A Comparison of Time Domain and Frequency Domain Approaches for the Fully Coupled Analysis of Deepwater Floating Systems

Abstract

The recognition of the need for a fully coupled analysis of deepwater floating production systems has led to the research and development of several coupled analysis tools in recent years. Barring a handful of exceptions, these tools and available commercial packages are invariably in the time domain. This has resulted in a much better understanding and confidence in time domain coupled analysis, but less so for the frequency domain approach. In this paper, the viability of frequency domain coupled analysis is explored by performing a systematic comparison of time and frequency domain methods using computer programs developed in-house. In both methods, a global coordinate system is employed where the vessel is modeled with six degrees-of-freedom, while the mooring lines and risers are discretized as lumped masses connected by extensional and rotational springs. Coupling between the vessel and the mooring lines is achieved by stiff springs, and the influence of inertia and damping from the lines are directly accounted for without the need for prior assumptions. First and second order wave forces generated from a random environment are applied on the vessel, as well as drag and inertia loading on the lines. For the time domain simulation, the Wilson-theta implicit integration scheme is employed to permit the use of relatively large time steps. The frequency domain analysis is highly efficient despite being formulated in global coordinates, owing to the banded characteristics of the mass, damping and stiffness matrices. The nonlinear drag forces are stochastically linearized iteratively. As both the time and frequency domain models of the coupled system are identical, a consistent assessment of the error induced by stochastic linearization can be made.