Large-scale Interactions between Wind Farms and the Atmosphere

The offshore wind energy sector is expected to see rapid growth in the coming years, to combat climate change. The performance of offshore wind farms is significantly influenced by turbine-generated wakes, which can pollute the inflow to downstream turbines and lower their power production. Therefore, it is crucial to understand how turbine and wind farm wakes behave and interact with eachother and the atmosphere. In this dissertation, we use large-eddy simulations (LES) to investigate how wind farm design and atmospheric conditions govern these interactions.

In Part 1 of the dissertation, we evaluate how wind farms are affected by atmospheric conditions found in coastal areas. Chapters 2 and 3 explore the effects of atmospheric baroclinicity: height-dependent pressure gradients, induced by horizontal temperature gradients. We find that baroclinicity affects wind shear and veer, low-level jets and turbulence levels in the atmosphere, impacting both the overall power production of wind farms and the relative performance of different turbine rows. Furthermore, we propose a new engineering model to predict wind farm wake recovery based on our findings. Chapter 4 investigates the effects of a terrain transition from land to sea on wind farms. We find that an abrupt change in surface properties induces a highly heterogeneous internal boundary layer, which boosts overall wind farm power production, but reduces the relative performance of downstream turbines.

The research in Part 2 investigates how a wind farm's design affects its interactions with the atmosphere. In Chapter 5, we investigate the transition from turbine-scale to wind farm-scale wake recovery. We find that wakes behind isolated turbines and wind farms recover through different physical mechanisms, and identify key factors that govern the transition. Chapter 6 evaluates the efficacy of rotor-tilt control in wind farms. We show that collective tilt strategies are effective in aligned wind farms, but can lead to power losses in staggered configurations. Furthermore, we discover that enhancement of large-scale entrainment is not easily achieved using tilt.

The findings presented in this dissertation advance our knowledge on wind farm-atmosphere interactions, and can be used to improve the tools used by industry to design and site wind farms.