A LiDAR-based pitch control strategy for ultra large wind turbines

Abstract

In this paper, a new adaptive predictive pitch control strategy is proposed. The proposed technique utilizes the future wind speed measurements provided by a light detection and ranging (LiDAR) apparatus. This apparatus supplies a feedforward controller with the incoming wind speed readings, which then generates an early control action that reduces generator speed variation and prevents suddenly generated power drop. The proposed feedforward controller is adaptive and got tuned online as well using the LiDAR wind measurements data. Moreover, a Robust Model Predictive feedback Controller (RMPC) is designed to augment the pitch control action with a component that reflects the fluctuation of the generator speed. The proposed control strategy is compared to that of a conventional gain-scheduled PI controller and tested on a 5-MW realistic wind turbine model against extreme turbulent wind profiles. Simulation results have proved the superiority of the proposed controller in improving both the power quality and the generator speed regulation as well as increasing the harvested power.