Study On Active Power Control Of Wind Farms With DFIGs Based On Model Predictive Theory

In recent years, with the development of wind power generation technology, the installed wind-power capacity has grown rapidly. Wind power has become one of the most important usage of clean energy. But, because of the complexity of wind power generation system as well as the intermittent and volatility of wind power, it has brought great difficulties to the active power control of wind farms. Meanwhile, large-scale wind power connecting to the grid has also brought a great threaten to the safety and stability of the grid.In order to reduce the mechanical load to extend its life time of wind turbines, keep the operation stability of the unit, improve the active power output characteristic of the wind farm and be more adaptable to the grid, this paper studies the control methods for both a single unit and the whole wind farm focusing on wind farms with doubly fed wind turbines. The main achievements are listed as follows.1) With a deep analysis of the structure and operation principle of doubly fed wind turbines, the mathematical models of the wind blades, gear boxes, doubly fed generators and variable pitch mechanisms are built. Because of the strongly nonlinear characteristics of blade, it was linearized according to wind speed. Reorganizing them, we can finally get the continuous-time state-space model of the unit which directly reflects the relation between the unit output power, generator speed and wind speed, pitch angel, generator torque. Furthermore, the running states of the unit are also included in the model.2) Model predictive control theory is introduced into the active power control of doubly fed wind turbines. And a multiple model predictive control(MMPC) strategy based on state-space model is proposed. The core of it is a multiple model predictive controller. With the change of wind speed, the weighs of the controller are adjusted to make the most use of wind energy. The control method can also be used when the output power of the wind farm is limited.3) An active power control strategy for wind farms based on units real-time adjustable capacity is proposed. All units are classified according to the real tine output active power and the predictive active power. Considering the adjustable capacity and response time of each kind of unit, the active power increase and decrease control method are proposed which can not only improve the control accuracy but also reduce the start and stop times of the units.