Research-based Adaptive Speed Observer For Doubly Fed Wind Power Generation Control System

With the increasing scale of wind power generation system, the investigation on wind power generation technique has been focused on improving the operation efficiency of wind power generation. For this reason, the technique of variable speed constant frequency(VSCF) operation which is capable of capturing the maxim wind power, becomes research hotspot whose representative is the double-fed wind generation System. However, the speed sensor does harms to the system's stability and maintainace. So it is significative to study the variable speed constant frequency wind generator system without speed snesor .This dissertation analyzes double-fed induction generator(DFIG) operation theory, including mathematic model of DFIG, coordinate transformation, operation characteristics and power relationship of wind turbine, which establishes the theoretical foundation for succeeding study. A grid controller and a converter controller are designed, which separately achieve the unit power factor rectifier and the maxim power capture. The maxim power capture control model is emphatically researched, in which a general control strategy based on stator flux oriented vector control is proposed. A 15KW simulation double-fed generation system is built in Matlab/Simulink, which can obtain the cutting-in and power capture functions.An speed estimation scheme based on adaptive stator observer is proposed. In the scheme, a new Lyapunov function and a speed estimation law is induced. Based on iterative LMI method, observer gain matrix is conducted by solving two bilinear matrix inequalities, which ensures the global stability of the system. At last, builds a simulation model of the sensorless double-fed generation system in Matlab/Simulink. By the comparison, the simulation results will prove the feasibility of this estimation method.