Designing porous materials with specific properties requires advanced inverse design methods to handle the relationships between microstructure and material behavior. This thesis presents a conditional variational autoencoder (CVAE) model that learns these relationships using Minkowski functionals, geometric descriptors including porosity, surface area, and Euler characteristic, as conditioning inputs. By using the CVAE with convolutional neural networks (CNNs), the model captures morphological features of 2D porous microstructures and links them to the target Minkowski descriptors. A Statistical Representative Elementary Volume (SREV) approach is adopted in data preparation to ensure the descriptors are representative of bulk material behavior, allowing the model to account for natural variability in microstructures. The trained CNN-CVAE can generate realistic porous microstructures that meet specified descriptor targets, effectively performing inverse design. Key findings demonstrate that the model reproduces target Minkowski metrics with high fidelity and maintains plausible connectivity and pore structure. Notably, the CVAE exhibits strong generalization, accurately generating microstructures for descriptor combinations not seen during training. These results highlight the potential of Minkowski functionals as effective design descriptors in machine-learning-driven material design. In addition, they confirm that the CVAE is capable of generating and tailoring porous structures to meet targeted descriptor requirements.

The need for more sustainable and cost-effective energy generation has increased the interest in offshore wind development in deep-water regions. Floating wind turbines offer a promising solution at water depths where bottom-fixed structures are not feasible. Currently, the implementation of floating wind turbines remains limited due to high costs and complex system dynamics. There is a need for efficient, automated optimisation tools that can efficiently identify cost-effective design spaces early in the development process. While optimisation studies have been conducted for semi-submersible and spar-type platforms, limited optimisation related research exists on tension leg platforms (TLPs). This study addresses that gap by developing an optimisation framework for TLPs supporting 15MW wind turbines, allowing for efficient, accurate and cost-effective design iterations while considering platform specific dynamics,.

To achieve this, a frequency-domain model was developed by extending the open source RAFT software. The extensions incorporate features particularly relevant to TLP modelling, including tower flexibility, a sum-frequency force approximation and an analytical tension leg mooring module. The resulting frequency-domain model achieves an error margin within $\pm16\%$ of non-linear time-domain simulations for TLPs with (near) vertical tendons, while reducing computational time by 98.5\%. This makes the model suitable for dynamic analysis within optimisation studies, while enabling TLP-specific response characteristics to be captured at a fraction of the computational cost.

The developed model was integrated into a multi-step genetic algorithm-based optimisation framework to explore the design space, with the objective to reduce the levelised cost of energy (LCOE). The framework, built around a single-column TLP design with four pontoons with tendons at their ends, took into account six design variables to describe the essential system properties. The resulting six dimensional complex design space considers the main column diameter and draft, pontoon diameter and length, tendon angle and tendon pretension. To enable complete design performance evaluations and potentially provide early-stage insights that support certification preparation, each concept was assessed against an extensive set of load cases according to standards, covering both ultimate and fatigue loads in operational and extreme conditions.

The results of the optimisation study identified draft, tendon pretension, tendon angle and pontoon length as the most influential parameters for dynamic performance due to their strong influence on platform stability and mooring stiffness. The column and pontoon diameter had the most significant influence on platform mass and therefore platform cost, while pretension dominated mooring system cost. Despite the identification of these most influential variables, the highly coupled nature of TLPs requires to take all identified design variables into account.

A detailed cost model was implemented, enabling comparison of design concepts within the study, and comparison with floating wind platform designs in other research. To achieve this, the model combines variable platform and mooring costs with fixed lifecycle costs. Multiple optimisation runs revealed several distinct, cost-efficient design spaces, with convergence toward lower LCOE values with increasing iterations of the optimisation. The most cost-effective designs achieved LCOE values around 65 €/MWh, making them competitive with other floating wind concepts across different studies.

Altogether, this work provides an efficient optimisation framework for TLPs in the context of floating offshore wind. It enables accurate and cost-effective design iterations that account for TLP-specific dynamics while significantly reducing computational time. By considering a wide range of load cases and maintaining a balance between speed, adaptability and physical accuracy within a limited degree of uncertainty, the framework offers a holistic approach. This makes it well suited to support early-stage design decisions and concept selection for future deep-sea wind farms using TLPs.

Offshore renewable energy deployment faces a fundamental materials challenge: synthetic fibre mooring lines offer an 8-fold weight reduction and tension reduction compared to steel chains, yet their certification requires predicting 20-year operational behaviour from necessarily limited experimental observations. Current approaches employ polynomial functions to represent stress-dependent material behaviour within established viscoelastic frameworks. Still, these rigid mathematical forms constrain material characterisation by forcing complex behaviour into predetermined equations rather than discovering true material physics. This research develops a neural-physics integration framework that replaces polynomial constraints with adaptive neural networks whilst preserving established Schapery-Prony physics principles. The core innovation employs functional decomposition, where neural networks learn stress-dependent nonlinearity functions (g0, g1, g2), while the proven Schapery-Prony framework handles temporal evolution, enabling the discovery of true material behaviour rather than mathematical fitting to predetermined forms.

