Floating wind turbines (FWTs) are considered a new lease of opportunity for sustaining growth from offshore wind energy. In recent years, several new concepts have emerged, with only a few making it to demonstration or pre-commercialisation stages. Amongst these, the spar-buoy based FWT has been extensively researched concept with efforts to optimise the dynamic response and reduce the costs at acceptable levels of performance. Yet, there exist notable lapses in understanding of these systems due to lack of established design standards, operational experience, inaccurate modelling and inconsistent reporting that hamper the design process. Previous studies on spar-buoy FWTs have shown inconsistencies in reporting hydrodynamic response and adopted simplified mooring line models that have failed to capture the coupled hydrodynamic behaviour accurately. At the same time, published information on drive-trains for FWTs is scarce and limited to geared systems that suffer from reliability issues. This research was aimed at filling the knowledge gaps with regard to hydrodynamic modelling and drive-train research for the spar-buoy FWT. The research proceeds in three parts, beginning with numerical modelling and experimental testing of a stepped spar-buoy FWT. A 1:100 scale model was constructed and tested in the University of Edinburgh’s curved wave tank for various regular and irregular sea states. The motion responses were recorded at its centre of mass and nacelle locations. The same motions were also simulated numerically using finite element method based software, OrcaFlex for identical wave conditions. The hydrodynamic responses were evaluated as Response Amplitude Operator (RAO) and compared with numerical simulations. The results showed very good agreement and the numerical model was found to better capture the non-linearities from mooring lines. A new design parameter, Nacelle Magnification Factor, was introduced to quantify coupled behaviour of the system. This could potentially encourage a new design approach to optimising floating wind turbine systems for a given hub height. The second part of the research was initiated by identification of special design considerations for drive-trains to be successfully integrated into FWTs. A comparative assessment of current state of the art showed good potential for directdrive permanent magnet synchronous generators (PMSG). A radial flux topology of the direct-drive PMSG was further examined to verify its suitability to FWT. The generator design was qualified based on its structural integrity and ability to ensure minimal overall impact. The results showed that limiting the generator weight without compromising air-gap tolerances or tower-foundation upgrades was the biggest challenge. Further research was required to verify the dynamic response and component loading to be at an acceptable level. The concluding part of research investigated the dynamic behaviour of the directdrive generator and the various processes that controlled its performance in a FWT. For this purpose, a fully coupled aero-hydro-servo-elastic model of direct-drive FWT was developed. This exercise yet again highlighted the weight challenge imposed by the direct-drive system entailing extra investment on structure. The drive-train dynamics were analysed using a linear combination of multi-body simulation tools namely HAWC2 and SIMPACK. Shaft misalignment, its effect on unbalanced magnetic pull and the main bearing loads were examined. The responses were found to be within acceptable limits and the FWT system does not appreciably alter the dynamics of a direct-drive generator. Any extra investment on the structure is expected to be outweighed by the superior performance and reliability with the direct-drive generator. In summary, this research proposes new solutions to increase the general understanding of hydrodynamics of FWTs and encourages the implementation of direct-drive generators for FWTs. It is believed that the solutions proposed through this research can potentially help address the design challenges of FWTs.

