Research On PMSM Sensorless Control And Its High Performance Operation

Permanent magnet synchronous motor(PMSM)is an excellent motor design.It has simple structure,high power density,excellent control performance and wide application prospects.It has attracted the attention of relevant researchers at home and abroad.Especially in recent years,China has attached great importance to the development of rare earth resources and domestic PMSM.It provides a lot of high quality and cheap raw materials for the design and manufacture of permanent magnet synchronous motors in China.Under the influence of market economy,it develops the design and control technology has made considerable progress.PMSM control can not be separated from its rotor position information.Traditional control systems often use mechanical position sensors such as encoder or resolver to obtain position information.This way of measuring position information directly ensures the control accuracy of motor,but there are some problems in volume,cost and robustness,which limit the application of the system in more occasions.In the past few decades,a great deal of research has been done on sensorless control technology.For example,the high frequency injection method proposed by Robert D.Lawrence et al.has achieved good control performance at first,and then has been further studied.The purpose of this paper is to optimize and integrate the existing sensorless control technology of PMSM in order to improve its speed control accuracy.The specific research contents are as follows:The traditional Extended Kalman Filter(EKF)is optimized.By extending the state feedback vector of the Kalman Filter in the iteration process,it can combine the discrete position signals of Hall elements,and then enrich the information density of the feedback vector,which can be used to improve the position and the position of the system.The accuracy of velocity estimation.In the design of the system,by setting the variables in the iteration process of Kalman filter,the estimation of the final position and velocity of the system is flexibly correlated with the position information obtained by Hall signal,even if the contribution of Hall signal to the two estimates is kept within a reasonable range.At the same time,this control method is combined with the high frequency injection method,which has good control effect at low speed.Through the reasonable arrangement of the transition process of the two systems,the computer simulation shows that the above mentioned methods can effectively improve the sensorless control.Control accuracy of the system.In the appendix and article,the simulation model and the method used are given.