Assessment of the accuracy of engineering methods for predicting dynamic and quasi-static wave loads on monopile-supported offshore wind turbines

Abstract

Meaningfully confronting the climate crisis requires, among many other things, a transition from an economy driven by the burning of fossil fuels to one driven by renewable energy. Development of offshore wind energy is an important aspect of the effort to decarbonize the energy sector because wind farms with large generating capacities can be constructed in the abundant offshore areas near densely populated coastal regions with high energy demand. In the United States, the offshore wind industry lags behind that of countries in Europe and Asia, but growth is on the horizon: the Atlantic Coast is projected to host thousands of offshore wind turbines (OWTs) with a cumulative capacity in the tens of gigawatts in the coming decades. It is likely that many or even most of the OWTs along the U.S. Atlantic Coast will be supported by monopiles. Monopiles support nearly three quarters of global offshore wind capacity and the U.S. Atlantic Coast has nearshore depths that are shallow enough for monopiles to be a viable foundation type. The scale of these offshore infrastructure installations will be dictated by trends in OWT capacity. Currently, the largest installed OWTs have capacities of 8 MW and monopile diameters of 7.5m, but with next-generation OWTs projected to have capacities in the 12-15 MW range, monopiles with diameters of 10m and greater are considered feasible and even necessary for future development. Structural engineers have an important role to play in this development. Assessing the risk to the offshore renewable energy infrastructure planned for the U.S. Atlantic Coast will rely in large part on the ability of structural engineers to accurately predict wave loads on large diameter monopiles. Structural engineers face a modeling problem and a computational problem: many uncertainties in the wave loading of monopile-supported OWTs persist because the offshore environment is so complex, but high-fidelity simulations that accurately model that complexity are computationally expensive and difficult to implement with confidence in a practical setting. As a result, comprehensive analysis and risk assessment for the total number of OWTs along the Atlantic Coast is challenging. Engineering methods of wave load prediction based on simple wave mechanics and hydrodynamics models offer a means to circumvent the computational limits imposed by high-fidelity simulations. If used judiciously, such engineering methods can expedite comprehensive analysis of many scenarios of practical interest. The goal of this dissertation research is to assess the conditions for which these methods for the prediction of wave loads on large diameter monopiles can preserve a minimum level of accuracy that is necessary to make useful design recommendations and risk predictions. This assessment is made by comparing predictions of wave loads yielded by the engineering methods with simulated wave loads and with measurements of wave loads from wave flume experiments. Such comparisons can also demonstrate which site, environmental, and structural characteristics are most important for the accurate prediction of wave loads. In this way, the work provides further insight into the physics governing the response of monopile-supported OWTs to wave loading. --Author's abstract