On the Use of a Domain Decomposition Strategy in Obtaining Response Statistics in Non-Gaussian Seas

Abstract

During recent years, thorough experimental and numerical investigations have led to an improved understanding of dynamic phenomena affecting the fatigue life and survivability of offshore structures, e.g., ringing and springing and extreme wave impacts. However, most of these efforts have focused on modeling either selected extreme events or sequences of highly nonlinear waves impacting offshore structures, possibly overestimating the actual load to be experienced by the structure. Overall, not much has been done regarding short-term statistics. Although clear non-Gaussian statistics and therefore higher probabilities of extreme waves have been observed in random seas due to wave–wave interaction phenomena, which can impact short-term statistics for the structural load, they have not been studied extensively regarding the assessment of the dynamic behavior of offshore structures. Computational fluid dynamics (CFD) models have shown their viability for studying wave–structure interaction phenomena. Despite the continuously increasing computational resources, these models remain too computationally demanding for applications to the large spatial domains and long periods of time necessary for studying short-term statistics of non-Gaussian seas. Higher-order spectral (HOS) models, on the other hand, have been proven to be efficient and adequate in studying non-Gaussian seas. We therefore propose a one-way domain decomposition strategy, which takes full advantage of the recent advances in CFD and of the computational benefits of HOS. When applying this domain decomposition strategy, it appeared to be possible to deduce response statistics regarding the impact of nonlinear wave–wave interactions.