A systematic five-phase methodology guides development from baseline optimisation through operational validation. Phase 1 employs 13-dimensional Latin Hypercube Sampling across neural architecture, training protocols, and physics parameters. Phase 2 sensitivity analysis identifies optimal methodological choices: Percentage Difference loss functions for scale-invariant learning, Multiple-Steps loading patterns for comprehensive material exercising, and enhanced relaxation spectra for improved temporal coverage. Phase 3 capability testing reveals model limitations through progressive validation scenarios. Phase 4 diagnostic analysis identifies physics model resolution as the primary constraint, leading to targeted improvement from 10 to 14 relaxation terms that address slow relaxation modes critical for long-term accuracy. Phase 5 validates robustness under realistic marine loading using industry-standard JONSWAP wave spectra. The implementation employs a shared-head neural architecture with shared backbone layers (3 layers, 96 neurons) feeding specialised heads (2 layers, 128 neurons each) for individual nonlinearity function learning. Two-stage training coordinates neural network and physics parameter optimisation through differential learning rates whilst maintaining physical consistency. The twin experiment methodology leverages validated HDPE material parameters to generate unlimited synthetic training data spanning stress ranges and loading patterns that are impossible to achieve through physical experimentation alone.Results demonstrate that the neural enhancement approach enables reliable material characterisation from limited experimental data. Temporal validation achieves 85-95% accuracy across 128-fold duration scaling, enabling decades-long predictions from short-term training data, following established principles for temporal extrapolation in physics-inspired neural networks. Marine validation under realistic JONSWAP loading conditions achieves 5.7-13.2% error across diverse sea states spanning significant wave heights 1.5-7.0m and peak periods 8-15s, demonstrating robustness comparable to validated approaches for viscoelastic modelling under complex loading conditions. The diagnostic insight that physics model completeness rather than neural network architectural sophistication determines model capability guides efficient improvement strategies applicable across computational materials science domains. The validated framework transforms synthetic fibre characterisation from an experimental constraint to a predictive capability, while establishing reproducible principles for neural physics integration across materials science applications that require improved understanding from limited observations. This enables accelerated deployment of offshore renewable energy systems whilst preserving the physical foundations necessary for reliable engineering applications.

Keywords: Neural-physics integration, Synthetic fibre characterisation, Nonlinear viscoelasticity, Schapery theory, Materials discovery, Offshore renewable energy, Computational materials science

The rapid expansion of offshore wind energy, especially into intermediate depths and more non-linear wave environments, has increased the need for accurate prediction of wave-structure interactions, particularly wave run-up and the associated hydrodynamic forces on monopile foundations. These predictions are essential for ensuring the structural integrity, safety, and cost-efficiency of offshore wind turbines. Existing semi-empirical formulas are limited in their applicability, often overestimating run-up heights and failing to capture the complex non-linear interactions that occur under steep (So > 0.03) or near-breaking wave conditions typical of intermediate water depths.

This thesis presents a numerical model based on Computational Fluid Dynamics (CFD), developed using OpenFOAM and the waves2Foam toolbox, to predict wave run-up and the associated hydrodynamic forces on monopiles under varying hydrodynamic conditions. The model was validated against experimental data from de Vos et al. (2007), achieving excellent agreement with less than 5% error in peak wave run-up predictions. Following validation, the model was applied to assess the influence of scale effects, turbulence, monopile geometry, and key wave characteristics on run-up behavior and force magnitude.

The results confirm that scale effects on surface elevation and wave run-up are limited, while hydrodynamic forces show some sensitivity, particularly viscous forces in the vertical direction. Steep, high-energy waves near the breaking thresholds (So ≈ 0.04 to 0.05) produce significantly larger run-up heights and slamming forces, underlining their importance in design considerations. The geometry of the monopile also impacts localized flow patterns and wave wrapping, while turbulence contributes to increased lateral and viscous forces. However, its effect on the critical run-up at the front of the structure remains minimal.

By integrating turbulence modeling, scaling, and mesh optimization, this study establishes a robust, scalable, and accurate numerical framework for simulating complex wave-structure interactions. The model demonstrates significant improvements over existing empirical approaches, especially in intermediate water depths where non-linearity, shoaling, and breaking are present.

While the model shows excellent predictive capabilities, areas for further improvement remain. These include local mesh refinement around the monopile, higher-order discretization schemes to better capture breaking waves, and more extensive validation with recent experimental datasets. However, these enhancements would come at significant computational cost. Given the already high level of accuracy (>95%) in capturing peak wave run-up, the current model offers a practical balance between precision and efficiency. Future efforts should focus more on applying the model to irregular wave conditions and additional hydrodynamic parameters rather than fine-tuning. Researchers with greater computational resources may explore these improvements but should weigh the trade-off between added accuracy and computational expense.

