Research On Speed ​​Sensorless Control Of Permanent Magnet Synchronous Motor

In order to achieve close-loop control, the speed sensor needs to be installed to return the rotate speed signal in the high-performance servo system of Permanent Magnet Synchronous Motor (PMSM). However, with the increasing requirement of the control accuracy, the issues caused by the application of speed sensor, including rising cost, difficulty of installation, increasement of system size and incapability of adapting to the severe environment, bring significant limitations to the application of PMSM. To solve the problems, the study about PMSM sensorless control is one of the most important research directions in the AC servo field. Therefore, in this thesis, the theory research about the sensorless control technology of PMSM is undertaken, and the outcomes are examined and analyzed through simulations and experiments. It generally implements the sensorless control for PMSM.Firstly, the mathematical model of PMSM and SVPWM algorithm are studied; the mathematical model of the three-phase inverter is established and connected with that of PMSM; the controller based on the current feedforward decoupling control strategy and Lyapunov method are analyzed and designed, then, the differences between the two methods are compared in the PMSM vector control system.Secondly, sliding mode variable structure control is studied and applied in the sensorless control of PMSM. The methods and principles of the sensorless control for the conventional Sliding Mode Observer (SMO) are analyzed and deduced with details. Sigmoid function instead of sign function is adopted in SMO, which simplifies the structure of SMO, weakens chattering and improves the stability and reliability of the system.Thirdly, regarding the problem that the traditional SMO has a relatively larger observation error, an adaptive SMO based on the static coordinate system is introduced, which eliminates the error and improves the steady accuracy effectively. To solve the problem that is unable to realize the reverse control, it also proposes a digital phase-locked loop (PLL) with ellipse fitting method considering frequency change and realizes the reverse control.Fourthly, regarding the slow dynamic response in system and massive calculation due to PLL, a new sensorless control method for the adaptive SMO based on rotating coordinate system is proposed. The influence from the dead-time effect on the sensorless control of PMSM algorithm triggered by dead-time of inverter is analyzed. The dead-time compensation is introduced into the proposed algorithm, which solves the problem that there is a phase difference between the estimated and the actual rotor position angle under the big dead-time situation.Fifthly, the experiment platform for the sensorless control of PMSM is established. The proposed theoretical algorithms, which implement the sensorless control of PMSM, are examined through experiments.