FIELD VERIFICATION OF LINEAR AND NONLINEAR HYBRID WAVE MODELS FOR OFFSHORE TOWER RESPONSE PREDICTION

Accuracy of the prediction of the dynamic response of deepwater fixed offshore platforms to irregular sea waves depends very much on the the ory used to determine wave kinematics. A common industry practice cons ists of using linear wave theory, which assumes infinitesimal wave ste epness, in conjunction with empirical wave stretching techniques to pr ovide a more realistic representation of near-surface water kinematics . The current velocity field is then added to the wave-induced fluid v elocity field and the wave-and-current forces acting on the structure are computed via Morison's equation. The first objective of this study is to compare the predicted responses of Cognac, a deepwater fixed pl atform, obtained from several popular empirical wave models with the r esponse Cognac predicted based on the hybrid wave model. The latter is a recently developed higher-order, and therefore more accurate, wave model which satisfies, up to the second-order in wave steepness, the l ocal mass conservation and the linear free surface boundary conditions at the instantaneous wave surface. The second objective of this study is to correlate the various analytical response predictions with the measured response of Cognac. Availability of a set of oceanographic an d structural vibration date for Cognac provides a unique opportunity t o evaluate the prediction ability of traditional analytical models use d in designing such structures. The results of this study indicate tha t (i) the use of the hybrid wave model provides predicted platform res ponse time histories which overall are in better agreement with the me asured response than the predictions based on the various stretched li near wave models, and (ii) the Wheeler stretching technique produces p latform response time histories which overall are more accurate than t hose obtained by using the other stretching schemes considered here.