Boskalis subsidiary Dockwise is the global market leader in heavy marine transport. Their fleet of over 20 semi-Submersible Heavy Transport Vessels (SHTV) is capable to carry the heaviest and largest structures. Until today the load or discharge operation of a floating body has only been performed in sheltered waters. Here, the influences of the environmental loading (wind, waves and current) are limited. Floating Production Storage and Offloading vessels (FPSO) are subjected to hull maintenance and repairs in the near future, which causes a major downtime due to transportation. If the maintenance of FPSOs could be performed offshore, it could reduce this downtime significantly. Due to the environmental loading the SHTV and FPSO move separately. In order to perform a safe offshore load/discharge operation their relative motion should be within the width of the cribbing. The aim of this thesis is to analyse the influence of the location of the connections between the two vessels on the relative transverse motion at keel level of the FPSO. Sway, roll and yaw are considered as the main degrees of freedom (DOF) that have influence on the transverse motion. A numerical model is built to solve the equations of motion in the frequency domain for a 3DOF system. This system consists of a rigid body which represents the FPSO. The rigid body is connected to a set of springs and dashpots which influence its behaviour. For every combination of location and parameters of the spring/dashpot system the numerical model calculates the response. The outcome of all these runs are brought together in contour graphs that give insight to a possible optimal location for the connections and their corresponding parameters. It is found possible to get the transverse motions at keel level within the limits of the cribbing. By positioning the spring/dashpot systems in the corresponding optimal location the minimum required stiffness can be reduced. A fraction of the critical damping at each spring location appears to be adequate in reducing the motions during resonance. The significant forces in both spring/dashpot locations are very high but workable.

During offshore installations in harsh sea conditions, the involved gangway system must satisfy rigorous requirements in terms of safety and accuracy. The forces resulting from the vessel motion have an extensive effect on the overall gangway structure and its lifetime. Moreover, vessel motion handicaps the operator during fine positioning of the gangway during connection to a platform. Hence, an active motion compensation system for the gangway is used. An impression of the gangway structure can be seen in Figure 1 below. In this thesis, a 2D-model for the gangway system is developed in order to simulate the Active Motion Compensation (AMC) required to compensate wave induced vessel motions. The model can be used to estimate the power required by the system, needed to compensate the motions in order to minimize the movements of the connection point. The model is developed using software Matlab Simscape.
The gangway is capable of performing single stage telescoping motion, activated by a winch that pulls the telescoping section back and forth, luffing motion by activating of the hydraulic cylinders and slewing motions by means of a slewing bearing. In order to develop an as accurate as possible model, the whole gangway system is divided in three sub-systems: the mechanical, the hydraulic and the control sub-system. The main outputs of the mechanical sub-system are the movements, velocities and accelerations of the gangway system relative to its mass and center of gravity and based on the RAO’s of the vessel. For the hydraulic luffing system, all relevant hydraulic components are modeled. A directional proportional (control) valve in the hydraulic sub-system is used to regulate the amount of flow through the system. In order to compensate the gangway tip, a feedback on the position and a feedforward on the velocity of the compensation point is applied. The inputs from the MRU, which are the imposed ship motions (heave, sway and roll), are converted to the motion of the tip of the gangway by a PLC controller. These converted signals are used to determine the control set points for the luffing and telescoping actuators. A basic proportional controller is used for the Active Motion Compensation system of the gangway.
In order to validate the numerical model, the simulation results are compared with experimental results from quayside tests. During these tests, different sinusoidal signals are applied for the 3 possible motions: translation in vertical direction which mimics the heave motion, translation in the horizontal direction which mimics the sway motion and a rotation around the main hinge which mimics the roll motion of the vessel. When applying only the roll motion or the sway motion, the luffing function and the telescoping function respectively could be assessed separately. Besides these three signals, combinations of this motions are also provided to assess the total performance of the system. Several variables are logged during the tests of which the compensation error, the measured set points, and pressures in the hydraulic cylinder are analyzed thoroughly and compared with the simulation results. It is observed that the simulation results show important similarities in the dynamic response of the gangway on the imposed ship motions.
The current model provides a good insight in the behavior of the gangway system and can be used to simulate offshore tests in different virtual weather windows before testing the real gangway offshore. This is the main advantage compared to the case before this thesis is conducted. It is recommended to extend the 2D simulation into a 3D simulation by adding the slewing function. It’s also recommended to further study the controller design in order to improve the AMC performance.

