This research investigates the effects of mechanical loads on bottom-founded and floating offshore wind turbines (OWTs), specifically comparing different floater configurations (TLP and semi-submersible) with bottom-founded turbines (monopile). The study identifies three critical intersections for a fair comparison: the base of the blades, the main shaft, and the yaw system. These intersections allow for the prediction of stresses on components when specific material properties are known, which is often not the case in many turbines.

The modeling program Orcaflex is used to describe the motions and loading of TLP and semi-submersible floating offshore wind turbines (FOWTs). Bluewater Energy Services is currently designing a TLP platform for a wind turbine, and this design, along with a semi-submersible FOWT model, is compared with an IEA 15 MW bottom-fixed turbine. External loads such as waves and wind, generated from North Sea data, are considered. Additionally, the effects of design parameters like weight, waterline area, center of mass, and wind turbine generator (WTG) control settings are taken into account.

The study reveals that the semi-submersible platform is more susceptible to environmental loads, leading to some significant translational and rotational motions. Its stability relies on a large water surface area and a catenary mooring system, resulting in low system stiffness. In contrast, the bottom-fixed and TLP turbines exhibit lower motion fluctuations due to their higher system stiffness. The TLP experiences higher nacelle accelerations compared to the semi-submersible, except for heave acceleration, due to resonance with wave frequencies. Mechanical loadings are significantly influenced by wind speed and the turbine's controller. Before reaching the rated wind speed, mechanical loads increase with environmental loads, while post-rated wind speed, the loads stabilize or even decrease due to the controller's intervention.

Furthermore, the study identifies the driving factors for the lifetime of pitch, yaw, and main bearings. The pitch bearing's equivalent load is predominantly influenced by wind-induced moments, while the yaw bearing's load is largely governed by axial loads from the RNA's weight. The main upwind bearing's load is primarily affected by radial loads, with axial loads becoming more significant as wind loads increase.

The overall conclusion indicates that while platform motions influence system dynamics, their direct effect on mechanical loads is less significant compared to other factors such as wind loads and controller actions. The pitch controller plays a crucial role in managing mechanical loads, particularly for pitch bearings. Nevertheless, the relatively large mean angle of the semi-submersible platform impacts bearing lifetimes. The system's angle, combined with the weight of components, especially for the yaw bearing, is a critical factor in determining their lifetime. These findings are supported by existing literature, confirming the complex interplay between environmental conditions, system motions, and mechanical loadings in offshore wind turbines.

Offshore wind turbines (OWTs) play a critical role in renewable energy, requiring efficient methods to predict fatigue loads on their support structures under harsh environmental conditions. Traditional fatigue assessment methods are effective but costly and impractical at scale. Surrogate models (SMs) offer a cost-effective alternative, but site-specific SMs often fail to generalize across turbines in varying environments and locations. This study explores whether platform-generic SMs trained on a Simulation Database can serve as reliable replacements for site-specific SMs trained on Supervisory Control and Data Acquisition (SCADA) and Structural Health Monitoring (SHM) data for predicting fatigue loads in OWT support structures. The SMs, developed using both deterministic and Bayesian neural networks (NNs), were evaluated in three progressively complex model setups featuring SCADA signals, nacelle accelerations, and tower top strain gauges. Results showed that site-specific SMs generally achieved lower mean average percentage error (MAPE), with deterministic NNs at 47.1m lowest astronomical tide (LAT) in the fore-aft (FA) direction reducing errors from 16.8% to 7.2% and in the side-side (SS) direction from 27.0% to 4.5% as additional signals were introduced. Platform-generic SMs exhibited higher MAPEs due to the broader scope of the model and slight mismatches between simulated and real turbine dynamics, especially in the FA direction. Nonetheless, in the SS direction at 47.1m LAT, platform-generic SMs showed promising performance, achieving errors as low as 7.5% in one configuration. Although Bayesian NNs did not consistently lower errors compared with deterministic approaches, they provided valuable insights into how far test data deviated from the training distribution, helping identify the potential limits of the model. This capability has significant practical implications, as it can potentially serve as a tool for detecting sensor malfunctions or identifying irregular data, ensuring more reliable predictions in real-world applications. Overall, platform-generic SMs still require refinement before they can fully replace site-specific SMs, and Bayesian NNs should not replace deterministic NNs but rather serve as a valuable supplement.

