Marine structures can sustain damage due to violent wave impacts that are characterized by complex dynamics involving both water and air. Methods that predict impacts loads for structural design often assume the water as incompressible. These predictions can be inaccurate during impacts where air is entrained in water, as that greatly increases its compressibility. Initial impact forces may decrease due to a cushioning effect. However, the total severity of the impact may increase due to temporal spreading and oscillations of impact pressure. Previous research on aerated water impacts covers breaking waves against fixed structures, flat water slamming or green water events, but not the interactive motion of rigid marine structures in free surface waves. This work aims to evaluate the effect of aerated water impacts on the dynamics of floating rigid bodies in irregular waves. As a part of this work, a state-of-the-art numerical simulation method for aerated water impacts is extended with a monolithic coupling for rigid body motion. Additionally, boundary conditions for wave generation were implemented for a validation of the method during breaking wave impact. Results of the verification of both extensions compare well to other literature. It can be concluded from this thesis that the compressibility effects of aeration on penetration depth after cylinder slamming can be neglected for entry Mach numbers below 0.1. For further research, it is recommended that this numerical method is used for a range of experiments with various geometries and higher impact velocities. This could provide more insight into situations where aerated water could have important design implications.It can be concluded from this thesis that the compressibility effects of aeration on penetration depth after cylinder slamming can be neglected for entry Mach numbers below 0.1. For further research, it is recommended that this numerical method is used for a range of experiments with various geometries and higher impact velocities. This could provide more insight into situations where aerated water could have important design implications.It can be concluded from this thesis that the compressibility effects of aeration on penetration depth after cylinder slamming can be neglected for entry Mach numbers below 0.1. For further research, it is recommended that this numerical method is used for a range of experiments with various geometries and higher impact velocities. This could provide more insight into situations where aerated water could have important design implications. It can be concluded from this thesis that the compressibility effects of aeration on penetration depth after cylinder slamming can be neglected for entry Mach numbers below 0.1. For further research, it is recommended that this numerical method is used for a range of experiments with various geometries and higher impact velocities. This could provide more insight into situations where aerated water could have important design implications.

With the transition towards renewable energy, offshore wind power farms are being constructed or planned in many places involving placing wind turbines in the sea. The monopiles for these wind turbines are vertical cylinders which are situated in a flow with high Reynolds numbers and low to intermediate Keulegan- Carpenter numbers. The Reynolds and Keulegan-Carpenter number are dimensionless numbers for the quantification of respectively currents and waves. Previous research has not covered the Reynolds and Keulegan-Carpenter numbers in which energy generating devices are situated. The current research is focused on filling the gap and investigating the drag forces on these cylinders via means of experiments. Additional simulations focussing on the free surface and end effects were also conducted. The forces at play and the Morison equation to model these forces have been evaluated. It was discovered that in the investigated range, the drag and inertia coefficients depend on both Reynolds and Keulegan-Carpenter numbers. By analyzing the results of the experiments in the time and frequency domain, indications of vortex shedding and second-order harmonic wave forces, as well as forces from wave-wave interaction were found. The Morison equation itself is analyzed to find that it either underestimates or neglects these forces. By means of simulations, the effects of the free surface and the aspect ratio of the cylinder on the drag and inertia coefficients are found. Lastly, it is proposed to use a rewritten Morison equation to find the drag on a cylinder in a flow with irregular waves. This equation is found to describe the drag forces more accurately.

Surrogate modelling techniques such as Kriging are a popular means for cheaply emulating the response of expensive Computational Fluid Dynamics simulations. These surrogate models are often used for exploring a parameterised design space and identifying optimal designs. Multi-fidelity Kriging extends the methodology to incorporate data of variable accuracy and costs to create a more effective surrogate. This work recognises that the grid convergence property of CFD solvers is currently an unused source of information and presents a novel method that, by leveraging the data structure implied by grid convergence, could further improve the performance of the surrogate model and the corresponding optimisation process. Grid convergence states that the simulation solution converges to the true simulation solution as the grid is refined. The proposed method is tested with several problems involving both synthetic and realistic multi-fidelity data acquired with Computational Fluid Dynamics simulations. The synthetic results indicate that our method surpasses the most well-known multi-fidelity Kriging model in terms of both accuracy and cost given the proper conditions. The performances of both surrogates in the optimisation cases using Computational Fluid Dynamics simulations were similar. Data-supported conditions for the proposed method to be effective, and a procedure to cheaply recognise them, are provided. To further improve and validate the method, recommendations are given at the end.

