Improved Traceless Kalman Filter For Position Sensorless Permanent Magnet Synchronous Motor Conrrol

The speed control system based on Permanent Magnet Synchronous Motor(PMSM)is currently widely used in the fields of energy,manufacturing,automobile,aerospace,etc.However,mechanical sensors are susceptible to signal distortion due to environmental and electromagnetic interference.Sensor technology and improving its identification accuracy have become a current research hotspot.This article focuses on the Unscented Kalman Filter(UKF)in sensorless to carry out related research.The thesis first summarizes the advantages and disadvantages of related sensorless technology,introduces the working principle and coordinate transformation of PMSM,gives the mathematical model of PMSM in different coordinate systems,establishes and simulates the traditional vector control model based on PMSM,and then according to UKF The related working principle is combined with the PMSM vector control system to carry out the simulation under variable speed and variable load.The results show that the PMSM speed control system using the traditional UKF has the problems of insufficient identification accuracy and weak anti-interference ability.In order to study the influence of the internal noise matrix on the UKF identification results,the diagonal elements of the covariance matrix Q and R are changed to perform grouping experiments,where the Q and R matrices are 4*4 and 2*2 matrices,respectively.The simulation results show that the larger the matrix R,the closer the center of the identified speed to the given value,but the greater the error with the actual speed.Changing the first two parameters of the matrix Q will affect the convergence result of the identification curve.The larger the result,the closer the result is to the given value.The third parameter reflects the response speed of the entire system,and the increase will speed up the identification speed,while the fourth parameter only affects the identification and The size of the actual speed error range.Aiming at the problems of traditional UKF in the PMSM speed control system,firstly,the fuzzy algorithm is used to adjust the parameters of the noise covariance matrix of UKF in real time,which can have better speed tracking effect under variable working conditions,and improve the speed of the motor.At the same time,the accuracy of the identification at speed,and the peak error between the estimated speed and the motor speed has been reduced by nearly 10 times.Secondly,a load observer was built to perform feedforward control based on the current and speed identified by UKF,reducing the overshoot of the system from the original 16% to 3%,and the adjustment time when the system load changes from 0.05 s to 0.03 s,which is a large margin.Improve the anti-interference ability of the system.Finally,the variable speed and variable load experiments are verified on the RT-LAB semi-physical platform.The results show that the PMSM speed control system that uses fuzzy algorithm to improve UKF and introduces feedforward control has better dynamic performance.