Model Predictive Control Application And Research In Permanent Magnet Synchronous Motor

With the continuous improvement of the level of the scientific productivity,human have gradually realized the mechanization and automation of the production process,and greatly improving production efficiency.Permanent Magnet Synchronous Motor(PMSM)has been widely used in various driving fields such as industrial production because of its high conversion efficiency,high power density,and high reliability.With the popularization and application of the permanent magnet synchronous motors,the shortcomings of their own nonlinear,multivariable and strong coupling are also becoming more prominent.Selecting the optimal control system to satisfy the complex industrial requirements has become an urgent problem in the field of permanent magnet synchronous motor control.Based on the vector control theory,this paper deeply studies the high-performance rotational speed,the current control strategies of the permanent magnet synchronous motors and the key techniques of suppressing torque ripple.Firstly,this paper status the development of the permanent magnet synchronous motor at home and abroad,status the existing problems of various control strategies and pointe out the chapter arrangement.Then status the basic structure classification and basic theory of vector control of permanent magnet synchronous motor.Finally,in order to improve the rotational speed response and robustness of the motor,a control algorithm of model predictive control in permanent magnet synchronous motor is proposed in this paper.Based on the mathematical model of the permanent magnet synchronous motor and the basic idea of vector control,this paper designs a double-closed-loop cascade model predictive controller based on current learning compensator.Because of the torque ripple caused by the stator cogging,the air gap flux and external measurement error will affect the stability of the motor,so it is more difficult to control the permanent magnet synchronous motor accurately.According to the characteristics of pulsation periodicity,based on the iterative learning control principle,this paper status a current learning compensation algorithm with the original cascade model predictive controller,and the cross-axis reference current compensation link is added to suppress the torque ripple.The simulation results show that the proposed double-closed-loop model predictive control system based on current learning compensator can effectively reduce the torque ripple of the motor and the dynamic response of the rotational speed response.The robustness is stronger,the controller parameters are easier to adjust and the utility is strong,which provides an effective way for engineering implementation.