Søren Christiansen defended his PhD successfully on the 6th of March 2013 at Aalborg University at the department of Electronic Systems. The title of his thesis is "Model–Based Control of a Ballast–Stabilized Floating Wind Turbine Exposed to Wind and Waves" and is a contribution to Work Package 3 in NORCOWE.
Below is the abstract of his thesis:
The wind turbine is a commercial product which is competing against other sources of energy, such as coal and gas. This competition drives a constant development to reduce costs and improve efficiency in order to reduce the total cost of the energy. The latest offshore development is the floating wind turbine, for water depths beyond 50 meters where winds are stronger and less turbulent.
A floating wind turbine is subject to not only aerodynamics and wind induced loads, but also to hydrodynamics and wave induced loads. In contrast to a bottom fixed wind turbine, the floating structure, the hydrodynamics and the loads change the dynamic behavior of a floating wind turbine. Consequently, conventional wind turbine control cause instabilities on floating wind turbines.
This work addresses the control of a floating spar buoy wind turbine, and focuses on the impact of the additional platform dynamics. A time varying control model is presented based on the wind speed and wave frequency. Estimates of the wind speed and wave frequency are used as scheduling variables in a gain scheduled linear quadratic controller to improve the electrical power production while reducing fatigue. To address the problem of negative damped foreaft tower motion, additional control loops are suggested which stabilize the response of the onshore controller and reduce the impact of the wave induced loads. This research is then extended to model predictive control, to further address wave disturbances.
In the context of control engineering, the dynamics and disturbances of a floating wind turbine have been identified and modeled. The objectives of maximizing the production of electrical power and minimizing fatigue have been reached by using advanced methods of estimation and control.
The thesis was supervised by Professor Thomas Bak and Associate Professor Torben Knudsen, both at Aalborg University.