| dc.description.abstract | For analyzing mooring systems, like other structural response analysis, simulation in both Frequency domain and Time domain has been carried out. Frequency domain approach uses load spectrum and transfer function to predict the response spectrum, and thereby the characteristic response is estimated. This method is quick, and gives quite good first hand estimation of results. On the contrary, the Time domain approach makes realization of wave elevation, and response is predicted by solving equation of motion numerically, at each time instant. While doing this, different realization could be achieved by considering different seed numbers.
In practice, only potential loads are considered to be acting on a large volume structure. Formulation of viscous loads estimation was done by numerous researchers. However, these non-linear loads have been neglected by the industry, due to the assumption that these loads are too low to influence structural response. In addition, simulation of viscous loads by a software tool is cumbersome work, but, after investigation on some recent mooring line failure accidents, viscous overloads has been found as the primary reason.
Chakrabarti (1984) suggested approximate expression for viscous drift forces, acting in a vertical cylinder. Two loads are of importance here. The presence of current causes an added mean force to the structure. Moreover, change in free surface results in an additional dynamic force. An interesting feature of these viscous loads is that they induce additional damping and dynamic excitation to the structure.
MARINTEK has suggested solution for both potential and viscous load prediction by correction of potential wave drift coefficients. This correction includes both potential correction according to Aranha formulation and viscous correction by empirical approach. But, while considering additional viscous drift force, one must undergo the penalty of additional viscous damping forces, which is absent in this formulation.
A base case was established with ULS loads based on Metocean design basis for Hydrun installation (Statoil, 2014). Floatel superior was considered as the reference vessel. All necessary coefficients for the vessel were provided by the thesis supervisor. A simplified catenary mooring system with sixteen mooring lines were established. For the established base case, analysis was done both in Frequency domain and Time domain, and subsequently, predicted responses from these two methods has been compared. As software tools, MIMOSA and SIMA were used for Frequency domain and Time domain analysis respectively. Damping contributions from different sources were investigated. In addition, an effort has been made to estimate the damped frequency and natural frequency of the model.
A model with corrected wave drift coefficients (according to MARINTEK formulation) was established and experimented with. With a view to establishing a better modelling technique, four slender elements were introduced at the four vertical columns, to consider added viscous forces at the splash zone, which is considered as the main source of viscus loads. Different combinations of drag coefficients for those columns and environmental loads were considered and experimented. At the end, comparison has been made between MARINTEK formulation and slender element model. | en |