The Zeelandbrug serves as an important connection between Noord-Beveland and Schouwen-Duiveland. The bridge completed construction in 1965 using the balanced cantilever method and has been in operation for nearly 60 years. After the passage of these years, the current status of the bridge’s bearing capacity is unknown. Monitoring devices can be applied to obtain more information about the bridge’s condition and to help extend its lifespan. Hence, this study aims to investigate the application of monitoring strategies for the Zeelandbrug. As part of the process of obtaining the monitoring plan, this research also aims to gain a better understanding of the bridge’s structural behaviour. To achieve this, a linear Finite Element Model using one-dimensional beam elements was developed for the Zeelandbrug. This model was used to evaluate how the settlement of a support affects the magnitude of shear forces within the bridge’s superstructure. A structural assessment on the bridge’s superstructure was conducted, with in longitudinal direction incorporating the additional shear forces resulting from a support settlement. Through this assessment, failure modes were identified together with their associated physical parameters. Lastly, a brief investigation was performed on monitoring strategies that can be applied to monitor these parameters.

The finite element method has proven itself to be an effective method of performing both linear and non-linear structural analyses. This numerical method, however, has several limitations, of one which is that the discretization of complex geometry requires human effort. The rise of the application of machine learning models has opened new possibilities of approaching challenges within scientific fields. The neural network, and more specifically, the physics informed neural network, is a promising method and allows for performing analyses without any discretization that is necessary. This type of neural network, which is less of a black box than the standard neural network, utilizes the concept of partial differential equations to accurately predict solutions.

The objective of this thesis is to perform a structural analysis of a geometrically non-linear Timoshenko beam using a physics informed neural network. The network is built using a variational principle (the principle of virtual work) and a force residual. Furthermore, two optimization algorithms, the adaptive weight loss algorithm and the adaptive activation function are separately used in conjunction with the model to examine the potential improvements on the convergence rate.

It is found that the geometrically non-linear Timoshenko beam can be accurately modeled (relative error of below 2% with respect to the finite element output) with a physics informed neural network. This accuracy can be achieved with a model possessing a relatively shallow size of four hidden layers containing eight neurons each. The adaptive weight loss algorithm and the adaptive activation algorithm both improve the convergence rate of the model, though they are not necessary to maintain the practicality of the model, as the convergence rate is adequate without these. It is recommended that the hyperbolic tangent function is utilized in conjunction with the Adam optimizer. The adaptive activation function can be incorporated into the model to improve the convergence rate significantly without substantially increasing the computational cost of the model.

In this study, we establish a foundational framework for controlling flexible floating structures by integrating state-of-the-art Model Predictive Control (MPC) with a high-fidelity Finite Element model representing these structures. Our specific objective is to enhance the stability of floating structures in offshore engineering applications. We commence with a comprehensive literature review to identify potential challenges, which informs the development of a well-structured research plan that addresses these challenges comprehensively.

From the insights gained during the literature review, we formulate a theoretical framework that serves as the basis for our methodology. We subsequently examine two distinct control scenarios—regular and irregular wave conditions—serving as a Proof of Concept, and we discuss the significant observations derived from these experiments. In conclusion, we summarize our findings and provide recommendations for future research directions.

We utilize a two-way monolithic Finite Element formulation of the flexible floating structure as a representation of the real-world system we aim to control. For our control strategy, we implement Model Predictive Control, which facilitates the incorporation of advanced functionalities, including the application of constraints. To enhance computational efficiency, we employ a Reduced Order Model (ROM) through Dynamic Mode Decomposition with Control (DMDc), trained using open-loop data derived from the Finite Element model. Additionally, we implement a Kalman filter to reconstruct the system's state from sparse and noisy measurements obtained from the floating structure.

We design two Reduced Order Models specifically for controlling the floating structure under the aforementioned conditions. Prior to executing the control strategies in these distinct scenarios, we conduct an in-depth investigation of DMDc, exploring its relationship with Koopman theory. We generate open-loop data free from pollution using the Finite Element model, which is subsequently utilized to derive the Reduced Order Model. A convergence study is performed by analyzing the eigenvalues and amplitudes of the DMDc, following methodologies established in prior research. We validate the DMDc models against validation datasets, allowing us to select the model exhibiting the smallest validation error. Furthermore, we ensure that the training datasets inherently encompass the wet modes by employing Proper Orthogonal Decomposition.

Our initial control scenario involves managing the floating structure under regular wave conditions. This case is pivotal for acquiring fundamental insights into the control mechanisms. We observe that the frequency of the control input aligns precisely with the excitation frequency. Subsequently, we extend our study to encompass the control of the floating structure in irregular wave conditions, characterized by a sea state defined by the JONSWAP spectrum. Consistent with our findings from the regular wave scenario, we discover that the control mechanism exploits the natural frequencies of the floating structure, which closely correspond to the energy-dense region of the sea state's spectrum. This is achieved by amplifying the reflected wave, thereby counteracting the incoming wave and reducing energy input into the system.

