Research On Sensorless Control Of Permanent Magnet Synchronous Motors Based On Extended Kalman Filtering Algorithm

In recent years,with the continuous improvement of the performance of permanent magnet materials,Permanent Magnet Synchronous Motor(PMSM)have developed rapidly.Due to its advantages such as small size,high power factor,and ease of maintenance,PMSMs have been widely used in fields such as robotics,aerospace,and electric vehicles.In the scheme of PMSM sensorless control,the Extended Kalman Filter(EKF)algorithm,with its strong anti-interference ability,is widely used for rotor position estimation of the motor.However,in the PMSM sensorless control system based on the EKF algorithm,there is a problem of inaccurate statistical process noise during the linearization and discretization of the PMSM mathematical model,which affects the performance of the filter and leads to inaccurate estimation of the motor speed.To solve this problem,this thesis proposes a Kalman filter algorithm that does not rely on the system process noise covariance.The main contributions of this thesis are as follows:Firstly,based on the characteristics of PMSM,this article establishes a mathematical model in the PMSM three-phase coordinate system,analyzes the working process of space vector pulse width modulation,completes the tuning of the PI controller parameters for the speed loop and current loop,builds a PMSM vector control system model based on mechanical sensors in Simulink,and analyzes the simulation results.Secondly,this article deeply studies the basic principles of the EKF algorithm,and based on the linearization and discretization of the PMSM nonlinear mathematical model,establishes the corresponding EKF mathematical equations for PMSM.According to this mathematical equation,a PMSM sensorless control simulation model based on the EKF algorithm is constructed in Simulink,and the hardware and software design of the PMSM control system is completed to transplant the relevant programs of the EKF algorithm in the simulation to the PMSM control unit.Through the analysis of the simulation results of the rotor position and speed curves,the EKF algorithm can estimate the motor speed and position well and has good control performance.Finally,this article addresses the issue of the uncomputable approximation error that arises during the linearization and discretization process of the continuous nonlinear control system for PMSM.An adaptive EKF algorithm that does not rely on the system process noise covariance is proposed.By establishing a mathematical model of the proposed algorithm,the improved algorithm is compared with traditional EKF algorithm and existing EKF parameter adaptive methods in the simulation model and experimental platform of PMSM without position sensor control system.The simulation and experimental results show that using the improved EKF algorithm can adaptively predict the prior estimation covariance,significantly reduce the negative impact of system process noise statistical mismatch on motor speed estimation,effectively reduce the steady-state jitter of the filter,and improve the adaptability of the Kalman filter algorithm.