The model configurations are publicly available to support future research and industry adoption: https://github.com/hiddded/MSc-thesis-wave-run-up-on-monopiles

Fiber-reinforced polymer (FRP) composites are used in various engineering applications due to their many advantageous properties, but predicting failure remains challenging due to complex failure behavior. The Computational Mechanics group at TU Delft uses an inter-element cohesive zone method, referred to as the Ortiz model, to analyze microscale crack propagation. Although effective for modeling complex failure processes, the method suffers from severe mesh dependency, restricting large-scale simulations. This research implements and investigates the shifted fracture method (SFM) as an alternative to reduce mesh dependency while maintaining computational complexity. The SFM modifies key aspects of the Ortiz model, including an area correction term, crack propagation algorithm, and shifted weak form equations and cohesive zone conditions using Taylor expansions. Three versions of the SFM were tested: (1) full SFM implementation, (2) reduced SFM with only the area correction and crack propagation algorithm, and (3) an extended Ortiz model with only the area correction term. Numerical simulations show that the reduced SFM is robust, mesh-independent, and efficient in basic fracture tests. The area correction term proved critical in reducing mesh dependency. The extended Ortiz model, on the other hand, showed less accurate results in predicting cracks, despite the additions of the area correction term. Furthermore, the maximum principal tensile stress criterion for determining the crack direction was unsuitable for mixed-mode scenarios, resulting in inaccurate paths. Although the reduced SFM shows promising results, it requires further development to handle complex fracture scenarios like FRP composites. Features such as crack initiation, merging, and termination were implemented and showed potential but require further validation. This research concludes that while the reduced SFM is effective for basic testing, further refinements are needed for broader application to FRP composites.

This study investigates the aerodynamic interaction between two neighbouring offshore wind farms operating in a Conventionally Neutral Boundary Layer (CNBL), where the atmospheric boundary layer (ABL) is neutrally stable, capped by a stable inversion layer and a less stable free atmosphere above. While previous studies have investigated: 1) wake effects between two neighbouring wind farms in a Truly Neutral Boundary Layer (TNBL) with neutral conditions throughout the ABL and above, and 2) the impact of a single wind farm on the flow field in a CNBL, there is a lack of research on the interaction between neighbouring wind farms in a CNBL. Improving our understanding of wind farm interactions in a realistic atmosphere is important for offshore wind farm planning and operation. In this study the performance of the two wind farms under two atmospheric conditions, a CNBL and a TNBL. While the TNBL assumes neutral conditions throughout the ABL and above, the CNBL represents a more complex and realistic modelling approach.

When the wind approaches a wind farm in a TNBL, the combined induction of the turbines creates a zone with increased pressure just before the start of the wind farm, redirecting the flow laterally and vertically. In a CNBL, the vertical component of the redirected wind interacts with a stable inversion layer aloft, which creates an enhanced high pressure area at the start of the wind farm, a phenomena commonly known as global blockage. At the end of the wind farm, the wind is directed down towards the low pressure region in the wake of the wind farm, lowering the inversion layer and creating a zone with even lower pressure. The combination of the two processes described, reduces the wind speed at the start of the wind farm and accelerates it towards the end.

In this study, we look at the interaction of two neighbouring wind farms at varying distances in both a TNBL and CNBL using the relatively fast Multi-Scale Coupled model (Stipa, Ajay, Allaerts, & Brinkerhoff, 2024) which uses a simplified mesoscale model to determine a background flow field which is then used to drive an engineering wake model. Each wind farm consists of a regularly aligned array of five turbines in the spanwise and ten turbines in the streamwise direction with a five rotor diameter (5D) spacing in both dimensions. The turbines are based on the 5 MW NREL reference (Jonkman et al., 2009) and simulations are run at a wind speed of 9 m/s at hub height, on the plateau of the thrust curve. For the CNBL simulations, we choose a Froude number of 1.0 for the inversion layer, resulting in maximum inversion layer displacement due to a phenomena called choking, and a Froude number of 0.10 for the free atmosphere.

Key findings:
1. For infinitely spaced wind farms in a CNBL, lower power output is seen compared to a TNBL due to stronger global blockage effects at the start of each wind farm despite a small speed-up towards the end of the farm.
2. For small wind farm separations (up to ±72D), the speed-up effect at the downstream end of the first wind farm enhances the power output of the second wind farm compared to a TNBL, whilst the blockage effect of the second wind farm negatively impacts the first wind farm.
3. The combined output of the two wind farms at small wind farm spacings is higher than when the farms are infinitely spaced, though lower than in a TNBL.

Offshore floating photovoltaics (OFPVs) show great potential for future renewable energy generation by complementing existing wind energy. To maximize power output, it is essential to accurately predict the response of these structures under wave-induced forces.

Although finite-element fluid-structure interaction (FE-FSI) models can simulate the response of very large floating structures (VLFSs) with high accuracy, they are computationally expensive and time-intensive. To address this challenge, this thesis develops and optimizes three surrogate deep learning models: a Convolutional Neural Network (CNN), a Long Short-Term Memory (LSTM) network, and a hybrid CNN-LSTM model. An iterative trial-and-error approach was employed to fine-tune the hyperparameters and improve model performance.

