Low Speed Performance Improvement Of Sensorless Vector Control For Induction Motor Drive Based On EKF

The speed sensorless control can effectively reduce the cost of the motor driver and has high reliability.Therefore,it can replace the speed sensor in the traditional vector control for closed loop speed control.In recent years,the development of high-performance processors has continued to mature,making the extended Kalman filter(EKF)widely used in the field of speed sensorless vector control.However,when the motor runs at low speed for a long time,the stator resistance parameters of the motor are prone to change,which leads to an increase in the voltage drop of the stator resistance and the accuracy of speed estimation is affected.In addition,sudden load changes will also cause the motor speed to fluctuate.Excessive load changes can even cause the motor to run instability,resulting in irreversible consequences.In order to improve the EKF’s speed estimation performance and the ability to resist load disturbances at low speed of the motor,the induction motor speed estimation method with stator resistance parameter identification based on EKF and load disturbance observer is mainly studied in this paper.The main contents are as follows:Firstly,the mathematical model and vector control principle of induction motors are analyzed,including the state equation of induction motors in different coordinate systems and the vector control principle based on the orientation of the rotor field.Secondly,the theory of induction motor speed identification based on the extended Kalman filter and the problem of reduced speed estimation accuracy under low speed operation arc elaborated.Thirdly,the traditional extended Kalman filter is processed to expand the order,which the stator resistance and motor speed are identified at the same time.And the observability of the extended order EKF algorithm on the stator resistance and the influence of stator resistance change on the speed estimation are qualitatively analyzed,the load disturbance observer is designed to suppress the speed fluctuation caused by the sudden load change.Fourthly,the induction motor speed estimation method with stator resistance parameter identification based on EKF and load disturbance observer is simulated and validated by using Matlab/Simulink software.Finally,the experimental platform based on TI’s TMS320F28377D as the main control chip is built,and the proposed algorithm is experimentally verified.The simulation and experimental results show that,compared with the traditional extended Kalman filter algorithm,the induction motor speed estimation method with stator resistance parameter identification based on EKF and load disturbance observer proposed in this paper effectively improves the identification accuracy of the induction motor at low speed,reduces the speed estimation error.And the motor speed is also accurately identified in the case of motor parameter mismatch and load change,which significantly improves the steady state and dynamic performance of the induction motor sensorless vector control system in low speed operation.