The operation of Floating Offshore Wind Turbines (FOWT) poses several challenges due to the added dynamics of the floating platform. Amongst them, the appearance of the platform pitch instability at above rated wind speeds limits the traditional PI controllers’ bandwidth. Therefore, several solutions have been proposed, including controller detuning or parallel compensation. This work explores the application of a linear Model Predictive Controller (MPC) at high wind speeds to limit the platform pitch motions and the loads. The MPC is compared to a gain-scheduled Proportional – Integral (PI) controller with parallel compensation. The linear MPC uses a fairly simple model with 3 independent Degrees of Freedom (DOF), namely platform pitch and surge, and rotor speed which allows for a thorough study on the tuning procedure. On the other hand, using a simple model limits the performance improvements close to rated wind speed due to inherent aerodynamic non-linearities. The controller objectives are explicitly stated as tracking the rotor speed and minimizing the platform pitch velocity, while a penalty to the demanded blade pitch rate is imposed. Turbulent time-domain simulations are used to generate Pareto fronts where objective function weights are swept to find optimal tradeoffs. The mean-wind speed, and turbulence intensity are found to impact the tuning procedure, and thus a set of weights is provided per wind speed. These optimal weights per wind speed are used to perform a load analysis. Great improvements in electrical power output, platform motion minimization, and shaft torsion and tower base load alleviations are achieved.
No takes yet. Share an insight, caveat, or question.
C. Moreno (2024) studied this question.