In conclusion, we advocate for further research to bolster the proposed method by examining various structural properties and wave environments, thereby providing robust evidence to validate the approach presented in this study. Our findings indicate that the integration of advanced control strategies, such as MPC and DMDc, holds significant promise for optimizing the performance and stability of flexible floating structures in dynamic offshore conditions.

Structural joints influence the design strength, material requirement of a structure. Structural joints experience damping dissipation due to friction damping or hysteresis damping. Damping is often used for reducing the vibrations in a structure. However, large amount of energy dissipation leads to deterioration of the material used for constructing the joint. Hence it is important to identify the system parameters like stiffness, viscous damping, friction force as well as the hysteretic restoring force that cause the energy dissipation in the structure.
For identifying the uncertain system parameters like stiffness, viscous damping and magnitude of friction force, the SINDy algorithm is extended by using stick and slip temporal constraints. This is done by segregating the data of external forcing and response of SDoF system, applying the existing SINDy algorithm and applying the sticking and slipping conditions in the time domain. The proposed Extended SINDy approach estimates the system parameters more accurately compared to the existing SINDy algorithm.
For studying the hysteresis in the structural joints, a pinned column base-plate was considered in an elastic region. Further, the Dahl model with different slope parameter for each branch of moment-rotation hysteresis is employed. The correct values of parameters are estimated using the Bayesian Optimization technique. This procedure yields a functional form representing a resisting hysteretic moment-rotation behaviour in a structural joint with good accuracy.

This thesis presents a novel methodology to determine the reliability of cantilever grandstands under dynamic crowd loading. No method currently exists to evaluate this, despite the occurrence of significant vibrations in such structures, for example in the Feyenoord Stadium in Rotterdam. These vibrations can negatively affect both the safety and serviceability of such structures, so a method able to evaluate both is desirable. The methodology consists of four components: a model of a grandstand, a model of the dynamic crowd load, a failure criterion, and a reliability analysis. Of these, the first and third component are case-specific, as a wide range of different grandstands and different criteria could be considered. In this thesis, the methodology is applied on a case study based on the Feyenoord Stadium, with a model of the trusses beneath the main grandstand of this stadium, and a criterion based on the intensity of the vibrations of the upper cantilever of this grandstand, a serviceability issue. The second component represents the most novel aspect of the methodology: a model describing the dynamic crowd load, defined in the frequency domain, and generally applicable regardless of the structure being considered. The core of this model is a parameterisation of the amplitude spectrum of the load applied by a group of jumping spectators. The parameters describing the spectrum are considered as stochastic variables, with distributions fitted using samples from a state-of-the-art database of measured loads. As opposed to the first and third components, this load model should be generally applicable.The parameters of the load model, together with other stochastic variables in the grandstand model, form the input of a Limit State Function. The output of this function determines whether the structure loaded by the crowd meets a certain criterion, and forms the basis of a reliability analysis together with the distribution of the stochastic variables. The probability of \emph{not} meeting the criterion, the failure probability, is calculated through a number of reliability methods in the final component of the methodology. These methods should be generally applicable as well, though which methods fits the best could still depend on the case being considered. This methodology was successfully applied on the case study, and a failure probability was found with two different methods. One of these, Crude Monte Carlo, allowed for a more fluid view of the concept of `failure' when a serviceability criterion based on a subjective limit is considered. The other method, SDARS, returned a slightly larger failure probability, in a much shorter runtime than Monte Carlo. This method is therefore more fit for safety criteria with a small (expected) failure probability, for which Monte Carlo would require an unreasonably long runtime. In addition, the wider applicability of the methodology has been investigated. An important requirement for this is flexibility with regards to the applied reliability methods. Two other methods, Directional Sampling and SDARS, were also attempted but did not yield usable results. This is caused by the manner in which phase angles are defined in the frequency-domain model. Random values are drawn, which causes noise to appear in the limit state function: a constant input does not lead to a constant output. Being able to use different reliability methods is an important requirement for applying the methodology in a wider variety of cases. In order to apply these methods, the phase angles need to be defined in a manner which does not introduce noise. Another important requirement is the ability to consider different failure criteria. While the criterion applied in the case study could be evaluated in the frequency domain, many others will require the results to be transformed to the time domain. Applying this transformation to the frequency-domain results of the models currently does not yield usable results, which is likely also caused by the definition of the phase angles in the load. In conclusion, while the methodology is applicable on the case study, the definition of the phase angles requires attention before it can be applied on other cases.

