Research On Speed And Rotor Time Constant Identification Technique Of Induction Motor Based On The Sliding Mode Observer

As the development of the power electronic converter technology and control theory, induction motor speed control performance is also improved. Both general industrial applications and emerging applications such as electric cars, induction motor still maintains a leading position because of its advantages, such as simple and reliable structure, wide speed range, and low manufacturing cost. On the basis of the induction motor vector control, this paper studies induction motor speed and rotor time constant online identification technology for electric vehicle applications, in order to improve the performance of speed sensor-less control system and the accuracy of field orientation in a speed sensor system.While considering running performance, most electric cars currently are installed with speed sensor, speed sensor-less control is demand of the electric car operation in the speed sensor fault cases not only, but also is the research direction to further improve technology. So this paper discusses about speed and flux estimation for speed sensor-less control strategy. The present observer estimates motor speed and flux simultaneously with sliding-mode surfaces, which are chosen as a combination of current mismatches and estimated fluxes in the stationary reference frame. According to the mathematical model of induction motor and the design principle of state observer in the synchronous rotating frame, this paper demonstrates the problem of steady-state error in traditional single sliding-mode surface observer and the uniqueness of compound sliding-mode surfaces, by deducing design principle of the compound sliding-mode surfaces in the stationary reference frame; it is also important to perfect the design theory and method of observer with compound sliding-mode surfaces. The stability of the proposed observer in the stationary reference frame can be proved by Lyapunov Theorem, and results are shown to verify its effectiveness. Based on the analysis of the proposed observer, a sensor-less induction motor drive control system has been designed.The accuracy of the motor parameters, especially rotor time constant, is the key to restricting the performance of vector control, in order to make the vector control system be robust to the parameter variation.A fourth-order sliding-mode flux observer in the stationary reference frame is developed in this paper to online identify rotor time constant. The proposed observer estimates the rotor resistance and the inverse of the rotor time constant simultaneously with sliding-mode surfaces, which are chosen as a combination of current mismatches and estimated fluxes in the stationary reference frame. According to the Lyapunov Theorem, it can be proved that the observer is convergent. It also can be proved that the equivalent controls (low-frequency components) are the rotor resistance and the inverse of the rotor time constant. The prominent advantage of the observer is that it can identify the rotor resistance and rotor time constant simultaneously with strong robustness to leakage inductance,and the identification result is not affected by mutual inductance. Its dependence on the stator resistance parameters can be solved through online compensation based on the built-in temperature sensor.Finally, an experimental platform of induction motor is set up to verify the key technologies, such as vector control of the induction motor, sliding-mode speed observer for sensor-less control.