The CNN model accurately captured spatial features but struggled with generalization across varying sea states. The LSTM model captured temporal patterns well, but had lower accuracy and longer training times. The hybrid CNN-LSTM model combined the strengths of both approaches, achieving a low weighted absolute percentage error (WAPE) of 0.98 %. All models performed well under typical wave conditions but exhibited reduced accuracy in extreme scenarios, particularly for low-amplitude tilts. A sensitivity analysis indicated that these errors were likely caused by insufficient representation of extreme sea states in the training data.

The hybrid CNN-LSTM model was selected as the model that most accurately predicts the response of VLFSs due to its balanced performance, minimal prediction time, and generalization capability across typical operating conditions. While the model shows high reliability for typical sea states, its performance declines for extreme conditions, such as extremely large waves or very small tilts. Despite reduced reliability in extreme conditions, the model provides a promising tool for real-time OFPV response prediction and preliminary studies.

Offshore floating photovoltaics (OFPV) have emerged as a scalable alternative to land-based PV by utilizing vast ocean space near densely populated coastal areas. In addition to wind and wave loads, these systems are exposed to tidal currents, where the resulting current-induced loads scale with the floater’s reference area. As OFPV systems scale in size, accurately predicting these loads becomes increasingly crucial for safe and efficient design. This study investigates the influence of the floater’s shape, specifically its draft and leading-edge bow angle, on current-induced loads.

To this end, a two-dimensional Computational Fluid Dynamics (CFD) model is developed using OpenFOAM to simulate the flow around a single OFPV floater under current-only conditions. The model applies a URANS (Unsteady Reynolds-Averaged Navier-Stokes) approach with the k − ω SST (Shear Stress Transport) model. Hydrodynamic performance is evaluated through pressure drag, viscous drag, lift and pitching moment coefficients.
Results indicate that decreasing the draft, thereby streamlining the overall shape, reduces the downward lift and pitching moment coefficients, while pressure drag remains constant and viscous drag increases. In contrast, streamlining the leading edge of the floater reduces pressure drag, downward lift and pitching moment coefficients, at the cost of a higher viscous drag.
Overall, these findings suggest that streamlining the edges of the floater is the most effective strategy to improve hydrodynamic performance of OFPV floaters.
While the model successfully captures these trends, it is subject to several limitations that present opportunities for future refinements. These include the absence of experimental validation data, simplified environmental conditions, and constraints of the numerical modelling approach.

The growing demand for renewable energy has positioned Floating Offshore Wind Turbines (FOWTs) as a promising solution for harnessing wind energy in deeper waters. Unlike fixed foundation turbines, which are limited to depths of around 50 metres, FOWTs can operate in greater depths. However, this relatively new industry faces numerous challenges, with high costs currently limiting its large-scale deployment. While chains are widely used in mooring systems for the Oil & Gas industry, their required size and supply constraints for floating wind applications have made fibre ropes an attractive alternative. This thesis identifies several factors influencing the use of top chain in a hybrid mooring system, with a primary focus on determining the minimum required length for hook-up operations of semi-submersible FOWTs.

The relationship between the connection height of the top chain to the FOWT and the required top chain length, as well as the relationship between sea state variations and top chain length, were investigated through numerical simulations. Complications related to the depth below the sea surface, fibre elongation behaviour, and the overall performance of hybrid mooring systems were reviewed through existing literature. The dynamic behaviour of the top chain during the hook-up operation was analysed using OrcaFlex, with simulations performed in both frequency and time domains.

Dynamic analyses primarily focused on a dry chain link connection method, revealing that top chain lengths between 61 and 76 metres are required during hook-up operations. These lengths depend on wave conditions, connection height, and the installation vessel. The results showed that the required top chain length increases with wave height and period, while the length itself minimally influences its dynamic behaviour. For the lower segment of the mooring line, at least 10 metres of chain are required to prevent contact between the line and the rudders and thrusters of the installation vessel.

The study concludes that the top chain experiences minimal dynamic tensions relative to the static tension during hook-up, provided it includes sufficient sag and submersion. A bottom connection height offers the most benefits but necessitates a longer top chain during hook-up compared to a middle connection height. While fibre materials show promise as alternatives to chains, further investigation is required for their use in shallow waters. The dry chain link method is considered financially advantageous but it is expected to require the longest top chain length for hook-up operations compared to other methods. An in-depth analysis of applying pre-stretch to a 3-line mooring system is recommended to verify this assumption. These findings contribute to enhancing the economic feasibility of floating wind energy solutions, particularly for large-scale deployment.

This research contributes to a novel non-contact installation method for offshore wind turbines, utilizing electromagnetism for positional control. Floating wind installation requires efficient control methods for payload positioning. Current methods use tuggerlines that are physically attached and work through tensile attractive forces. In contrast, electromagnetic control offers a non-contact solution and allows for control through both attractive and repulsive forces. This research explores the applicability of multiple data-driven control algorithms on a magnetically controlled pendulum system. These algorithms are evaluated for both positional control and motion attenuation under pivot-point excitation, using simulations of a numerical model of the electromagnetic pendulum. The model-free control algorithms considered in this study assumed that changes in the system’s state are solely due to the control output. These methods aimed to optimize the control output based on either the state error or estimates of the state gradient with respect to the control output. However, this assumption was found to be valid only for systems dominated by the control response with limited dynamic behaviour over time, which was not the case for the electromagnetic pendulum system.