La fonction principale des systèmes d'ancrage des éoliennes offshore flottantes est de limiter les mouvements du support. Les lignes d'ancrage qui les composent sont typiquement constituées de chaînes, de câbles aciers, de câbles synthétiques ou d'une combinaison de ces composants.Dans cette thèse, on se concentre sur les câbles en acier qui permettent de réduire le poids et d'augmenter la résistance en tension par rapport aux chaînes. Leur dimensionnement dépend des chargements en tension et flexion, liés aux mouvements du flotteur sous l'action de la mer et du vent.L'objectif de la thèse est le développement d'un nouveau modèle numérique pour prédire la durée de vie en fatigue des câbles d'ancrage d'une éolienne offshore flottante. Il doit notamment simuler les glissements relatifs entre les fils au cours d'une flexion du câble. Des résultats d'essais de tension-flexion de la littérature ont en effet montré que la première rupture est localisée près du plan neutre de flexion, où ces déplacements relatifs sont les plus grands. Cet effet majeur sur la durée de vie du câble n'est pas pris en compte par les lois de fatigue en tension-tension des normes de design offshore actuelles.Il faut aussi remarquer que l'utilisation d'un modèle détaillé de câble dans une démarche de dimensionnement à la fatigue représente un vrai défi. Le nombre élevé d'interactions de contact à modéliser, de l'ordre de plusieurs milliers par mètre de câble, et le grand nombre de cas de chargement rendent ce type de calculs très coûteux.Les chargements qui sont utilisés dans le modèle local de câble sont issus de calculs globaux réalisés à l'aide d'un logiciel multiphysique (Deeplines). Ce logiciel permet de simuler les conditions environnementales (vent, houle, courant) appliquées sur l'ensemble de la structure offshore.Nous montrons que le comportement non linéaire en flexion du câble, lié aux interactions de contact entre les fils, n'influence pas significativement les résultats du modèle global. Cette observation justifie une démarche de type descendante, les calculs globaux pouvant être réalisés en première étape. Les évolutions temporelles des tensions et courbures globales sont appliquées uniformément sur le fil central du modèle local du câble. La continuité du câble est représentée par des conditions de périodicité reliant les sections de bord à des points internes du modèle situés sur la même position circonférentielle. Les fils sont modélisés par des éléments poutres. On obtient les contraintes généralisées sur les fils, les forces de contact et les glissements relatifs. Des premières analyses ont montré que les déplacements relatifs entre les fils restent petits dans notre cadre d'application. Afin de réduire le coût calcul, nous avons développé un nouvel élément de contact entre poutres non parallèles, avec un appariement fixe de contact, dans l'hypothèse de petits glissements mais en grands déplacements et grandes rotations. Des tests numériques montrent l'amélioration obtenue, avec un résultat plus proche d'un modèle de référence qui considère un contact surfacique. De plus, le nouveau modèle réduit significativement le coût calcul et se montre plus robuste en convergence, ce qui s'avère crucial pour un calcul de fatigue. Les sorties du modèle local sont ensuite utilisées pour prédire un état de contrainte 3D, en exploitant des solutions analytiques de contact entre corps cylindriques. Finalement, un critère de fatigue multiaxial de la littérature est appliqué pour évaluer le risque en dommage. / The main function of mooring systems of floating offshore wind turbines is to ensure station keeping. The mooring lines can be composed of chains, wire ropes, synthetic ropes, or even a combination of them. In this thesis we focus on wire ropes, whose advantage over chain is to sustain high tension at a lower weight. Their design must consider the successive tension and bending loading induced by the floater movement for various wind and waves conditions.The thesis purpose is to develop a new numerical model, dedicated to the prediction of fatigue damage in mooring wire ropes of a floating wind turbine. In particular it has to simulate the relative movements between the wires when the rope is bent. Results from free-bending fatigue tests in the literature show the importance of these effects, since the first rupture is localized near the neutral plane, where fretting is more important. This phenomenon affecting the fatigue life is not considered by fatigue criteria of current offshore standards, which are related to tension-tension loading.It is worth noting that the use of a detailed model of wire rope in a fatigue design procedure represents a real challenge. The high number of contact interactions to be modeled, which are several thousands per meter of rope, and the large amount of loading cases make this type of computations extremely time-consuming.The loading used in the developed local model of wire rope is obtained from global computations performed with a dedicated multiphysics software (Deeplines). This software allows to simulate the environmental conditions (wind, waves, current) applied on the whole structural system.Some preliminary computations showed that the nonlinear bending behavior of the wire rope, linked to the wire contact interactions, does not significantly affect the output of the global model. This observation justifies the use of a top-down scheme, with a prior computation of the global scale.The global scale tension and curvature are then uniformly imposed on the central wire of the local model. The continuity of the rope is represented by periodic conditions which link the end sections to points within the model, at the same circumferential locations. The wires are modeled by beam elements. The outputs at the local scale are the stress resultants on the wires, and the contact forces and relative displacements at contact locations.Small sliding between the wires has been observed from first numerical analysis, for a representative loading case. Therefore, in order to reduce the computational cost of the wire rope model, a new node-to-node contact element has been developed, dedicated to the modeling of contact between non-parallel beams with circular cross section. It assumes fixed contact pairing and finite rotations. Numerical benchmarks and experimental tests on wire ropes show the improvement with results closer to a reference surface-to-surface model, when compared to standard algorithm for the simulation of contact between beams. Moreover, the new model reduces significantly the CPU cost and is also more robust, which is crucial for fatigue life estimates.The outputs of the local scale model are then used to obtain the complete 3D stress state by means of analytical solutions of contact between solids with cylindrical shape. Finally, a multiaxial fatigue criterion is applied in order to assess the safety of the system.

