Over the past few decades, there has been a significant increase in global energy consumption. To address the risks associated with climate change, there is an increasing urgency to transition towards renewable energy sources. With this growing demand, the interest in offshore wind energy has increased significantly, leading to the construction of larger bottom-founded offshore wind farms. In order to fulfill this rising demand, The International Renewable Energy Agency has estimated that a total of 2000 GW of installed offshore wind power is required to achieve net-zero emissions by 2050. However, the current installed capacity stands at approximately 35 GW, indicating a substantial gap that needs to be addressed. The offshore industry faces a major challenge in meeting this demand and making a substantial contribution to the supply of renewable energy sources.

As the size of wind turbines continues to increase, their monopile foundations also grow in weight and dimensions. Conventionally, monopiles are installed by upending them on the vessel, lifting them into the air and lowering them onto the seabed. However, this installation procedure is not feasible for extra-large (XXL) monopiles. This is because these XXL monopiles exceed the crane capacity of the, relatively new, installation vessels. To prevent the vessels from becoming outdated, an alternative approach to upending has been devised. This method involves utilizing the buoyancy of the monopile itself to compensate for the insufficient crane capacity on board of the vessel. This innovative upending technique is referred to as the trapped air method. This research explores this method as the influence of imposed buoyancy on the system's behavior in such operations has not been addressed in earlier research.

The main objective of this thesis is to examine the dynamic behavior and determine the natural frequencies of a monopile experiencing an upward facing buoyancy force. Moreover, it is desired to quantify the workable limits of the concept. First, a literature study is conducted, followed by a study that investigates the mechanics of the system. The monopile suspended from a crane, can be simplified using double pendulum models. Through the analytical approach, the influence of each force, mass and inertia component in the complex system can be examined individually. This assessment results in the equations of motion that serve as the foundation for the numerical model.

To gain initial insights into the behavior of the monopile, exploratory tests are conducted in a purpose-built experimental setup. It is observed that as the buoyant force increased, the monopile searches for an equilibrium to stabilize the system. Hydrodynamic effects induce a noticeable shift of the center of gravity of the monopile towards the waterline. Additionally, when a current is applied to the partially submerged monopile, motions in the sway direction are observed and suggests the presence of vortex induced motions.

Building upon the exploratory tests, decay tests are conducted at TU Delft to investigate the behavior and loads experienced at the cranetip. The results revealed that the presence of buoyancy significantly reduces the side-lead load to approximately X\%-Y\% of the maximum allowable load in the horizontal direction, which nearly half of the load observed without induced buoyancy. Additionally, the vertical loads reached approximately X\%-Y\% of the maximum allowed vertical loads. The buoyancy in the system effectively reduces both the vertical and horizontal loads in the cranetip. Furthermore, the presence of buoyancy leads to a significant decrease in the natural frequency of the system.

The numerical model is based on the equations of motion derived through the analytical approach. Initially, validation of the model is performed by using the scaled model dimensions. The frequency alignment indicates that the numerical model accurately models the natural frequencies on model scale. The model is then used to simulate a full-scale scenario. It can be concluded that numerical approximations of the frequencies closely align with those derived through the analytical approach and those observed during the model experiments. The close agreement among three different approaches provides strong validation of the accuracy and reliability of the numerical model in predicting the natural frequencies in full-scale scenarios.