The Second-Order Iterative Learning Control Of Permanent Magnet Synchronous Motor Based On Sliding Mode Observer

Permanent magnet synchronous motor(PMSM) with its high power density, high efficiency, high torque and simple structure, has been widely used in the tendency of electric drive system. However it's output torque ripple limits its application. Iterative learning control(ILC) is suit for the controlled system with repetitive motion, that can achieve optimal tracking in a limited time without accurate mathematical model of control object and has good effect in restraining the periodic torque ripple. In obtain the accurate information of rotor position and speed of permanent magnet synchronous motor, the wildly method is use high accuracy position sensor at present, but the method is restricted by the cost and application condition. Therefore, a sensorless control method with high reliability, high tracking accuracy and easy to maintain is urgent need. At present, the widely used sensorless control methods such as direct calculation method, it has the advantages of simple calculation and fast response, but it also has poor anti-interference ability and low reliability due to the open loop control; Sliding mode observer method,it has the advantages of strong robustness, well immunity, but the chattering problem limits its application.On the basis of the summary of PMSM torque pulsation and sensorless control, the analysis and design of the motor control system are studied in this paper. In order to master the character of the PMSM more, new ideas is analyzed and proposed in this paper. The research work and innovation of this paper are as follows(1) Initially, this paper analyzes the physical structure of PMSM and obtains the mathematical model of PMSM through the coordinate transformation. Then determines the observability of PMSM by analyzing the observability of nonlinear system.(2) A kind of second-order PD-type(proportional-derivative-type) iterative learning control law is proposed for the problem that periodicity pulsations exist in PMSM. Iterative learning controller is designed in the case of nonlinearity and disturbance, sufficient condition for the stability of the closed loop control system is obtained by lyapunov stability theory in the sense of Lebesgue-p norm.(3) Sensorless control based on two-order sliding mode observer is proposed for the problem of position and speed tracking. Firstly, the proportional integral sliding mode surface is designed, and then the method of combining equivalent control and switching control is adopted to ensure the speed convergence and the error reduction. Which can estimates the rotor speed and position information quickly and accurately.(4) The iterative learning control method is used to track the torque, current and speed of PMSM, two-order sliding mode observer is designed to observe the rotor position and speed based on the Lyapunov stability theory. Then, feedback state observer value to the iterative learning controller, to achieve the precise tracking of the PMSM system. Finally, the proposed method is verified by MATLAB/simulink simulation system and experiment.