Stochastic Model Predict Control Of Megawatt Wind Turbines Control Systems

Under the wind load of random, intermittent, megawatt wind turbine showed the typical characteristics of stochastic hybrid system. Due to the randomness and uncontrollability of the wind speed, wind turbines in different wind speed area have different control requirements for power and speed. Generally, when the wind speed is less than the cut-in wind speed, wind turbine does not start; when the wind speed is greater than the cut-in wind speed and less than the rated wind speed, wind turbine need to achieve optimal wind energy capture efficiency; when the wind speed is exceeds the rated wind speed, wind turbine is required to limit the rise of power and keep the system stable. Under the action of control system, the whole wind turbine switches between several discrete states, such as partial load, full load and system stop states. The mutual conversions between the part load, full load and no load of the wind turbine satisfy the markov chain model. Thus, conducted to megawatt wind turbine stochastic hybrid system modeling and stochastic predictive control system research is of paramount importance.In this thesis, the first part discussed the discrete stochastic hybrid automaton modeling of wind turbine. According to the mechanism model and the real time data of wind turbine, by using a combination of mechanism modeling and data driven approach, a switched affine system model for the evolution of the continuous state of wind turbine is established. Then calculate the markov chain state transition probability matrix P of the wind turbine discrete state (including the shut down, optimal capture, constant power running), so as to build a DHSA model of wind turbine under normal mode. The second part introduced a control algorithm that can realize the closed loop stochastic stability, and use it to control the discrete stochastic hybrid automaton model of the wind turbine. Simulation results show that the control algorithm can doing well. The third part introduced a stochastic predictive control algorithm based on scenario enumeration method and its toolbox, then a simple simulation example is given out,and the application prospect of this algorithm in the megawatt wind turbine model are discussed.