Design Of PMSM Vector Control System Based On Improved SMC Control And Unscented Kalman Filter

Permanent magnet synchronous motor(PMSM)has the advantages of simple structure,superior performance,small size,high efficiency and good stability.It is widely used in electric vehicles,servo robots,aerospace projects and other fields.Although the traditional vector control strategy can support the control requirements of most occasions,in some cases which are susceptible to sudden load disturbance,only relying on the feedback regulation of PI by speed loop can not achieve the purpose of rapid speed adjustment.At the same time,the sensor control system equipped with permanent magnet synchronous motor is also relatively higher in cost than the speed sensorless control system.The speed sensorless control system has lower cost and stronger adaptability to the environment.Therefore,the research and design of PMSM control system with high performance and relatively low cost has considerable practical application value.Firstly,this paper briefly introduces the mathematical model and physical structure of PMSM.Because of the strong coupling relationship of PMSM itself,coordinate transformation is needed to achieve easier manipulation.Therefore,this paper introduces the common methods of coordinate transformation.On this basis,the vector control system model of PMSM is built by using the mature FOC vector control method.Secondly,the Sliding Mode Variable Structure Control(SMC)is used to replace the PI speed modulator in the system.In order to improve the system's ability to resist external disturbances and suppress the self-chattering of SMC,the disturbance observer is adopted to feed forward the output of the system to the input of the current regulator.Simulations and experiments show that this method can weaken the buffeting problem of SMC itself,and improve the anti-load disturbance ability and robustness of the system.Finally,the unscented Kalman filter and the adaptive method are introduced to solve the problem of divergence caused by unknown statistical characteristics of system noise in the filtering process.An adaptive Unscented Kalman Filter(AUKF)is formed,and the convergence of the new algorithm is analyzed and proved.The AUKF is used to design the speed and rotor position angle observer of PMSM,instead of the original vector control encoder as the output of feedback speed signal,to realize the modulation of speed sensorless FOC system.Through the analysis of simulation and experimental results,the new method has good observation accuracy in speed and rotor position estimation.