Intercomparison of dynamic wake modeling frameworks for offshore wind farms

Wind energy plays a crucial role in the transition towards sustainable power generation. However, the intricate flow dynamics within densely clustered offshore wind farms pose significant challenges leading to efficiency losses due to wake interactions, larger fatigue loading due to enhanced turbulence within the wind plant and aeroelastic phenomena. The estimation of the performance of a wind plant can be done using different wake models. Analytical models are often used to predict the wake effects and to assess control-oriented strategies with a lower computation time compared to higher fidelity wake models. However, few of these engineering tools are able to perform unsteady wake simulations. There also exist medium-fidelity solvers which aim to balance the need for accurate dynamic wake modelling while keeping low computational cost. This thesis compares various wake simulation models of different fidelities, ranging from analytical steady to multi-body unsteady, to investigate how these models relate to each other and underlying physical phenomena. The study focuses on the low-fidelity wake modelling frameworks UFLORIS and FLORIDyn, which are an extension of the steady wake framework FLORIS. The comparison also involves the medium-fidelity FAST.Farm framework which is a multi-physics engineering tool for modeling power performance and structural loads. First, the intercomparison is carried out on a simplified wind turbine layout which consists of three turbines in a row with yaw control for the upstream turbine to assess the routine, accuracy, and flexibility of each model. Secondly, an analysis in detail of the unsteady VKI solver, UFLORIS, is carried out and some improvements are proposed. The effects of these changes are illustrated on a simple simulation, yet retaining all physics of interest. Finally, this solver is used to simulate a low-pressure system in the North Sea considering the full Belgian-Dutch offshore cluster. The results compare different models within UFLORIS and the real operative data of the turbines of the Norther offshore wind farm. They exhibit a good agreement between the models which, however, show notable discrepancies from the real data, only capturing the general trend in the strongly dynamic condition (wind direction and amplitude change); while converging in steady conditions. This project contributes to advancing the understanding and prediction of wake effects in offshore wind farms when dynamic features are important. Such insights are useful for optimizing wind farm layouts, enhancing energy production, and guiding the development of efficient control strategies, ultimately driving the sustainable growth of offshore wind energy.