Bayesian system identification has been extensively adopted in Structural Health Monitoring as a way to probabilistically infer unobservable parameters of the physical model of a structure using measurement data. Combining the Bayesian approach with distributed optic fibre sensors can potentially improve the accuracy and reduce the uncertainty in parameter estimation problems, given the large amount of quasi-continuous data produced by this sensing technology; however, its computational cost could be prohibitively high when using conventional methods since Bayesian inference typically involves a large number of samples, each of which entails a physical model evaluation.
The focus of this work is on performing Bayesian system identification for real-world civil engineering structures within an acceptable running time, while using optic fibre measurements with a high spatial resolution. The proposed methodology employs a cheap-to-compute Gaussian process (GP) surrogate that replaces the main bottleneck of the Bayesian workflow for these type of problems: the evaluation of the log-likelihood. The GP surrogate is actively built by sequentially selecting new training points in areas that are expected to highly contribute to the accuracy of the posterior distribution. Once convergence is achieved, the surrogate is used to obtain the parameter estimates via Markov chain Monte Carlo (MCMC) sampling. Additionally, in order to accelerate the Bayesian workflow, cloud-based parallelization is used to perform multiple finite element analyses simultaneously.
A first synthetic case with an inexpensive frame model is used to test the methodology for problems with two and five probabilistic parameters. An encouraging outcome is obtained with the actively learned GP surrogate, with posterior distributions very close to the full MCMC procedure while requiring a number of physical model evaluations orders of magnitude lower. 
After that, a second case study consisting of an existing reinforced concrete bridge with real measurements and a relatively expensive finite element model is investigated. A subset of discrete strain and translation sensors are used to perform an initial parameter estimation that almost exactly resembles the results from previous research on this bridge by Rózsás, et al. (2022), successfully validating the proposed procedure. Then, another parameter estimation is computed using available high resolution optic fibre measurements, after which is shown that the optic fibre provides the best improvements in model predictive capacity among all sensor groups, confirming its potential when combined with Bayesian system identification.
The results of the case studies indicate that the approach presented in this thesis has the capacity to greatly reduce the wall-clock time of Bayesian parameter estimation for real world civil engineering structures with optic fibre measurements, while maintaining a high degree of accuracy. Nevertheless, additional research is required for cases where the statistical parameters governing the measurement and model uncertainty are inferred along with the physical parameters.

Dynamic behaviour often is a topic of concern in the design of pedestrian bridges. It is time-consuming to assess and little is known about the way it is influenced by the characteristics of a bridge. In many cases, a bridge needs testing after is has been finished to determine the specifications of a damper.

Striving to reduce the environmental impact of bridges, there lies a great potential in using materials with a low environmental impact, such as timber. This research combines the lack of knowledge about dynamic behaviour of footbridges with the need for using timber instead of other materials. It consists of two parts. The first part, the parameter study, investigates the influence of three preliminary design parameters on the dynamic behaviour of a long-span timber footbridge, namely the pylon height, the pylon shape and the amount of cables. The second part, the optimisation study, examines to what extent it is possible to design a long-span timber footbridge that does not need dampers to control excessive vibrations.


To this end, a parametric model of a bridge was made in which parameters can be varied and optimised to create realistic design variants. To be able to optimise taking into account dynamic behaviour, a python script was written to automatically determine the type of modes. The results of the parameter study show that the dynamic behaviour can be influenced by the parameters, although the results depend on the specific model, dimensions, parameter values and damping value. The results of the second part show that a with a 14% increase in mass a design variant that does not need dampers to control excessive vibrations can be obtained.

