Predictive Control of Wind Turbine Using Preview Wind Speed Information

In this work, a preview wind information-based model predictive control (MPC) is applied to a linearized wind turbine obtained through high fidelity aeroelastic model (i.e., DNV-GL Bladed). The nonlinear dynamics of wind turbine is linearized into state space model using Bladed. The challenges in extracting energy from wind arise from unpredictable nature and turbulence in wind field. Hence, a preview wind speed based enhanced feedforward predictive control strategy for pitch and torque control is implemented using information obtained through LIDAR. Simulations are carried out in Matlab/SIMULINK to demonstrate the effectiveness of MPC based optimal controller in comparison to the reported baseline feedback (PID) and feedforward controller. Significant reduction in generator speed variations and pitch demand at high wind speed (above-rated) is observed. It also gives improvement in terms of frequency response (stability) of system; concurrently inherently eliminates oscillations without incorporating drive-train dampers. Thus, life span of wind turbine is improved.