Research On PMLSM Sensorless Control Based On Improved Cubature Kalman Filter

Permanent magnet linear synchronous motors(PMLSM),with the advantages of high thrust density,high power,low loss and fast dynamic response,are applied in many fields,such as modern industrial production,transportation,carrier-based aircraft lifting,electromagnetic ejection and logistics sorting extensive,and its control strategy has received extensive attention from domestic and foreign researchers.The traditional high-precision closed-loop control uses a mechanical sensor to collect the speed and position signals of the motor.However,the increase of mechanical sensors will bring many problems,such as increasing the size of the motor installation and the installation cost of the drive system.It is difficult to operate normally under severe working environment,which has a serious impact on the reliability,dynamic and static characteristics of the control system.Therefore,the sensorless control technology can well solve the above problems and achieve high-precision control of the motor.In this paper,a sensorless vector control system of PMLSM is built based on the cubature Kalman filter(CKF),which is widely used in nonlinear state estimation.The simulation results show that the algorithm is effective.Because the standard CKF algorithm lacks the adaptive ability to deal with the noise change,and the state covariance matrix is easy to lose the positive definiteness,resulting in the filter accuracy reduction and filter divergence defects,there is a large estimation error between the actual speed and the estimated speed.In order to improve the accuracy of motor state estimation,an adaptive square root cubature Kalman filter(ASRCKF)algorithm is proposed based on CKF and square root filtering algorithm and noise statistical estimator,which can be applied to sensorless control of PMLSM.This method can ensure the nonnegative nature of state covariance matrix in the process of filtering,and can adapt to the noise change.When the state dimension of the system is high,the noise covariance matrix is improved.The time-varying biased noise covariance matrix estimator can ensure that the noise covariance matrix is always non negative in the filtering process.The effectiveness of the algorithm is verified by two simulation examples.The PMLSM sensorless control system based on the improved CKF is built by using the improved cubature Kalman filter algorithm.The simulation results show that the improved CKF has a significant improvement in the speed and position estimation accuracy of the permanent magnet linear synchronous motor.The maximum tracking error percentages before and after load mutation are 0.4286% and 0.1468% respectively,and the error percentages after stable tracking are 0.0457%,Moreover,the stability of the filter structure has been greatly improved,and it can also track quickly when the load of the motor changes suddenly.