The study culminates in the development of a controller based on the online estimation of the magnetic interaction force. This controller, named Proportional-Derivative Force estimating neural network controller (PD-FeNN), is a neural-network based state-dependent PD-controller. The neural network is trained during an operational learning phase on estimations of the magnetic interaction force, which are derived using the model of a linear undamped pendulum. By incorporating the neural network, this approach eliminates the requirement of the previous state-of-the-art controller to manual model the magnetic interaction force based on experimental data. The PD-FeNN controller is tested on both numerical and physical models of the electromagnetically controlled pendulum. The results demonstrate efficient control across a wide range of excitation frequencies and amplitudes for both motion attenuation and positional control. As a successor to the modified PD controller, the PD-FeNN controller improves upon its predecessor by enabling positional control without requiring a model of the magnetic interaction. This advancement enhances the applicability of non-contact motion control for payloads in the offshore wind industry.

Offshore wind farms, critical for sustainable energy production, face the challenge of optimization among many parameters influencing the objectives in conflicting ways. In response, this research introduces the novel Integrative Maximized Aggregated Preference Wind Farm Optimization (IMAP-WFO) framework—a comprehensive tool designed to enhance the flexibility, accuracy, and uncertainty quantification of offshore wind farm design and operation.

Existing methods often fall short due to limitations in adaptability and precision, especially when modeling complex multi-physical behaviors under uncertain conditions. IMAP-WFO represents a fundamental change by combining advanced statistical techniques and simulation methods.

At its core are parametric design performance functions, capturing critical aspects of wind farm behavior, including energy production, material usage, and structural fatigue. These functions rely on Kriging meta-models, known for their accurate predictions of unknown values. To address inherent uncertainty, Monte Carlo simulations provide a probabilistic assessment of outcomes. IMAP-WFO’s true innovation lies in translating technical functions into socio-economic objectives. These include sustainability metrics, Annual Energy Production (AEP), Capital Expenditure (CAPEX), Operational Expenditure (OPEX), model uncertainty, and lifetime fatigue. Stakeholders can dynamically weigh these objectives based on their preferences, showcasing IMAP-WFO’s versatility.

A validation process described in this research ensures the accuracy of design performance functions, comparing simulated results with real-world data from operational wind farms. IMAP-WFO’s application is demonstrated through case studies: optimizing the Levelized Cost of Energy (LCOE) and exploring wind farm control strategies. Overall, IMAP-WFO bridges the gap between technical challenges and socio-economic goals, helping offshore wind farm design to increase in efficiency

This thesis is a about a novel float-over concept for the installation of offshore substations in the Dutch North Sea. This new technology is a topside jacking system integrated into the H-851 float-over barge which should be able to jack the topside to a higher elevation before a float-over installation.

First, a high-level concept design of the novel float-over technology is presented. Thereafter, the stability for the jacking stage is investigated via hand calculations. Analytical models are developed to model the jacking system during select stages of the float-over and jacking of the topside. These models are used to verify the assumed design loads on the jacking system and other components required to install the offshore substation. As of now, the design loads are based on guideline for a different type of float over making the design loads a guestimate.

The float-over barge H-851 proved to be very stable without ballasting during the jacking of the topside. Initial ballast errors did not affect the barges stability significantly, the effect was only a 30% increase in heeling angle for the worst case. The effect for wind loading was more profound as the heeling angle increased with 100%. Due to the large initial stability of the barge, the heeling angle after jacking is still of no concern. For the investigated cases, however, care should be taken to ballasting the barge before jacking as initial errors become larger after jacking.

It was found that the driving systems for the jackings system can be designed in the global horizontal direction using less than 5% of the topside weight, given the sea states in this thesis. A lower stiffness jacking system resulted in lower horizontal loads on the driving system. The jacking system showed a difference in local horizontal loads over the jacks from stern to bow as great as 150%. This is due to the position of the topside with respect to the centre of the barge and in combination with the roll, pitch and heave behaviour of the barge and topside.

The mating analysis had multiple findings. It was found that the impact loads on the LMU’s were significantly lower by using the jacking system compared to the DSF for North Sea waves. However, the effect of the horizontal load (compression force) was less significant which resulted in lower or equal loads on the LMU’s. The horizontal loads at the interface between topside and barge were larger for a jacking system compared to a DSF.

Increasing the stiffness of the jacking system resulted in an increase of loads on the jacks and LMU’s. This was found due to the lower eigen period in the dominant motion for stiffer jacks. An interesting finding is a torsion moment with its centre at the interface between barge and topsides. This point was found to be dependent on the stiffness of the interface.

The offshore wind industry has grown significantly since the installation of the first offshore wind farm in Vindeby, Denmark, in 1991. In recent years, the need to expand offshore wind capacity has led to an increased interest in floating offshore wind turbines (FOWTs), especially for deeper waters where bottom-fixed structures are impractical. This research explores the feasibility and behavior of daisy-chaining FOWTs to reduce mooring costs and anchor points, which currently account for a significant portion of installation expenses and are challenging to install.

