Cost-conscious control strategies for wind turbine systems

Wind energy is one of the most abundant renewable energy sources that can meet future energy demands. Despite its fast growth, wind energy is still a marginal player in electricity generation. The key issues preventing wider deployment of wind turbines include low energy conversion efficiency, high maintenance cost, wind intermittency and unpredictability etc. These issues lead to considerably higher cost of wind power compared to that of traditional power sources. This work is focused on control designs to overcome the above challenges. First, control algorithms are developed for energy capture maximization. During partial load operation, wind turbine rotor speed is continuously adjusted to remain optimal operation by manipulating the electromagnetic torque applied to the generator. In this dissertation, a dynamic programming based real-time controller (DPRC) and a gain modified optimal torque controller (GMOTC) are developed for faster convergence to optimal power operation under volatile wind speed and better robustness against modeling uncertainties. Secondly, fatigue loading mitigation techniques are developed to reduce the maintenance cost of a wind turbine. During partial load operation, a generator torque-based fatigue mitigation method is devised to reduce the impact of exacerbated tower bending moments associated with the resonance effect. During full load operation, a H₂ optimization has been carried out for gain scheduling of a Proportional-Integral blade pitch controller. It improves speed regulation and reduces drivetrain fatigue loading with less oscillations of turbine rotor speed and generator torque. Thirdly, battery energy storage systems (BESS) have been integrated with wind turbines to mitigate wind intermittence and make wind power dispatchable as traditional power sources. Equipped with a probabilistic wind speed forecasting model, a new power scheduling and real-time control approach has been proposed to improve the performance of the integrated system. Finally, control designs are oriented to wind turbine participation in grid primary frequency regulation. The fast active power injection/absorption capability of wind turbine enables it to rapidly change its power output for stablizing the grid frequency following an sudden power imbalance event. In addition to quick response to grid frequency deviation event, the proposed controller guarantees turbine stability with smooth control actions.