On the long-term response of marine structures

This paper addresses some important issues related to the estimation of long-term extreme responses of marine structures. Several convolution models to establish the long-term distribution of a marine structure response parameter are available in the literature. These methods are typically based either on all short-term peaks, all extreme short-term peaks or all short-term upcrossing rates. The main assumptions and simplifications of the five models most usually found in the literature are discussed in this paper. A linear single-degree-of-freedom (SDOF) system along with a bi-lognormal probability model for significant wave heights and zero-crossing wave periods have been used for numerical tests. An improved approach to efficiently evaluate the long-term convolution integrals is also proposed in this paper. It is shown that a combination of the Inverse First Order Reliability Method (IFORM) and an Importance Sampling Monte Carlo Simulation (ISMCS) approach can be used to obtain a very good result for the exact solution of long-term integrals.