Robust Model Predictive Control for Collective Pitch in Wind Energy Conversion Systems

Abstract When wind speeds exceed the rated values, wind turbines operate in region 3. In region 3, collective pitch control (CPC) is the main tool to regulate the turbine’s speed and generated power. The main challenges that face a CPC design are the modeling uncertainties, constraints on the control actions, and immeasurable system states. A tube-based model-predictive output-feedback controller is proposed here to design a CPC. The proposed controller is an optimal controller that respects constraints and accommodates uncertainties without a need to measure all states. Applications to a typical 5-MW offshore wind turbine show through simulations the superiority of the proposed controller.