Mooring analysis for a TLP supporting a 15MW wind turbine

Abstract

With the pressing matter of climate change, the importance of renewable energy sources is increasing day by day. A large portion of this renewable energy is to be generated with the use of wind turbines, which are commonly located onshore. In recent decades, great effort has been put in the development to place these turbines offshore, because of a higher yield in energy production, as a result of higher wind speeds, greater consistency and less interference from human-made objects. These offshore turbines can be placed on floating or bottom-founded structures. Only 20% of all possible wind farm locations are suitable for the application of a bottom-founded wind turbine, due to its depth restriction. Therefore, a great deal of development is focused on the application of floating wind turbines. In essence, there exist three different types of floating structures: semi-submersible, Single Point Anchor Reservoir (SPAR) and Tension Leg Platform (TLP).

In this thesis, the TLP is considered for a floating foundation because to its superior motion characteristics compared to the other structures. The TLP obtains its stability from its mooring system due to excessive buoyancy, which causes the mooring lines to be under tension. This in turn causes the TLP to have compliant and constrained Degrees of Freedom (DOFs). These constrained DOFs cause the eigenfrequencies of the system to be out of reach of first-order wave loading. However, from the oil & gas industry it is known that this type of structure does have a high-frequency response. This is known as springing and ringing, where this thesis is devoted solely to springing. Springing is identified as the resonance of the constrained DOFs of the TLP caused by the subsequent wave loading on the floater.

From initial evaluations of the motion characteristics of the case model TLP, it is seen that there are eigenfrequencies of the system that coincide with both the high frequency tail of the first-order wave load spectrum and the second-order wave load spectrum. Therefore, first- and second-order wave loads are considered. These forces are evaluated with the use of potential theory and calculated with OrcaWave and Hydrostar, which are commercially available diffraction programmes. Springing is subsequently evaluated using time-domain analysis with OrcaFlex, which is a commercially available software capable of performing time-domain analyses with aero-hydro-servo-elastic coupled models.

From initial simulations, it is seen that springing events are indeed present in the system and are mainly affecting the fatigue life of the tendons. Additionally, slack tendon events are present to a larger degree and are correlated to the whipping of the tower and RNA. This mainly causes the exceedance of structural limits. As springing is a resonance problem, only adjustments to the mooring system are considered that influence the location of the eigenfrequency. In this thesis, adjustment of the axial stiffness and inclination of the tendon are considered. Adjusting the axial stiffness did not produce the desired effect, because the range of common values in the mooring industry is rather limited to invoke any considerable change in the motion characteristics of the TLP. Furthermore, the TLP showed a large number of slack tendon events. that are correlated with the whipping of the tower and consequently caused the exceedance of structural limits. Adjusting the axial stiffness is not able to prevent these events. Adjusting the tendon angle did show promising results. For the case study, an optimum angle of 16° is found, which on average halves the tendon tensions. A reason for this is that with the introduction of a tendon angle, the TLP rotates around a certain focal point. At 16°, this focal point is located at the Rotor Nacelle Assembly (RNA). Thus, the TLP tends to rotate around the RNA instead of the other way around. This change of motion behaviour also prevents the occurrence of slack tendon events. Although not preventing the occurrence of springing events, the optimal tendon angle does reduce the effect and rate of occurrence of springing to a satisfactory level.