Adaptive Tracking Control of Partially Known Signals with Application to Large Wind Turbines In Region II Operation

Abstract

In some applications it is necessary for the active controller to track a signal that is only partially known or measured. This arises particularly in active control of variable speed horizontal axis wind turbines for electric power generation, and is called Disturbance Tracking Control (DTC), where a plant is persistently disturbed, and the controller must cause the plant output to track a linear function of this disturbance. This control theory is related to Tip Speed Ratio Tracking for wind turbines operating in Region II. In this paper the theory of adaptively tracking a partially known signal is developed with the use of a simple estimator to approximate the unmeasured part of the tracking signal. This estimator uses no plant parameter information. The adaptive tracking control is shown to converge under fairly simple hypotheses, and is applied to form an adaptive version of the DTC Theory for wind turbines with poorly known parameters and no direct wind speed measurement. We apply the theory to create an adaptive tip speed ratio tracking controller for a horizontal axis wind turbine generator based upon a model of the NREL Controls Advanced Research Turbine (CART).