Study On The BLDCM Sensorless Control System Based On Kalman Filter

Brushless DC motor(BLDCM) is widely used in automotive electronics, household appliances, aerospace, industrial automation and other fields with its advantages of high efficiency, simple control and good speed performance. Brushless DC motor uses electronic commutation instead of mechanical commutation, which increases reliability and service life. But Brushless DC motor needs to install the position sensor to provide commutation signals, which adds complexity and cost of the system. Therefore, the abolition of the sensor device and improvement of the motor performance are very important. It also gradually becomes the current research focuses of Brushless DC motor.This paper studied the sensorless BLDCM control system based on the extended Kalman filter(EKF) algorithm, and achieved the sensorless estimation of rotor position and motor speed. This method has good dynamic performance and is not sensitive to the variation of motor parameters, which can realize the sensorless control of Brushless DC motor well. However, the traditional EKF method only has a first-order Taylor series of the nonlinear system accuracy when the equation is linearized. Therefore, in practical applications, the further improvement of the EKF is necessary. Aiming at this problem, this paper deeply studies an improved Kalman filter UKF. The results show that UKF algorithm performed better than the EKF in application of sensorless BLDC control system.Firstly, the mathematical model of BLDCM is given. The speed and current double closed-loop system simulation model based on Matlab / Simulink is built.Secondly, EKF algorithm is described in detail. Sensorless BLDCM control system simulation model based on EKF is set up combining the state equation of BLDCM. The simulation results show that the application of EKF algorithm in sensorless BLDC control system is feasible.Thirdly, aiming at the shortcoming that EKF algorithm only has the first-order accuracy, an improved Kalman filter algorithm Unscented Kalman Filter(UKF) is descripted and analyzed in detail. This paper applies UKF in BLDCM and designs a sensorless brushless DC motor control system based on UKF. The simulation results show that this method can improve the tracking performance in sensorless BLDCM control system.Finally, the hardware platform of sensorless BLDCM control system is designed and the software program is written. The experiments verify the feasibility of Sensorless BLDCM control system based on EKF.