The Sensorless Speed Estimation Based On Extended Kalman Filter For Permanent Magnet Synchronous Motor

In recent years, Motors' effective utilization and control gradually become the main topic of all kind of researching and development sectors. Furthermore, the Permanent Magnet Synchronous Motor(PMSM) with its advantages of high power density, high efficiency, wide application domains and needless external magnetic excitation devices which had been widely used in many alternative current varying frequency systems, and for the PMSM drive, whether hardware or control logistics, also have been more deeply studied. But in conventional motor control theories, most of them need the exact information of rotor speed or position to calculate the rotor model for aligning the control model with actual rotor position, therefore, using various photoelectric, mechanical encoder or sensors to detect the actual rotor speed and positon is indispensable. This kind of devices not only increase the cost of motor assembly, and also lead to bad result of reliability for some special motor running environments, these disadvantages seriously limit the application and domains of PMSM. For removal the accessary sensor or speed detective devices, the aim of sensorless motor control become more popular major topic of all kinds of scholar and academics. This thesis introduces from the classical motor control theory, briefly inscribe the principle of permanent magnet synchronous motor and related basic control algorithm. Introduced and summarized the common control method of sensorless AC motors, and base on the static coordinate system and the revolving coordinate system and Vector Control principle. After looking up the published paper and literatures, this thesis discussed the existing motor speed estimation and speed regulation methods for sensorless PMSM control.In the past, most of papers about discussing of Kalman Filter for motor speed estimation, which based on the steady state or accurate motor parameter models to discuss. These assumptions are too theoretical significance to implement in the practice. This paper aim at the practical condition based on some common motor working condition, through using Matlab/Simulink simulation software to made some simulation, and improved the effectiveness of EKF for motor sensorless speed estimation control. And then, for the motor parameter varying issue, the EKF algorithm has good Robust and fault-freedom properties.