Speed ​​sensorless Induction Motor Variable Frequency Speed Control And Position Servo Control Technology

The subjects of this paper are the realization of speed-sensorless fast point- to-point position orientation system and to analyses the properties of state estimator based on Extended Kalman Filter (EKF) for field-oriented controller(FOC) system of induction motor. The former put stress on the research of time optimal control method , fuzzy discursion and moving track, and the latter on the study of EKF recursion algorithms , noise model and noise characteristic identification. Some research about sensorless control of induction motors in recent years has been introduced in chapter one.In chapter two, a field-oriented control system with current regulator and voltage source inverter (VSI) is presented firstly, which is proved in theory that the voltage feed-forward decoupler can be omitted , the good decoupled control properties of flux current and torque current is also validated by experiments with speed-sensor based on F240DSP real system. The second, it is proposed that a FOC mentioned above can be modeled with a fast time optimal position control system based on the Pontryagin's minimum principle, which may become a speed-sensorless time optimal one by using speed estimator. A bang-bang controller may exists limit-loop or oscillation in vicinity of origin of the plane, so the stability of system destroyed. Here a low cost optimal position system with speed-loop open and position-loop closed based on fuzzy control is presented, it employs fuzzy controller to verdict rotor velocity approximately and a fuzzy switching curve is substituted for real one to achieve bang-bang switch, no limit-loop or oscillation occurs. At the end of this chapter, a DC break curve is discussed.Chapter three is about the experimental research of the real time optimal position system(See chapter two), emphasis on the high performances of 196MX PTS interrupt response and safe design of IPM module. The analysis of perfect experimental waveforms and basic algorithm are also provided.Chapter four focus on the properties and application of EKF estimator. The precision and convergence of both rotor speed and flux estimation with EKF largely depend on the accuracy of the models of system noise and measurement noise. A global optimized noise covariance estimation method based on Genetic Algorithms (GA) is applied to EKF in FOC system, A GA training simulation system of optimal parameters of EKF is given and the simulation results proves the effectiveness and robust of this algorithms.The end of this paper comes a conclusion of my study work during doctor's day, includes some suggestions for next work.