Hierarchical control of floating offshore wind farms with repositionable turbines

Floating offshore wind farms (FOWFs) are expected to play a central role in expanding offshore wind energy into deep-water regions, where fixed-bottom wind farms are no longer viable. Their deployment, however, is held back by a levelized cost of energy substantially higher than that of onshore wind or fixed-bottom offshore wind. One pathway to reducing this cost is to mitigate wake effects, whereby upstream turbines generate regions of reduced wind speed and elevated turbulence that lower the power production of downstream turbines and increase their fatigue-relevant loads. Wind farm control offers a promising approach to mitigating wake effects and reducing the cost. For FOWFs, wind farm control can exploit an opportunity unavailable to fixed-bottom wind farms. Because each turbine floats and is anchored to the seabed by mooring lines, designing these mooring lines with sufficient slack allows turbine positions to be controlled within a wide horizontal range through aerodynamic forces exerted on the rotor. By reducing the overlap between upstream wakes and downstream rotors, turbine repositioning provides an FOWF-specific wake-effect mitigation strategy. This thesis develops a hierarchical control framework for turbine repositioning, in which a farm-level controller determines where turbines should move, while turbine-level controllers determine how individual turbines track those reference positions. At the farm level, an optimization-based controller coordinates turbine repositioning with three wake-mitigation strategies, namely wake steering, power derating, and Helix wake mixing, either to maximize farm power or to track a prescribed power setpoint. At the turbine level, two repositioning controllers are developed to realize the farm-level reference positions. The first is a supervisory model predictive controller that drives each turbine toward its reference position while jointly managing power operation, platform-oscillation suppression, actuator usage, and repositioning responsiveness. The second is a gain-scheduled linear quadratic regulator that coordinates nacelle-yaw and individual-blade-pitch actuation to achieve turbine repositioning while reducing reliance on the maintenance-intensive nacelle-yaw actuator. Together, the simulation results demonstrate that turbine repositioning is a promising wind farm control method, providing a new pathway for improving the performance, operational flexibility, and cost-competitiveness of FOWFs.