The dynamic behavior of structural components can largely change in the presence of damages. Understanding this behavior is of particular importance for critical engineering systems, and in particular bridges. Damage identification methods forms a key objective in structural health monitoring of bridges so many researches have been conducted in this area. In this thesis, damage identification techniques on beam bridges under moving vehicle loads will be presented in order to produce useful conclusions about the assessment of existing bridges by investigating various numerical applications of different scenarios.
The first objective of this thesis is to derive the analytical expressions needed to be able to predict the dynamic response of many different cases of bridges so that as many real scenarios as possible can be treated. This means that these expressions would be used to investigate damaged beam bridges that can be modelled as an assembly of beams with any number of different material properties, any type of interface or boundary conditions and any number of cracks. For this reason an approach to analyze the bridge as an assembly of n piecewise homogeneous damaged Euler-Bernoulli beams jointed at their edges, will be presented, using the generalized functions to obtain a single expression of the solution which depends on the 4 integration constants associated with the boundary conditions. The closed-form expressions of these 4 constants will be provided. Furthermore, in the presence of internal or externals springs, translational or rotational, additional constants representing the discontinuities have to be taken into account and are computed by considering one additional condition for each discontinuity. The feasibility of this approach and the corresponding analytical formulations is shown with two numerical applications that include all the different capabilities mentioned. Moreover, the implementation of these expressions in a deterministic approach for damage localization is presented, mainly as another example of the many possibilities of the use of analytical formulations instead of other approaches and as an introduction of the so called Inverse Problem with deterministic and probabilistic methods.
The second objective concerns the optimization of damage identification on bridges by comparing different quantities that are evaluated while measuring the response of the bridge (direct monitoring) and the response of the moving vehicle when it passes along the bridge (indirect monitoring). First, the governing equations for the dynamic response of these models are derived, considering the crack(s) as a rotational spring, the bridge as an Euler-Bernoulli beam (or multiple with different properties) and the moving vehicle as a spring-mass system. In this manner, the dynamic response of the bridge is calculated (modal characteristics and displacement) as well as the one of the moving oscillator (displacement and acceleration) and the reaction force acting on the surface of the beam from the moving vehicles. Numerical applications with different beam properties and different number of cracks are performed, using MATLAB for the analytical expressions and SAP2000 for the finite element model, to derive the optimal quantity to be used for damage identification. Lastly, the results are validated by considering and comparing an alternative way of modelling crack, namely as a zone with reduced rigidity, for the same numerical examples, leading to the same conclusions about the crack(s) identification.
Last but not least, the third objective of this thesis is to be able deal not only with the widely used time-invariant damages, namely the always-open crack model, but also with time-variant damages and in this case with the switching crack model. To achieve this, the analytical expressions for the closed-form solutions of the mode shapes derived for the always-open crack are modified to be able to tackle the switching crack model by introducing a Boolean switching crack array which identifies open cracks, modelled as rotational springs. These new expressions would still be able to be used for any number of Euler-Bernoulli beams, any type of interface or boundary conditions and any number of switching cracks. Then, the governing equations for the dynamic response of this model are derived, considering the moving vehicles as moving masses in order to validate the approach with numerical examples existing in the literature and then by introducing its new capabilities. Further, as the computational strategy has been validated, a comparison between time-variant and time-invariant damages is performed concerning crack identification, so that the reader would recognize the importance of understanding the dynamic behavior of different ways of modelling damage in complicate engineering systems like bridges.

Bayesian system identification, including parameter estimation and model selection, is widely used to infer partially known, unobservable parameters of the models of physical systems when measurement data is available. A common assumption in the Bayesian system identification literature is that the discrepancy between model predictions and measurements can be described as independent, identically distributed realizations from a univariate Gaussian distribution. However, the decreasing cost of sensors and monitoring systems leads to more frequent structural measurements in close proximity to each other (e.g. fiber optics and strain gauges). In such cases, dependency in modeling uncertainty could be significant, both in space and time, and the assumption of uncorrelated Gaussian error may lead to inaccurate parameter estimation.

The aim of this thesis is to explore how Bayesian system identification can be feasibly performed using large datasets when spatial and/or temporal dependence might be present and to assess the impact of considering this dependence. A pool of models, each assuming a different correlation structure, is defined and Bayesian inference is performed. In particular, stress measurements obtained on a steel road bridge are used to update the parameters of the corresponding FE model and the parameters of the correlation structure. The results are compared to a reference model where only measurements of the response peaks are used under the assumption of independence. Nested sampling is utilized to compute the evidence under each model and Bayesian model selection is applied. The question of efficiently performing system identification for large datasets (N > 10² for temporal dependencies and N > 10³ for combined spatial and temporal dependencies) is investigated, and a novel approach for efficiently calculating the exact log-likelihood is derived. An approximation based on the Fisher information matrix is used to efficiently calculate the information content of measurements.

