random number generator and the Fast Fou
rier Transform allows for efficient simulations of long records of the
slow-drift motions, and the study of its statistical parameters. The
sensitivity upon the simulation length, transients, drag-coefficient a
nd directional spreading are demonstrated. Copyright (C) 1996 Elsevier
Science Limited