The Bristol Channel has one of the largest tidal ranges in the world, with a mean tide of nearly 10.5 m at Avonmouth, and speculations on how to harness this energy date back well over a century. One method is tidal range by the use of tidal lagoons, which operates on principles akin to hydropower. The lagoon is essentially an enclosed reservoir near the coast, which can maintain the inside water level while the outside rises and falls with the tide. When a sufficient head difference between the inside and outside water level is achieved, the turbine gates open and the flow drive the turbines.

This thesis explores a novel modelling approach consisting of a 1D channel based on the inertial form of the Saint-Venant Equations, coupled with 0D tidal lagoons. The model was built in Modelica. The 1D channel subsystem was modelled first, and validated by comparison with tidal constituents at multiple measuring stations in the domain. The channel model’s accuracy was deemed sufficient, with root-mean-square deviations ranging from 0.29 to 0.67 m. With the model validated, the nature of the tidal resonance in the channel was investigated, where the natural period was determined to be located between 7-8 hours. It was shown that the natural period is heavily dependent on the tidal wave propagation velocity, which is in accordance with the quarter-wavelength resonance theory.

The 0D tidal lagoon subsystem was coupled with the 1D channel and validated by considering the proposed Swansea lagoon design by Tidal Lagoon Power and comparing the results with the current literature. The lagoon was found to be operating as expected in dual generation, and the annual energy generation was 425 GWh. The energy production is slightly conservative compared to literature, but within reason. The lagoons' impact on the channel hydrodynamics were also investigated, and it was found to have a significant effect, particularly in the narrow section of the estuary. The influence of the lagoon-channel coupling on energy generation was also explored: As the hydrodynamic impact increases, the energy generation from the system as a whole decreases.

A 2D model was also utilised to explore the higher-order and near-field effects that would potentially be lost in a 1D model, like localised dips in water elevation, reduced circulation zones, jets and more. The 2D results indicate that consideration of the localised effects is crucial when designing tidal lagoons.

The tidal lagoons were also implemented, and the Swansea lagoon had an annual energy generation of 408 GWh, closely matching the 1D model. Three larger tidal lagoons were also implemented, however not successfully scaled to match the 1D model due to time constraints. Operational tidal lagoons were also shown to have a tremendous influence on the water elevation in the channel.

In conclusion, the 1D modelling approach is potentially as a powerful tool to supplement more detailed 2D models, and it can potentially replace the 0D models ordinarily used for optimisation and sensitivity analyses in the preliminary investigations.

Trends towards higher capacity offshore wind turbines (OWTs) and deeper offshore wind farm sites, make wave-induced vibrations more dominant in its design. During idling phases of these next-generation OWTs, additional damping is required. This can be achieved through dynamic vibration absorbers (VAs), i.e. dampers. More importantly, wave-induced vibrations can decrease the installation workability of these OWTs up to 50%. VAs which are designed for the operational life of an OWT, are not effective at natural frequencies typical for partially installed OWTs. To reduce installation costs to acceptable values, higher damper effectiveness is required.
The main objective of this thesis is, therefore, the reduction of wave-induced vibrations during OWT installation by non-invasive modification strategies of off-tuned VAs.

In the first part of the research, a passive damper modification is proposed. A numerical model was developed of the stand-alone passive modified VA. Using data from small scale-experiments, the steady state response of the model was validated. However, the passive method was shown to be insufficient in supplying additional damping at installation frequencies.

In the second part, semi-active control strategies were deployed. This semi-active controller, which cannot add mechanical energy through the actuator, results in an inherently stable system. A hybrid control law and a clipped-LQR control law were investigated. The damper its performance was tested for a deep-water site case study. The semi-active damper modification was shown to be a robust approach for frequency tuning. Wave-induced vibration reduction of up to 60% could be achieved, independent of the off-tuning of the unmodified damper. Actuator dynamics were included and shown to have insignificant effects on the performance.

It is therefore concluded that, the semi-active damper can be effective over the whole installation and operational range of an OWT support structure. For the representative case-study, the installation workability can be increased from only 50% to approximately 90%.