It is found that the choice of correlation function can significantly affect the posterior distribution of the model prediction uncertainty. Additionally, it is shown that using large datasets and considering dependence makes it possible to perform system identification for a larger number of parameters compared to the reference model. The results of the case study indicate that using measurements from multiple sensors under combined spatial and temporal dependence and additive model prediction error yields reduced uncertainty in the posterior and up to 29% reduction of the posterior predictive credible interval range compared to the reference case. Furthermore, the efficiency of the proposed likelihood evaluation method is assessed. Using this method, exact calculation of the log-likelihood can be performed for >10⁶ points in under a second in the case of correlation in one dimension. For combined spatial and temporal correlation it is shown to be approximately 900 times faster than naive evaluation for a 64 by 64 grid of observations. The results of the case study indicate that the described approach can be feasibly applied to real-world structures and can potentially improve parameter estimation and reduce prediction uncertainty. These findings suggest that further research into the approach could yield improvements over current methods.

A new semi-analytical model is developed to predict the dynamic response of slender gates in combination with an overhang, which are subjected to standing waves. Still little is known for these types of gates for this specific forcing type. The semi-analytical model is still in its development phase and must be validated on its performance. This thesis aims for validation of the modal calculations performed by the Semi-Analytical model for submerged gates, which include fluid structure interaction. Several experiments were executed in the wave flume at the faculty of Civil Engineering at the Technical University of Delft. Dry hammer tests, wet hammer tests and wave tests were executed for different water levels. Different plates were investigated in the experiments: Solid Plate and Reinforced Plate. A measurement plan was designed to obtain reliable results from the measurement devices (strain gauges and accelerometers). The experimental data was subjected to an Experimental Modal Analysis algorithm. The Frequency Domain Decomposition turned out to be most suitable for this situation. The mode shapes that were found were subjected to a Modal Assurance Criterion (MAC), in order to compare them with the semi-analytical prediction. The results from the wet modal analysis showed that the Solid Plate had good correspondence with the prediction of the semi-analytical model. The mode shapes turned out to have a high MAC values, while the natural frequencies showed small relative errors for the modes under consideration. The Reinforced Plate was less accurate. In the wave experiments the first three modes were found, which were also believed to have the highest energy input. Considerable high MAC values assured that the identified modes were indeed the same as the modes of the semi-analytical prediction. The natural frequencies showed larger errors for especially the first mode (approx. 30%). It was observed that several measurement errors might have influenced the results of the Reinforced Plate. After finalizing the experiments for this plate, I discovered that the stiffeners and front plates came loose from the U-frame. The dynamic quantities of the Reinforced Plate were therefore adjusted during the experiments. The datasets of the experiments that were most trustworthy were selected for the different analyses. The Regular Wave Impact experiments showed good correspondence and where therefore assumed to be correct. The Solid Plate was concluded to behave as predicted by the semi-analytical model. High correlation between the predicted and identified modes and small errors in the natural frequencies were observed. The data from the Reinforced Plate showed that the high energy modes were identified for the wave experiments. Modal shapes had high correlation between predicted and identified ones, while the natural frequencies had somewhat large errors. It was observed that small natural frequency errors for the input modes resulted in relatively small errors for the calculated modes. Further validation of the model should focus on the prediction of maxima and time series of the response. The step from modal analysis to a time series is a final step in the semi-analytical model.

As buildings get taller and lighter, structural engineers are increasingly faced with the consequences of the dynamic response of high-rise buildings to wind. Damping is an important property for the dynamic response of high-rise structures, but is a combination of many mechanisms, which makes it a complex phenomenon to account for in the structural design. Empirical damping predictors currently exist, but a large scatter is found among predictors, as well as between predictors and identified damping from measurements. Damping values are prescribed in codes, but are not consistently conservative. Therefore, there is a strong desire from both structural engineers and researchers to obtain further understanding of damping behaviour in high-rise structures.

Damping identification techniques, such as the Half-power Bandwidth method and the Random Decrement technique are commonly used. However, these are not applicable to buildings with closely spaced modes, and require extensive measurements. Besides, they only provide a damping value, and cannot find damping of separate components of a system. A novel technique, the Energy Flux Analysis, approaches damping from an energy point of view, making it more widely applicable, and allowing for damping identification in components of a structure. The Energy Flux Analysis has been verified to lab structures, but its performance when applied to a high-rise structures using in situ measurements is still unknown.

The aim of this research is to investigate the sensitivities of and prerequisites for the application of the Energy Flux Analysis to high-rise buildings excited by wind using spatially limited measurements. The sensitivities were sought for in the uncertainty of required input for the Energy Flux Analysis: structure motion, which includes internal forces, wind load, data acquisition, and structural properties. The research was performed through application of the Energy Flux Analysis to the New Orleans tower in Rotterdam.

