Offshore installation vessels are growing in size in order that they are capable of handling the next-generation 15 and 20MW wind turbines. At the same time, it is important to gain a better understanding of their structural behaviour using the finite element (FE) method. One approach that is often used is the full vessel FE analysis, which studies the vessel in its entirety. However, the drawbacks are that a coarse mesh must be adopted to have an acceptable simulation time and it takes months to create a 3D model of a vessel. Another approach is the partial vessel FE analysis, which considers only a section of the vessel. This research aims to develop a methodology for the partial vessel FE analysis of an offshore jack-up installation vessel.

The studied section of the vessel is between the four legs, which is where the highest bending moment is expected and the wind turbine towers are located on deck. The methodology was developed to analyse the vessel in still water and waves separately. The boundary conditions and hull girder load adjustments of the partial vessel model were described, which cause the vessel section to reflect the behaviour of the full vessel. The procedure to apply the loads on the model was also given. The finite element analyses were performed and the stress results compared to a simple analytical solution of the vessel based on Euler-Bernoulli beam theory.

The methodology of the partial vessel FE analysis was successfully applied to the offshore jack-up installation vessel. For the vessel in still water, the stress results were in agreement with the analytical solution. The results were accurate in the middle region for approximately 50% of the partial vessel model and thus it is advised to have a larger model for future use. The effect of torsion could not be fully justified as it is not captured in the analytical solution, but resulted in a significant improvement. For the vessel in waves, the stress results also showed a good agreement with the analytical solution. The accuracy depended on the equivalent design wave approach, which assumed a regular travelling wave. Lastly, it was proven that the presence of payload only leads to local changes in the stress field, but not in the global behaviour of the vessel.