The study focuses on spar-type FOWTs, investigating how daisy-chaining impacts the motion, stability, and mooring line forces in different wave and wind conditions. By conducting dynamic simulations with OrcaFlex software, two configurations, triangular and hexagonal, were analyzed to assess the effects of connecting multiple FOWTs. Results indicate that the daisy-chaining doesn’t increase the amplitude of roll and pitch motions compared to a single FOWT, but the system is susceptible to resonance in high sea states, causing excessive yaw motions around the z-axis. These resonances occur due to certain modes in the system being excited by peak wave frequencies, leading to amplified loads.

Despite these challenges, the study demonstrates that daisy-chaining can be feasible if the mooring and connection line design parameters are optimized to avoid resonance issues. The forces experienced in the mooring lines were found to be well within the available safe working load of industry-standard steel wire ropes, but careful design considerations are necessary to mitigate fatigue and ensure stability. This research contributes to the limited literature on shared mooring systems for FOWTs and offers insights into potential cost saving and design strategies for future offshore wind farm projects.

Pile run refers to the phenomenon that occurs during the installation of pile and monopile foundations in soils with low bearing capacity, where soil resistance is insufficient for the controlled driving of the monopile. The increasing frequency and associated risks of such occurrences in offshore installations underscore the necessity for effective pile run mitigation strategies.

This thesis investigates the feasibility of utilizing a mitigation structure to reduce pile run during the impact driving of monopile foundations. The primary objective is to model and analyze the dynamics of the hammer-pile-soil system under impact driving and pile run, with a focus on the impact of a mitigation tool in monopile response.

The research adopts a structured methodology, beginning with a qualitative study to identify effective mitigation measures. A 1-D finite element (FE) model is developed to simulate pile run, while a 2-D FE multiphysics model characterizes the fluid-structure interaction (FSI). The study explores the effects of various geometric parameters of the mitigation structure on forces and stresses experienced during pile run.

Key findings indicate that the implementation of a mitigation structure can significantly reduce pile run velocity, with its performance being influenced by the geometry of the structure and the velocity of the pile. A trade-off exists between velocity reduction and maintaining structural integrity. The research also highlights the critical role of FSI in determining the structural response of the mitigation tool and its interaction with the monopile.

Conclusions drawn from the results suggest practical applications for the mitigation measure, emphasizing the need for further investigation into the design and deployment mechanisms of the structure. Recommendations for future research directions are provided, aiming to enhance the understanding and application of pile run mitigation strategies in offshore engineering practices.

Hyperloop is a new high-speed transportation system currently in development. Utilising an electromagnetic suspension within a low-pressure tube, this approach eliminates the traditional wheel-track friction and drastically reduces the air resistance, targeting speeds up to 1000 km/h. Compared to traditional trains and aeroplanes, the Hyperloop is a more sustainable alternative as it produces no greenhouse gas emissions and uses ten times less energy than road and aviation transportation. The Hyperloop, therefore, has great potential to contribute to the goal of achieving climate neutrality by 2050.

Ensuring the stability of the Hyperloop is a critical challenge in its realisation. Multiple instability mechanisms are present in the system: (1) electromagnetic instability, (2) wave-induced instability, and (3) aeroelastic effects, such as galloping. Additionally, the imperfections in the guideway, make the Hyperloop system prone to parametric instability. This thesis primarily investigates how aeroelastic forces and parametric resonance interact with the electromagnetic and wave-induced instability mechanisms, impacting the overall stability of the Hyperloop system.

Based on the results, it can be concluded that the aeroelastic force destabilises the system, where its impact increases as the velocity rises. As the aeroelastic force continuously injects energy into the system, the unstable domain expands. Furthermore, parametric resonance, represented by ellipse-shaped indentations in the stability planes, is observed when the excitation frequency is twice the natural frequency of the system. Based on the control parameter range used for maglev trains, it is found that the stable domain is narrow, with only a small region prone to parametric resonance. Therefore, choosing the control parameters precisely is essential to prevent instability and parametric resonance.

When examining the combined effect of aeroelastic force and the irregular guideway profile, the results show that the aeroelastic force shifts the overall position of the stability boundaries but it does not affect the parametric resonance regions. Instead, the amplitude of the guideway’s irregular profile influences the size of the ellipses and the stable domain. A larger amplitude leads to an expansion of the unstable domain and an increase in the size of the ellipse; therefore, it is advised to maintain a smooth guideway.

This thesis further evaluates the capabilities of analytical expressions for the simpler 1.5 degree-of-freedom (DOF) electromagnetically suspended mass system by comparing it to systems that incorporate the beam dynamics. The results show that the analytical expressions from the 1.5 DOF system cannot approximate the position of the ellipses for the more complicated systems across all velocities. However, beyond the velocities of $1.3v_{cr}$, the analytical expressions effectively approximate the ellipse size. Notably, while the position and size of the ellipse are influenced by the guideway’s surface roughness amplitude, only the size of the ellipse is affected by the amplitude of the oscillations in the 1.5 DOF system. For the more complicated systems, the analytical expressions of the ellipse size relative to the amplitude provide accurate results, making the simplified system a valuable approach for estimating the ellipse size of parametric resonance in systems with beam dynamics.

