An Efficient Importance Sampling Method for the Long-Term Mooring Lines Response Estimation Considering Wind Sea and Swell

Abstract

Abstract Escalating demand in the oil and gas industry has led offshore structures to be installed in ever deeper waters and under severe environmental conditions. As the mooring system is a crucial element in floating offshore structures, a reliable estimation of its long-term response is a decisive step in any usual design procedure. In the long-term scenario, the environmental actions to which these structures are subjected to, such as waves, wind and current, are non-stationary processes. However, this long-term behavior is usually modeled as a series of short-term stationary conditions (typically 3-h). In a full long-term analysis approach, an estimate of the long-term N-year response can be obtained through a multidimensional integration over all these short-term environmental conditions. In this paper, this multidimensional integral is numerically evaluated by means of the Importance Sampling Monte Carlo Simulation (ISMCS) method, where the uniform distribution is used as the sampling function. Thus, all short-term environmental conditions have the same probability of being sampled, which assures that conditions with very low original probability of occurrence, but with knowingly higher contributions to the long-term response, are efficiently accounted for. The random variability of the short-term environmental parameters and their interdependencies are represented by a simplified joint probabilistic model which comprehends both wind sea and swell waves. The methodology is numerically validated for an idealized single-degree-of-freedom (SDOF) model and later investigated for a mooring line connected to an FPSO installed in Brazilian deep waters. It is shown that ISMCS provides good estimates for the long-term N-year response with a moderate amount of required simulations and can be a powerful tool in order to account for simultaneous occurrence of wind sea and swell waves in structural response evaluations.