Prospects of multivariable feedforward control of wind turbines using lidar

Abstract

Current advances in lidar-technology provide the possibility of including wind preview information in the control design. Lidar-assisted collective pitch control is a simple, but promising approach to reduce the rotor speed variation and structural loads for full load operation. This work extends this approach to the transition between partial and full load operations. A multivariable controller is presented, which provides a simple update for the generator torque rate and the minimum pitch angle based on a nonlinear system inversion. The feedforward signals of the generator torque rate and the minimum pitch angle can be combined with conventional feedback controllers and the collective pitch feedforward controller for full load operation. This facilitates the modular application on commercial wind turbines. Simulations with a full aero-elastic wind turbine model and a lidar simulator show improved rotor speed regulation and significant reduction of tower loads, while only slightly decreasing power. Further, possibilities to transform the load reduction into energy increase are outlined.