While the sensitivity to structural properties and the magnitude of measurements is limited, the Energy Flux Analysis demonstrated to be highly sensitive to the phase of structural motion, internal forces, and wind load. The first two points are relevant for computing the energy flux at the boundary of a system, when one is interested in damping in the superstructure and due to soil-structure interaction separately. The last point is relevant when one is interested in the total or superstructure damping.

The phase differences occurring between structure motion and internal forces are a direct result of damping. Material damping resulted in a phase difference between stress and strain in the numerical model, while a local damper resulted in a phase difference between structure motion at different locations. The many damping mechanisms occurring in a high-rise structure may each affect the phase of structure motion and internal forces differently. When these phase differences are not taken into account in the Energy Flux Analysis, for instance due to extrapolation of measurements, an erroneous result will be obtained. A brief investigation was performed as to whether these effects can be expected in true structures, but additional research is required.

The fluctuating wind load at the natural frequency of the structure is dominant for the flux of energy from wind to the structure, which is obtained by multiplication of the wind load with the structure velocity. Again, the phase of this wind load is highly important. When measured at one location, little is known about the phase of the wind load at other heights. Different approaches of extrapolating the measured wind load demonstrated a large scatter in the Energy Flux Analysis results. In this research, the Energy Flux Analysis found not to be repeatable, which was proven to be a direct result of the phase difference between the measured wind load and structure velocity. Possible causes for this varying phase difference were formulated.

It is essential, but due to the major advantages of the Energy Flux Analysis also profitable, to perform further research into its application to high-rise structures. Therefore, this study provides extensive recommendations mostly focused on simple numerical and lab experiments.

Under urban sprawl the trend of new established complex structures has rapidly increased. In this context little importance has been given to maintenance, even if this represents an important step in combating and avoiding disasters and developing improved future structural designs. Over the last years low-cost Global Navigation Satellite System (GNSS) equipment has faced rapid and important development opening a new door to reliable and high accuracy positioning applications such as structural health monitoring (SHM). This study focuses on assessing, from a geodetic perspective, the capabilities of a pair of low-cost dual frequency GNSS receivers for capturing the kinematic response of structures to wind. An experiment has been carried out with a stainless steel cantilever beam, aiming to highlight the advantages of employing a differential GNSS system for monitoring low frequency changes in the structure’s body. Hence, in this context the nominal precision of the GNSS system in East, North and Up direction of 4, 5 and 10 millimeter (1σ), was further improved to 3, 4 and 8 miilimeters in the presence of a Global Positioning System (GPS) based multipath (MP) correction. However, it is safer to consider that the true displacement retention potential of the low-cost GNSS receivers corresponds to 3 times (3σ) the aforementioned standard deviation values, resulting in slightly larger than 1 centimeter detectable horizontal displacements, and up to 2.4 centimeters vertical displacements. To support this, wind-induced beam displacements of up to 1.9 centimeters were identified and attested based on a cross correlation analysis with meteorological information. Next, the architecture of a GNSS based SHM system is proposed that can detect structural displacements in real time and rise safety alarms. Therefore, with real time kinematic (RTK) differential positioning and a position outlier and slip statistical testing procedure, a clear strategy for the estimation and identification of uni- or tri-dimensional displacement quantities in real time is proposed, to rise alarms about the magnitude and the direction of identified displacements. Hence, there is no doubt that newly released low-cost dual frequency GNSS receivers represent an alternative to high-end geodetic equipment for SHM, by offering an optimal balance between precision and cost efficiency.

In order to assist project developers in efficient offshore wind farm installation planning a decision-support tool has been developed. The tool combines Discrete Event Simulation and Black-box Optimization. The focus of the development was on providing the flexibility allowing a user to compare various installation strategies and impact on the total installation cost and vessel workability. Synthetic weather generation with Markov Chains was implemented in order to account for weather effects. Recomendations have been given for weather modelling (for the purpose of installation planning) and for applying different families of black-box optimization algorithms.

Heerema Marine Contractors (HMC) is a contractor in the international offshore oil, gas and renewables industry. It is specialized in transporting, installing and removing large offshore facilities. HMC operates three crane vessels. Two of which are semi-submersibles (Thialf and Balder), the other is the monohull Aegir. A third semi-submersible, the Sleipnir, is currently under construction. To ensure safe operations, make accurate fatigue predictions and extend operational windows, HMC relies on vessel motion calculations. Currently, vessel motion estimations are based on response amplitude operators calculated by the diffraction software package WAMIT. As HMC cannot rely on diffraction software in case of non-linear vessel motions, the need for a method capable of capturing non-linear effects arises.

