Research Of Sensorless Vector Control System And Its Speed Estimation

Induction motor is a non-linear, multivariable, high-order and close coupling system. It's torque isn't controlled as the DC motor. The problem of tranditional induction motor drive can be solved by vector control, while the sensor is eliminated in sensorless vector control system and this makes the relative inductor motor drive easier, more economical and more reliable. For this reason,the sensorless vector control is becoming the attractive researching project.Firstly, the basic theories of vector control is introduced.The dynamic mathematical model of inductor motor under static reference frame is given. Then through coordinate transforming, the mathematical model of inductor motor based on the synchronous reference frame is deduced. The rotor flux linkage control and torque control based on the rotor field orientation control are separately achieved.Secondly, based on the synchronous reference frame, the voltage equations in terms of flux linkage of induction motor is deduced , a vector-controlled approach based on the rotor field orientation control is proposed as well. Error signal of the stator T-axis current in the synchronous reference frame is utilized to attain the induction motor speed. This speed estimation method is simple and has self-adaptability. Because classical PI controllers is used, the structure of the sensorless vector control is very simple.Thirdly, based on MATLAB , a simulation model of the proposed vector control sysem is established. In this model, because of adopted PU of controlled variables, the system has high self-adaptability and parameters of PI controller are easy to be adjusted; The simulation model is a discrete control system and in this system sample period of different signal is different; To improve the system dynamic and static performance ,a compensation method of the phase shift caused by discrete system is presented .Finally, dynamic and static characteristic of induction motor are simulated. Simulation results demonstrate that the proposed vector control system is feasible and valid, and the presented speed estimation model has good speed estimation precision. Moreover, it is robust to the variation of motor parameters.