Research On Sensorless Control Of Direct-driven Permanent Magnet Synchronous Generators For Wind Energy Conversion

This thesis mainly discusses direct-driven permanent magnet synchronous wind energy conversion system(WECS) used back-to-back PWM converter. Mathematical models of system components are introduced. Maximum power point tracking(MPPT) is realized by effective electrical torque control strategy. Closed-loop current control system using rotor flux oriented control strategy is designed. Based on these steps, simulation model of vector control system for WECS with position and speed sensor is built up in Matlab/Simulink module. Simulation results show that MPPT control could be realized. System has good steady and dynamic performance.The general principle of space vector pulse width modulation(SVPWM) technique is analysized inαβγcoordinates. According to it, the impact of different zero-vector time distribution on output common-mode voltage, DC bus voltage utilization, power quality and switching loss in voltage source inverter is researched in this paper; and an effective common-mode voltage suppression method called zero sequence component cyclically balanced SVPWM methed is proposed.To reduce cost and overcome mechanical sensor's assembling problem in WECS site, sensorless vector control technique based on Simplified Kalman Filter(SKF) is researched in this paper. The rotor position and speed are chosen as state variables, two-phase stator current is chosen as input. Discrete state estimation model of permanent magnet synchronous generator is built up to estimate rotor position and speed. Simulation results indicate that rotor position and speed could be precisely estimated by using this method. Dynamic and steady state performances are good. MPPT control is also achieved based on the method. This paper also studies phase-locked loop estimation method based on extended electromotive force. Simulation comparison displays SKF method has good steady and dynamic properties. Firstly, this thesis accomplishes design of drive circuit for back-to-back PWM converter. Secondly, DSP programs of three SVPWM methods are compiled and experimental verification of the zero voltage vectors'impact on SVPWM is executed. Then, the system operating experiment at steady and step change wind speeds are carried out. Experimental results demonstrate the effectiveness of the proposed MPPT control scheme. At last, DSP program of SKF estimation method is compiled, sensorless vector control system operating experiment is carried out. Experimental results show the proposed SKF estimation method can accurately estimate rotor position and speed information.