The goal of this study is to determine the potential of a neural network based model in predicting hydrodynamic behavior of semi-submersible crane vessels. Hindcast weather data, vessel motion measurements and model test data are used to train several different neural network architectures. The research into the potential of neural networks in predicting hydrodynamic behavior is split into two main categories: the frequency domain and the time domain.

Within the frequency domain, wave forecasts can be used to predict a response spectrum. The neural network in this case acts as a conventional RAO. In an artificial environment, four architectures are tested and the results show that neural networks are able to make accurate predictions in a fully linear environment. When tested on project data, where the vessel sails at operational draft, the neural network predictions shows a slightly higher accuracy than the diffraction based predictions for the specific test case. Another network is tested on transit data, where the vessel sails at an inconvenient draft. The results from these tests show that there is potential for a neural network to be used as a substitute for Response Amplitude Operators.

The time domain models focus on predicting ship response based on surface height signals and/or hindcast vessel motion measurements. The first model is trained and tested on model test data from an SSCV. The input of the neural network is surface height measurements and the output is pitch motion prediction. The model shows that it is capable of predicting both first and second order pitch motions. Another time domain model has MRU roll measurements as input and it tries to predict the future 60 seconds of roll motion. Many network topologies and optimizer settings are tested but none are capable of predicting future motions.

Wind power to cold climate sites is attractive because of favorable wind conditions and low population density. However, icing of wind turbine blades remains one of the main challenges for cold climate sites. Ice formation on wind turbine blades causes several problems, such as the loss of power production, unbalanced loads in drive train and ice throw. Most of the existing detection methods need special sensors installed and would be expensive to achieve a satisfied accuracy. In this research, an alternative model-based ice on blade detection method is proposed. There are mainly two advantages of the method. Firstly there are no additional measurements are needed. Secondly the detection method can be implemented for any kind of blade aerodynamic changes, not only ice on blade.
The basis of the mode-based ice-detection method is a reliable linearized wind turbine model. The accuracy of the model is mainly influenced by the number of degrees of freedom (DOFs) and the number of operation points (OP). The most efficient model is the one with less DOFs and OP but without losing much accuracy. The relative importance of DOF and OP is revealed and suggestions are given for an optimum choice of DOF and OP for different quantities. Specifically, for power production properties, increase the number of OP is more efficient. While, for blade and tower related properties, increase the number of DOF is a better choice.
The ice on blade influence of aerodynamic force, power production and structure loads is studied for three ice conditions, which are start ice, light ice and moderate ice. Based on the ice on blade influence and the reliable linearized wind turbine model, the concept of model-based ice-detection is proposed as follows: Firstly ice on blade changes the aerodynamics and the blade mass. Therefore it can be considered as one wind turbine system changed to a different system. That means the outputs of the two systems, with ice and without ice, are different, even the inputs are exactly the same. From the ice-detection point of view, if the difference of outputs between iced system and cleaned one is much larger than modeling error, ice on blade can be detected.
With the proposed model-based detection method, the ice-detection capability of different output quantities is studied. It has been found that the power production related quantities usually have the highest ice-detection capability, while the blade and tower properties have relatively lower capability. In order to verify the ice-detection method for a broad working conditions, implementations are performed for above and below rated wind speed conditions. Successful detections have been achieved for both of them.
For possible measurement errors in reality, they are efficiently considered by introducing uncertainty factors. The ice-detection capabilities with measurement error have been analyzed for three ice conditions and detected based on different quantities. Although a drop in the detection capability has been observed if considering the measurement error, all the quantities still can detect the ice successfully to some degree.

One of the main drawbacks of offshore wind energy is its high levelized cost of electricity (LCOE). One explanation for these high costs arises from the large amount of uncertainties that come with such complex projects. As a result of this complexity, it’s difficult to foresee the course of events in advance. Decisions made during the preliminary design phase are often based on limited data while having a great impact on project outcome. A tool is needed to assist engineers during these early stages of design. This tool should deliver good results using limited input data and require little computational time. An example of such a tool is TeamPlay; an engineering model that evaluates offshore wind farm design using automated optimization techniques. In its current state, only a monopile type foundation is analyzed within this process. Since the current trend in the industry is to build larger farms further from shore, this foundation type will become less popular in the future. A jacket type foundation is better suited for far-­‐shore application. In this thesis, a new stand-­‐alone engineering model is created which can be added to TeamPlay’s functionality to extend future use...