Comparison of Simulated And Measured Motions of a Turret Moored Vessel

Investigations on prediction methods for extreme motions and loads in turret mooring systems have indicated a considerable scatter of results. Nielsen et al. [6] reported a considerable scatter between results from different analysis programs. The scatter is mainly observed m the low frequency motions and is to a large extent attributed to uncertainties. In damping models, and In extreme response distribution models. This paper describes results from an experimental and theoretical study of a turret moored vessel. To obtain reliable estimates of extreme motions, long time series are required. In a research project on Floating Production Systems in Deep Water, FPS2000, a series of model tests have been undertaken as a part of the programme on Mooring and Positing. The main test series cover up to 18 hours m full scale. The wave-frequency (WF) motions are normally computed m the frequency domain and the low-frequency (LF) motions the time domain. The total motion is then found by some sort of summation described later in this paper. This summation procedure is not straight forward when the WF and LF motions are dependent. Another alternative is to compute all forces acting on the vessel, and compute the total motion in the one domain. This method is used ill the present analyses. The model tests were carried out in scale 1:70 in a water depth corresponding to 200 m. The data required for preparing a time-domain summation model were partly obtained from model tests and partly from theoretical analysis. Wind- and current force coefficients were obtained from model test with the actual model. Hydrodynamic coefficients, including frequency-dependent added mass, damping and 1st and 2nd order wave excitation force were calculated with a 3-D diffraction program (MULDIF). This program also calculates the velocity dependency of the 2nd order wave drift forces.