When sailing in heavy seas, it happens that the water washes over the bow. This shipping of water is generally divided into two types. The term white water is used when the water is more of a spray and is generally harmless. Green water is more violent and can pose a danger to the crew and structures on board. Research on green water has mainly focused on the arrival of green water on deck and the different types. Impacts are generally considered as result of green water wave and tested by a dam-break problem. The impact of green water has not been studied as solely the impact of water on a structure. Drop test experiments are conducted in different ways for different shapes in the past. However, there have not been tests before with green water using drop tests. Drop tests are deemed suitable to test shapes with different deadrise angles and create aerated water. The aim of this work is to investigate the effects of green water on deck structures by studying the maximum impact pressure when considering the deadrise angle and the aeration. In this thesis the influence of shape and aeration on the maximum pressure at impact is investigated. To investigate the effect of aeration and shape, the impact tests are conducted using a drop tower experiment. For this experiment different increasing deadrise angles are chosen from 0± or a flat plate up to 30±. In addition test with aerated water are done up to 4% which is in range of real green water waves. The aeration and the shapes are chosen in such a way that non-compressible, linear compressible and non-linear compressible effects take place. It can be seen that with a flat plate, the impact increases almost instantaneously and produces a high peak. Adding aeration reduces this peak considerably. Fromexamining the impact of the wedges, it can be concluded that the pressure rises more slowly at a higher deadrise angle. The almost instantaneous drop in pressure seen with a flat plate changes to a smoother rise at a high deadrise angle. In addition to the experiment, numerical simulations are carried out to verify the solution. The numerical simulations also give a better insight into the pressure across the width of the object. Upon examining the numerical simulations, it is concluded that the results for deadrise angles of 15± and 30± are equivalent to the results of the results of the experiment. However, the simulations for a flat plate result in an high pressure which is probably due to a low number of data points which capture the peak. It can be concluded that the shape of the structure and aeration have a significant influence on the maximum pressure. However, the decrease due to a higher deadrise angle is more significant than the effect of additional aeration. With higher aeration, the maximum pressures are closer together and the difference in deadrise angle has less effect. When examining the impact pressures the effect of both deadrise angle and aeration can be seen. The maximum impact for a non aerated flat plate impact measured is 9.93[bar ] and is decreased by 8.8[bar ] for a wedge impact with a deadrise angle of 30± which drops in water that is 4% aerated. The influence of both shape and aeration can be reviewed independently which leads to an reduction of impact pressure for a flat plate of 59 percent when 4% air is added to the system. When only changing the deadrise angle to 30± a reduction of 86% is found.

