The full-order state observer speed-sensorless vector control based on parameters identification for induction motor

During the operation of speed-sensorless control system for induction motor, the stator and rotor resistance varies greatly with the change of temperature and the frequency of the rotor side, which affects the estimation of the stator flux and leads to the low accuracy of the speed estimation. A speed-sensorless vector control method based on parameters identification with the full-order adaptive state observer is proposed in this paper. In the model reference adaptive system of AC motor, the stator resistance and rotor flux are assigned as state variables to build the reference model, and a full-order flux observer is introduced to adjustable model. Lyapunov theory and Popov superstability theory are used to deduce the speed and rotor resistance adaptive rate. The feedback gain matrix is simplified to speed up the convergence rate of the system. The estimation values of speed and rotor resistance are taken as the proportional integral form, so that an interactive model reference adaptive system is constructed by speed and rotor resistance identification. While observing the rotor flux, it can not only ensure the accuracy of the reference model but also eliminate the disadvantages of the voltage model with integral terms, and the rotor speed can be estimated at the same time. The experimental results show that the accurate performance of speed and flux identification can meet the requirements of application; the proposed control method with the identification of speed and rotor resistance has little fluctuations phenomenon on motor torque in low speed and achieves better performance.