Sensorless Control for PMSM Drives Using the Cubature Kalman Filter based Speed and Flux Observer

This paper studies the design and implementation of Cubature Kalman filter (CKF) for sensorless permanent magnet synchronous motor (PMSM) drives. Considering the inherent shortcomings in using the extended Kalman filter (EKF), alternatively, the CKF is derived as an improved derivativeless nonlinear Kalman filter. In CKF, the state distribution is represented by a minimal set of weighted sample points provided by the third-degree spherical-radial cubature rule. These points, when propagate through a nonlinear transformation, present the posterior mean and covariance of a Gaussian random variable accurately up to the 3rd order Taylor series expansion for any nonlinearity. In this paper, a CKF-based speed and stator flux estimator is proposed for the space vector modulation (SVM) based direct torque control (DTC) scheme. The use of a new state variable representation of the motor dynamics deals with the multi-solution problem which may cause the diverging of the motor system during the start-up stage. The design procedure and properties of the CKF based control algorithm are studied carefully in this paper, and implemented by a digital signal processor. Both simulation and experimental test are carried out to verify the validity and effectiveness of the proposed sensorless PMSM drive.