Hybrid Reliability Predictions of Single and Advanced Double-Hull Ship Structures

Abstract

A hybrid approach based on the combination of the Monte Carlo simulation (MCS) technique and the first-order reliability method (FORM) is developed to quantify the small probability of failure of a hull girder under longitudinal bending. The ultimate strength (ULTSTR) bending solver, developed by the U.S. Navy, is integrated with the hybrid probabilistic analysis framework. Different from the conventional approach, where the probability distribution of the hull capacity is preassumed, the probability distribution of the ultimate bending strength is determined by propagation of the uncertainties associated with hull geometric and material property parameters into ULTSTR via a small number of Monte Carlo simulations. The FORM-based solution module is then applied in the reduced random variables space to compute the probability of failure. Both the computational efficiency and numerical accuracy of the hybrid approach are demonstrated through its application to a Navy frigate, designated Ship A. The validated hybrid approach is then applied to a Navy combatant, designated Ship B, in both its single-and double-hull configurations. The reliability results reveal the benefit in using the advanced double-hull configuration, assuming its design load is the same design load encountered by the corresponding single hull during its service life.