Research On EKF Permanent Magnet Synchronous Motor Speed ​​Control System Based On Particle Swarm Optimization

Permanent magnet synchronous motor (PMSM) are widely used in a lot of occasion where high performances are required due to its excellent performance such as high power density, high efficiency, high reliability and so on. Certain sensors are installed to capture speed and rotor position signal of the motor in the case of traditional direct control (such as vector control and direct torque control) of the PMSM. But the measurement error of the sensors will affect the accuracy and reliability of the system. In order to solve the above problems, some people propose solution methods, such as sensorless technology, which has become a hot topic in recent years. Typical sensorless technology is the Kalman filtere, whose estimation method based on its good dynamic performance and robustness feature.The paper analysises the basic principles of vector control based on the mathematical model of permanent magnet synchronous motor, lists several sensorless technologies aiming at problems brought by traditional vector control as installation of sensors, uses Kalman filter algorithm in permanent magnet synchronous motor control. The application of the Kalman filter algorithm in asynchronous motors, permanent magnet synchronous motors and linear motors are overviewed. The shortage of the Kalman filter algorithm and a number of technical improvements are listed, the perspectives of the Kalman filter in motor applications are indicatedExtend Kalman Filter (EKF) suffers from the noise covariance matrix impacting on the estimated accuracy and the difficulty in selecting the noise covariance matrix. A method to optimize the EKF based on particle swarm optimization (PSO) is proposed. PMSM sensorless speed control system model based on particle swarm optimization EKF is build to solve problems, which is brought by the traditional trial and error method to obtain the parameters, which is not always ideal. By analyzing the principle and implementation process particle swarm optimization, PSO and EKF will be combined to simplify the process of selecting the noise covariance. The simulation model on Matlab/Simulink platform about particle swarm optimization EKF permanent magnet synchronous motor control system is built and simulations are conducted. Simulation test results show that the use of particle swarm optimization EKF observer is contribute to the system to estimate torque and speed with high efficiency and to contribute to the system.