Research On Control Algorithm Of Permanent Magnet Synchronus Motor Based On Current Prediction And Kalman Speed Measurement

With the development of industrialization,the servo system plays an increasingly important role in the field of industrial production.Permanent magnetic syschronous motor(PMSM)has been widely used in medium and small capacity servo system because of its small volume,simple structure,high efficiency and high power density.The high performance and digital control of PMSM servo system are the hot spot in the current research of servo system.Therfore,it is extremely important to study various advanced control algorithms to realize the high performance PMSM servo system.The main goals of this project are to improve the dynamic charateristics and low speed performance of the servo system.The details are as follows:The deadbeat current predictive control algorithm has many advantages,such as fixed switching frequency,fast dynamic response and the optimal state.Therefore,the paper selects the deadbeat current predictive control algorithm.Deadbeat current predictive control is essentially a model predictive control algorithm that requires accurate motor parameters to ensure the performance of controller.In order to improve the robustness of the predictive control system,this paper adopts an improved deadbeat current predictive control algorithm.The current deviation contraint and the output voltage prediction method are modified so that the system can still be stable when inductance paramerer is mismatched.The control accuracy of the servo system depends mainly on the low-speed performance.The exsiting PMSM drivers focused more on the rated-speed control permanence,while the low-speed control performence is not satisfactory.Therefore,improving the low-speed performance is significant for improving the entire performance.In order to control PMSM precisely in the full speed range,accurate speed feedback must be obtained,especially when PMSM is running at low and ultra low speed.The traditional M method based on the incremental photoelectric encoder has a large detection error at low speed,which seriously affects the control performance of the servo system.In order to solve the problems exsiting in the traditional speed measurement method,the Kalman filter technology is used to improve the speed detection accuracy.Finally,the paper builds the experimental platform based on TI's TMS320F28335,and designs the hardware and software parts of the system.The correctness and rationality of the system's hardware circuit and software are verified by the detailed tests.Meanwhile,the validity of the algorithm is also proved.