Offshore wind energy has become a important contributor to global renewable energy efforts, particularly as the industry seeks to expand into deeper waters where traditional fixed-bottom turbines are no longer feasible. In these deeper waters, floating offshore wind turbines (FOWTs) offer a promising solution, allowing for the generation of renewable energy in locations with stronger, more consistent winds. However, with the increasing use of FOWTs comes a set of logistical challenges that need to be addressed to ensure their commercial viability and large-scale deployment. One of these challenges is the large-scale production of floating foundations. One significant challenge to the commercial success of FOWTs is the slow fabrication rate of semi-submersible floaters, which require extensive welding and assembly. In contrast to faster monopile fabrication, these floaters are more complex and take longer to produce. Since installation campaigns are typically limited to favorable weather periods, especially in the summer months, many floaters need to be ready in advance, placing strain on fabrication yards and causing space constraints. Wet storage provides a viable solution by allowing temporary storage of floaters in coastal areas, freeing up space in fabrication yards for continuous production. This research explores the existing literature
on wet storage mooring systems and identifies key design criteria, environmental considerations, and components such as mooring lines, anchors, and auxiliary equipment.
To evaluate mooring system designs, a Multi-Criteria Decision Analysis (MCDA) framework was developed, facilitating structured comparisons and trade-offs between various design objectives. Although the MCDA enables informed decision making, dynamic analysis is essential to obtain deeper insights into system performance. In this study, frequency domain (FD) analysis was employed for rapid evaluations of the behavior of the mooring system. FD analysis proved sufficient for preliminary studies and is an ideal tool for exploring design variations. It efficiently captures first-order loading effects, which are typically dominant in expected wet storage conditions. However, for detailed design stages, to account for nonlinear behaviors such as slow drift and mooring line stiffness, time domain (TD) analysis are recommended to validate final configurations and ensure compliance with design codes.
Dynamic analysis of the mooring systems compared catenary and taut configurations under various environmental conditions. Concept-specific constraints such as maximum fairlead tension, uplift allowance, mooring line angles, and minimum anchor distances were incorporated into each mooring concept. Furthermore, diffraction analyzes were conducted with varying water depths and floater drafts to accurately define floater behavior under different conditions, and a mesh validation was performed to ensure that the mesh was fine enough for this study. Further validation of the full model was done by comparing the calculated Load Response Amplitude Operators (RAOs) and natural frequencies with
the reference turbine to ensure model accuracy. Tidal ranges and line diameters significantly influenced mooring performance, impacting tensions loads, and floater displacements. The results indicated that the catenary system consistently met these constraints and demonstrated resilience across the full tidal range, where a clear trade-off between tensions and displacements can be observed. However, the taut system experienced tension overloads, particularly at high tide, where it often exceeded line breaking strength, with only a single taut configuration yielding valid results. Despite being more spatially efficient, the taut system’s reliability was compromised compared to the catenary system. Several recommendations for future research are identified to further the understanding of mooring systems in wet storage applications. These include conducting TD analyzes to capture non-linear behaviors, exploring alternative mooring configurations such as Honeymooring or pile fields, refining design standards, and evaluating the economic and logistical feasibility of wet storage solutions. Additionally, investigating different storage locations is essential, as this study focused on a single location, and variations in environmental conditions could significantly influence mooring performance.

This study focuses on optimizing the use of high-performance computing on public cloud infrastructure, along with information theory, for assessing water systems. These assessments are computationally intensive and can benefit from parallel computing and the evaluation of the collected data with information theory. A case study of a water system analysis for the Vlietpolder was conducted to test various cloud configuration settings using an embarrassingly parallel batch computation. The hydrodynamic simulations involved D-HYDRO 1D2D models with different precipitation events and model resolutions. The modelling results were quantified using normalized Shannon’s Entropy to facilitate the comparison of system configurations, evaluating the batch computation process to determine whether enough simulations have been performed and comparing individual simulations.

The study showed that public cloud infrastructure provides comparable computational performance to local computers and servers, and offers opportunities for vertical and horizontal scaling for parallelization. The study also provides insight into the impact of allocated resources, node size, and node type on cloud infrastructure performance. Furthermore, the quantified information derived from the simulations can be utilized to evaluate the batch
computation output and support cost-benefit analyses for selecting configuration settings and model decisions given modeling scenarios.

The study concludes that combining cloud infrastructure and information theory can enhance hydrodynamic modelling for batch computations in water system analysis. The findings provide insights into the potential benefits of utilizing public cloud infrastructure for large-scale computations of hydrodynamic simulations.

