Study On Rotate Speed Estimation Of Speed Sensorless Induction Motor Drive

With the qualitative leap to AC speed regulation system, made by rotor field directional control, the appearance of speed sensorless induction motor drive did greater effect on enhancing the simplification and robustness of system. There are two problems to be solved in such system. One is the rotate speed estimation, and the other is flux linkage observation of rotor. In this paper, three methods would be investigated separately on the speed estimation and flux linkage observation of the speed sensorless induction motor drive.The first one is the confirmation of self adaptive arithmetic. It is the key to design the MRAS (Model Reference of Adaptive System). Superstability theory and positive dynamic system were used to design adaptive rule in this paper. With the vector control system, model reference adaptive arithmetic was designed, and simulated as the rotate speed estimator in speed sensorless induction motor drive. Result indicated that this system had preferable dynamic and steady state capability.Besides, it is well known that the mathematics model of asynchronism motor is a set of nonlinear equation. According to the system identification theory, the efficiency way of state estimate to nonlinear equation is extended Kalman filter arithmetic. Based on such arithmetic, research was proceeded on nonlinear state equation of motor. It was found that even stator current and voltage could be used to estimate motor state in extended Kalman filter arithmetic. Simulation result proved that the rotor speed and flux linkage signal were much satisfied.In recent years, wavelet network is one of the popular nonlinear approximating function methods, which has strong study capability. Because it can approach nonlinear function with random precision after training, the third research of rotate speed estimation was carried through with wavelet network. Therefore, it improved the conventional...