Offshore floating wind turbines represent the future of wind energy. However, significant challenges must be overcome before these systems can be widely used. Because of the dynamics of offshore floating wind turbines -- surge, sway, heave, roll, pitch, and yaw -- and the resulting interactions between the rotor and generated wake, the aerodynamic analysis methods and design codes that have found wide use throughout the wind energy industry may be inadequate. Application of these techniques to offshore floating wind turbine aerodynamics may result in off-optimal designs, effectively handicapping these next-generation systems, thereby minimizing their full potential. This dissertation will demonstrate that the aerodynamics of offshore floating wind turbines are sufficiently different from conventional offshore and onshore wind turbines, warranting the use of higher fidelity analysis approaches. It will outline the development and validation of a free vortex wake code, the Wake Induced Dynamics Simulator, or WInDS, which uses a more physically realistic Lagrangian approach to modeling complex rotor-wake interactions. Finally, results from WInDS simulations of various offshore floating wind turbines under different load conditions will be presented. The simulation results indicate that offshore floating wind turbine aerodynamics are more complex than conventional offshore or onshore wind turbines and require higher fidelity analysis approaches to model adequately. Additionally, platform pitching modes appear to drive the most aerodynamically-significant motions, followed by yawing modes. Momentum balance approaches are shown to be unable to accurately model these dynamic systems, and the associated dynamic inflow methods respond to velocity changes at the rotor incorrectly. Future offshore floating wind turbine designs should strive to either minimize platform motions or be complementarily optimized, via higher fidelity aerodynamic analysis techniques, to account for them. It is believed that this dissertation is the first in-depth study of offshore floating wind turbine aerodynamics and the applicability of various analysis methods.

In this paper, a new algorithm for utilizing energy storage is proposed and applied on floating wind turbine economics. The proposed algorithm’s decision making on storing energy or selling electricity onto the grid is based on the electricity price, which makes it unique and different from similar algorithms. From the literature review, it was concluded Ocean Renewable Energy Storage to be most suitable with the Spar-Type and Semi-Submersible floating wind turbine to which the paper is based upon. The objective of this paper is to find the suitable ratio of energy storage versus wind farm, find the product of increase in wholesale, and evaluate whether the proposed method makes the hybrid economically sound. The algorithm was applied on spot-price data from Denmark due to its large share of wind energy with wind data from off the coast of Morro Bay in California, USA. Additionally, a sensitivity analysis is applied to evaluate to energy storage cost impact as well as evaluate the algorithm by lowering the required energy storage size.   Using the algorithm, the wind farm must account for nine days’ worth of energy production with a product of energy storage versus wind farm ratio of 1.42. The wholesale price increased with 11.9-21.5% for the four years studied, however, all financial results favored not utilizing energy storage. By the results derived from the sensitivity analysis, it was concluded that with future cost reductions, the algorithm will still favor no energy storage. However, by fine tuning the algorithm to reduce the need for storage, positive financial result might be achievable. The key to achieve a profitable result seems to rely on minimizing the need for energy storage, to which the proposed algorithm fail to achieve. Conclusively, spot-price decision-based energy storing is not economically sound.