Previous research into floating offshore wind turbines consistently shows some manner of discrepancy between numerical and experimental simulations when it comes to low frequency forces and motions. A part of solving this issue is the generation of accurate data for the calibration and validation of the numerical models, needed because of the coupled analysis introduced due to the relatively large wind force. The goal of this research thus became to generate accurate experimental data focussed on the low natural frequency of the mooring system in surge direction. This data had to be accompanied by an uncertainty assessment and should encompass multiple mooring systems for a good comparison.
In order to achieve these goals experimental model tests were set up using bichromatic wave sets in a long towing tank. The bichromatic wave sets allowed for the creation of beating patterns, targeting the specific low frequencies on and around the surge natural frequency via the difference frequencies. The long towing tank reduced the amount of reflections and clutter in the tank. Together these factors ensured accurate excitation of model at the desired frequencies.
The model used was a 1:96 semi-submersible model with three buoyancy columns supporting a central tower column. The depth in the tank being 1.25 m at model scale ensured a relatively deep testing environment. Two different mooring systems were tested, each with a different surge natural frequency but with the same semi-taut fibre rope-chain lay-out. A comprehensive measurement system consisting of force transducers in the mooring lines and wind force device and a camera tracking system enabled accurate measurement of both the force and motion of the model. Experiments revolved around examining the influence of the difference frequency in the wave excitation on the low frequency behaviour of the mooring system and floater.
From the lead mooring line force time series, it could be observed that during the surge natural frequency test the mooring line had triple the force compared to the wave tests with the monochromatic components, showing a clear effect of the bichromatic beating pattern. From the tests it also became clear that mooring system 1 had rear lines which were slack during wind and wave excitation. These slack lines had a number of effects on the behaviour of the model during the tests, such as increasing the mooring line force and surge response.
The tests thus show that the low frequency response comes from the difference frequency in the bichromatic wave sets, as was the goal. The heightening of the response around the natural frequency also shows that getting close to this frequency amplifies the effect. The comparison between the test results and the results from the OC5 project show that this is likely also the case there. At the same time the surge response does not show this heightened response. Which is likely due to the non-linear stiffness of the mooring system.
Furthermore, it was seen that the slack lines have a very large impact on the force in the mooring lines, especially on the lower frequencies and should thus be avoided at all times. The damping ratio as a parameter for the low frequency response is both physically and experimentally (excluding the slacked lined system) consistent and shows potential for development.

The first part of the thesis explores the process of homogenization, particularly for the permeability properties of rock samples. The Hill-Mandel postulate of energy consistency throughout the transitioning of scales is revisited since traditional homogenization methods rely on applying specific boundary conditions to enforce energy consistency. However, it is shown that the applied boundary conditions influence the effective physical parameter and provide upper or lower bound estimations. Recently, it is shown that these boundary conditions influence a layer near the boundary of the sample and that homogenization applied on the subsample away from this boundary layer is not affected by the boundary conditions.

The research focuses on energy consistency by studying the evolution of the energy within the intrinsic subsample, away from the boundaries. With the help of Finite Element simulations of Stokes-flow through idealized structures, the energy of the fluid is traced without the influences of grain properties on the energy dissipation. By plotting the ratio of the energy dissipation of the macro- and micro-scale, it is shown that the energy consistency is not found within small subsamples. Yet, with a growing subsample, energy consistency is achieved naturally, without the enforcement of boundary conditions. As a result, it is concluded that the energy consistency is found at the Representative Elementary Volume (REV), which is a similar requirement as for traditional homogenization methods. The study of the natural energy consistency in idealized microstructures is extended to real microstructures, which include more natural heterogeneities, such as grain properties. It is shown that energy consistency is also found with the natural heterogeneities included, albeit with a slower convergence.

For the homogenization of the permeability of a sample, the energy ratio is now known to be unitary, which can be used as an accurate indicator to determine the size of the REV.

The second part of the thesis explores the process of determining the size of the REV, which is a common, yet essential practice in Digital Rock Physics. Currently, this is an extensive exercise, involving many and large-size simulations to trace the convergence of the physical property and requires a lot of computational resources and time. Numerical-statistical studies have shown that the convergence of the REV visualizes in a cone-like shape. By plotting the convergence for both the permeability and the energy dissipation ratio for idealized microstructures, this study analyses the shape and evolution of the cone of convergence. From this, the generic evolution law of the convergence is determined.

It is shown that the asymmetrical convergence cone is described with a log-normal distribution, with a stable mean throughout the evolution of the cone and a variance for each sample size. The evolution of the variance is described with the law of large numbers, taking into account a reference value. This makes it possible to determine the size of the REV. Since the statistical method applies, information about the error of the fit, the error of the determined homogenized property, and the error of the size of the REV is provided.

The study is extended to real microstructures to validate whether the determined evolution law applies when natural heterogeneities are included. It is shown that the evolution law still accurately describes the REV's convergence. Therefore, the REV's size of real rock samples can also be determined. Even when the REV is not included within the sample, the evolution law can provide an estimate of the size of the REV or the homogenized property.

By using the cone of convergence, it is not necessary to run simulations on the full sample to find the REV, which is computationally expensive, instead running a number of simulations on small subsamples is sufficient, which saves both time and computational resources. It also unlocks the possibility to find the REV for high-resolution samples, as splitting the sample into subsamples allows for smaller simulations.