Parametric roll is a nonlinear phenomenon that can rapidly create large roll angles coupled with significant pitch motions. Such motions can cause large loads especially in fore and aft parts of the ship structure, severe stresses on containers and their securing systems, and can be a threat for the crew of the ship. The large roll angles may increase to such values that can even cause capsizing. Even though there are many studies focusing on the parametric roll phenomenon using theoretical, numerical and experimental approaches, topics such as uncertainty quantification and sensitivity analysis in parametric roll have not attracted enough attention. In this paper, results from a study on uncertainty quantification in numerical simulation of parametric roll are presented. Numerical simulations were carried out in irregular seas using the Korean Container Ship (KCS). Motion of the ship was simulated with a time-domain program. Several parameters were considered as sources of uncertainty such as the GM, the length of the simulation time window, the phase, and the number of components of the irregular wave. To propagate these uncertainties in the computational model, Latin hypercube sampling (LHS) and Polynomial chaos expansion (PCE) methods were used, and their performances were compared. The paper ends with a summary of conclusions and recommendations.

This content is only available via PDF.
You do not currently have access to this content.