Full Envelope Wind Turbine Control Systems Design

点击翻译按钮获取中文摘要

Wind turbines are a means of providing clean, safe and abundant energy. There is limitedenvironmental impact and the operating costs are low. On the other hand, initial capital costs forwind turbines are high. As such, long life and predictable operating patterns are desirable. Thisproposal aims to improve the life and predictability of operation of wind turbines through fullenvelope control. Wind turbines operate in two regimes: moderate wind speeds where the goal is toextract maximum power and higher wind speeds where power is regulated near capacity. In thetransition between moderate and higher wind speeds the system should efficiently and safely movebetween regimes. Most large wind turbines are of a variable speed, variable pitch configurationwhere two actuations are used: field excitation of the generator stator, governing rotation as abraking force; and blade pitch, which affects the aerodynamic lift and drag and generated forces.For lower operating costs, pitching of the blades is avoided as much as possible as this involvesmechanical components requiring lubrication and bearings with finite lifespans. Thus, at moderatewind speeds only field excitation is used, while in power regulation mode, both blade pitching andfield excitation are employed. Transitioning from one regime to the other involves a change in thenumber of actuators, along with changes in control system goals and in dynamic characteristics. Inthe transition region, overspeeding of the turbine can result in exceeding rated conditions forbearings and torques and forces in the blades and tower structure. All of these issues can greatlyreduce the life expectancy and increase maintenance costs.Promising methods for feedback control of wind turbines include linear parameter varying concepts,optimized disturbance rejection methods and model predictive control. Further, switched systemswill allow more flexibility in dealing with varying objectives, number of actuators, and varyingdynamics. Work here will consider the extrapolation of these methods to handle transition regioncontrol and full envelope operation of the turbine. Transition between regimes can induceoscillations in speed and torques and also increase wear on the pitch actuators. Intelligent choices ofcontrol strategy to avoid nonlinear oscillations would be of benefit to practical implementation.Finding peak power is fraught with difficulty. Varying dynamics due to blade wear and damage,uncertain power curve data, and complex interaction between turbines and with the environmentcan cause errors in estimating parameters for maximum power extraction. Employment of advancedfeedback control techniques as described in this proposal can improve the efficiency, maximumpower extraction and improve life of the turbine.