Wind speed estimation for wind turbine control

In the literature, proposed control strategies for the efficiency of wind energy conversion systems are usually based on gain scheduling PI controller using the pitch angle to control the rotor speed. Focus is then only made on power production. In previous work, we proposed a model based control strategy that highly decreases the mechanical solicitation of the structure while preserving power production. This control strategy is based on aerodynamical models that uses the measure of the wind speed as an input. For many applications such as wind resource assessment and power performance testing, it is the mean speed (usually averaged over 10 minutes) at a particular height above ground that is of interest. The most appropriate type of instrument for such applications is undoubtedly anemometers. However, for control purpose, instantaneous wind speed is needed. Such devices already exist as LIDAR (Light Detection And Ranging) for example. However, this measurement is not accurate for all turbulence wind conditions. Thus, wind speed estimation is needed to robustify the proposed control strategy. The proposed estimator is an observer estimating the wind speed and the aeroelastic torque for various wind conditions. The aim of this paper is to see the impact of using this estimator in the control strategy on performance and on the lifetime of the structure. The observer based control strategy has been validated using FAST, for several turbulence wind conditions showing the impact of the wind speed estimator on performance and the lifetime of the system for onshore and offshore wind turbine.