Adaptive Observer of Rotor Position and Flux for Salient Synchronous Motor

Abstract

An algorithm of adaptive estimation of rotor flux and angular position for the salient synchronous motor with permanent magnets is presented. A new nonlinear parameterization of the dynamic motor model is proposed. Due to this parameterization the problem of position estimation is translated to the task of identification of unknown constant parameters. During the synthesis of estimation algorithm the currents and voltages of the stator windings, as well as the rotor speed, are assumed to be known signals. Two variants of the adaptive observer based on the standard gradient estimator and the algorithm of the dynamic extension of the regressor are proposed. It is proved that the both versions of the observer provide global exponential convergence of estimation errors to zero if the corresponding regression function satisfies the persistent excitation condition. Also, the latter version of the observer provides global asymptotic convergence if the regression function is square integrable. The results of numerical simulation demonstrate that the algorithm with the dynamic extension of the regressor provides a better quality of estimation transient processes in comparison with the standard gradient estimator.