Speed Control Of Permanent Magnet Synchronous Motor With A Self-Adaption Kalman Observer

Permanent magnet synchronous motor(PMSM) has been widely used in the industrial areas due to its simple structure, high power factor, big torque current ratio, etc. In the process of vector control of permanent magnet synchronous motor system, rotor position and speed signal are the necessary feedback signals, which can be obtained by an optical encoder, rotary transformer and an algorithm of position and speed measurement. Among them, an absolute encoder can read the motor rotor position directly, which is suit to permanent magnet synchronous motor system. In addition, the measurement speed is used as a feedback signal to follow the given speed in the control process of the PMSM system, thus the quality of the measurement speed affects the motor working property necessarily. The paper mainly conducts some researches around the influence of the measurement speed on the motor working property and the relevant aspects such as motor control.The paper firstly detailed analyzes the measure mechanization of the absolute encoder, and the mathematical analyses about speed measurement error and time delay of traditional speed measuring methods are done. Then the influences of the speed measurement error and time delay on the motor working properties are analyzed in a theoretical point. In order to solve the problems of low speed measurement precision, long time delay of traditional speed measurement methods and improve the bad dynamic performance of the traditional Kalman observer, this paper proposes a measurement noise self-adaption Kalman observer, it is designed on the ideas of making up measurement noise of the absolute encoder with limited resolution ratio and improving the dynamic tracking ability of the algorithm. Based on the mathematics model of PMSM, the parameters of the observer are designed to observe the rotor position, speed and load torque. In the condition that the motor working state has a sudden change, the proposed observer calculates the current measurement noise in real time and updates the gain matrix which can ensure the precision and no-time-delay of the algorithm. In addition, to improve the anti-disturbance and speed tracking abilities of the motor system, a PMSM feed-forward and feedback control system is built using the observed signals as the speed feedback signal and load compensation signal.The paper builds an experimental platform of permanent magnet synchronous motor system to make an experimental research on the proposed self-adaptive Kalman observer using in the PMSM system. The experimental results show that the precision and time delay of speed measurement method affect the motor working property. The proposed self-adaptive Kalman observer can observe rotor position, speed and load torque timely and precisely. Using the observed speed and load torque into the feed-forward and feedback control can improve the dynamic and steady properties of the permanent magnet synchronous motor.