State Estimation of Two-Phase Permanent Magnet Synchronous Motor

Abstract

The goal of this paper is to estimate the states of two-phase permanent magnet synchronousmotor (PMSM). The system is highly nonlinear and one therefore cannot directly use any linearsystem tools for estimation. However, if one can linearize the system around a nominal(possibly time-varying) operating point then linear system tools could be used for control andestimation. Firstly, the error covariance matrices of measurement and process would be derivedwhen the system inputs and outputs are subjected to uncertain variations. Then, the corruptednoisenonlinear model of the system will be discretized and extended to be suitable for applyingstandard discrete Kalman filter (KF) for state estimation purpose. The entire state estimatedsystem has been modeled using MATLAB/SIMULINK blocks. The state estimation algorithmand the motor discretized model are coded inside special S-functions of m-file type.