Assessment of offshore wind turbines in extreme weather conditions: Wave nonlinearity effect, cyclic soil response and breaking wave forcing

As one of the most promising source of sustainable energy contributing to reducing CO2 emissions, offshore wind is gaining more popularity around the world. Despite its fast growing, the need for lowering the cost of offshore wind energy has been a great priority over the past years. This thesis is aimed for this goal by providing a more accurate and de-risked load assessment for the offshore wind turbines (OWTs) exposed to extreme weather conditions. More specifically, three main contributions were made by investigating the wave nonlinearity effect, the cyclic soil response and the breaking wave forcing. A more physically realistic fully nonlinear wave model, as an alternative to the widely used linear and constrained wave, was used to reduce the model uncertainties inwaveloadscalculation. Asexpected,largerandsteeperwavesweresimulatedusing the nonlinear wave description, further resulting in larger wave loads and a stronger response on the OWTs. This effect is more pronounced for the parked situation, through the so-called springing- and ringing-type response. The well-known ringingtype response may be triggered by an extreme wave in the nonlinear wave fields. The increment of the ultimate characteristic loads caused by the nonlinear waves is within the current design safety factor 1.35. As a result of the pronounced spring-type and ringing-type responses, assessment of the cyclic soil response was performed in a 39 hours storm. Generally, the soil deformation subject to cyclic loading is mainly determined by the soil types, the static ultimate capacity and the cyclic load magnitude. Given a pile, the largest load cycles, although normally very few in number, generate higher cumulative soil displacement thanthethousandsofsmallloadcycles. Additionally,thenumberofcycleshaslimited effect on the accumulated soil deformation for a given loading condition, due to the soil densification phenomenon exposed to the cyclic loading. Finally, the characteristics of breaking wave slamming force were addressed using the measurements data from the large-scale experiments on a monopile and a jacket. For the monopile, the parameters used in the existing slamming load model were calibrated, and a significantly larger impact duration was found. In terms of a jacket, the slamming force time history was parameterized, and a description of the global breaking wave force was developed. In summary, the outcomes of this thesis is foreseen to provide a more accurate load assessment tools to reduce the risks and costs related to substructures, for a safe, yet economic design.