Amazing Grace will be needed for the removal of offshore platforms that are out of Pioneering Spirit 's reach. The base case design of Amazing Grace is an enlarged version of Pioneering Spirit (Figure 1). Amazing Grace uses mainly Quick drop Ballast tanks, releasing a lot of water at once, for lifting. Pioneering Spirit uses mainly a pneumatic-hydraulic system for fast lifting. The idea is to lift bigger platforms with a less complex system. The behaviour of the base case design of Amazing Grace when lifting the upper part of an offshore platform (topside) from its supporting structure (jacket) is studied in this work and the feasibility of the new Quick drop Ballast system is tested. To perform a feasibility study on lifting topsides with Amazing Grace , a set of design requirements is generated. Before using these design requirements, a one tank model is required to simulate the emptying of a tank. The dimensions of an existing Pioneering Spirit Quick drop Ballast tank are used. The model is validated by comparing its output to existing physical test data of emptying the same tank. To lift by means of heave only, a bigger volume is used for the one tank model. Once the Quick drop Ballast concept shows to be feasible, the next step is to introduce waves. Waves are modelled using the linear superposition method. An estimation of the time series duration is done by computing the standard deviation of the heave amplitude. For a converged standard deviation, the time series' minimum length needs to be 1200 seconds. The maximum heave amplitude, derived from another statistical analysis, is applied to the connection plateau (when Amazing Grace connects and starts lifting the topside without creating clearance). The results show that Amazing Grace 's mass, added mass, damping and stiffness (the coefficients of the equation of motion) need to be included to give a more accurate estimation of the dynamics for connecting Amazing Grace to the topside and the possible rebounce to the jacket. Although this is a conceptual study, the dynamics are included to get a better understanding of Amazing Grace 's response. To include the previously mentioned coefficients of the equation of motion, the Cummins equation is implemented. The excitation forces are the sum of the wave- and the Quick drop Ballast force. The Quick drop Ballast force is derived from the one tank model. A convergence study shows which time step is needed for this model to work accurately considering its application. The implementation of the Cummins' equation is verified by showing agreement between time- and frequency domain. A sensitivity analysis is used to show the effect of changing the model's main variables. The peak period has the biggest influence on Amazing Grace 's heave motion. By varying the coefficients of Cummins' equation in the range of 35.5 - 38 meters draught, the computed vessel motions show that the variation of the draught only has little influence. The Quick drop Ballast system is feasible for both the pretension- and fast lift phase when applying the Quick drop Ballast force at the centre of gravity. The Quick drop Ballast force vector is partly relocated at the bow for applying trim during fast lift. A comparison shows that both the required volume and the number of valves reduce by one third compared to the heave only concept. The maximum allowable trim per length of Amazing Grace is 1.5 meter. This maximum is exceeded by 0.2 meters. The 0.2 meters needs to be included for a future topside lift system (TLS) design. It is recommended to apply trim during fast lift since this simplifies the complexity of the Quick drop Ballast system by reducing the required volume of water and the number of valves.

This thesis study is intended to research the effects on peak pressure during impact of a deformable structure with a liquid. To find an answer to this question, a novel wedge was designed to perform free fall wedge experiments at the TU Delft. This report describes the progress that lead to answering the research question: What is the effect of dynamic deformation of a structure on the maximum pressure at the wedge surface during impact? To create an overview of the problem, the various systems and knowledge required to answer the research question are discussed. Among others, a decision on wedge deformation type is made, the importance of multiple impact velocities and the sensors required to measure data of the impact are discussed. The proposed concept to perform experiments is with a deformable wedge that is allowed to fold around the keel upon impact with the free surface. Hypothesis is that the deformation on impact could lead to initial lower impact pressure but dynamic response could lead to higher peak pressures at later stage of impact. In the next phase, estimations of impact pressures, forces and accelerations are used to finalize the design of the wedge. Strength calculations were performed on the critical parts of the wedge. It is concluded that the analysed parts are strong enough to withstand repeated loading of the structure. Also during this phase, an extra experiment into dynamic calibration of the accelerometer was performed. By curve fitting the expected acceleration to the experiment data, the error of the accelerometer is determined. The accelerometer used in this experiment has an error of 0.48% which is within the limits of the manufacturers requirement of ≤1%. Before the start of the experiments the limitations of the setup are discussed and an uncertainty analysis is performed on the sensors. The experiment consists of a series of free falls with various wedge setups, either rigid or deformable. A rigid setup of the wedge is tested at impact velocities of 2, 4 and 6m/s and deadrise angles of 10 and 20 degrees to create a base line for maximum impact pressures. The results of the rigid experiment are compared against previously tested rigid wedges. It is concluded that the novel wedge showed similar peak pressures compared to previously tested rigid wedges. The deformable wedge is dropped at all impact velocities and deadrise angles with three different springs, varying in stiffness. Deformation of the wedge during impact resulted in decrease in peak pressure. Decreasing spring stiffness and increasing impact velocity resulted in increasing deformation of the wedge. With these results the research question set in the first chapter of the report is answered. The answer to the research question is that impact induced deformation of the structure leads to lower peak pressures. Dynamic effects were observed but did not lead to higher pressure than the rigid experiments. While further research into the matter is required, this result may already lead to improved ship design which could lead to lower manufacturing costs and lower emissions.