Digital implementation of full-order flux observers for induction motors

Abstract

This paper deals with flux estimation for induction motor drives by using a full-order flux observer. A problem of full-order flux observers is their need for computationally demanding discretization methods in order to work stably and accurately at high speeds. An implementation of the full-order flux observer using the stator and rotor fluxes as state variables in the stator reference frame and in the rotor reference frame, respectively, was recently proposed. This paper describes how an observer gain can be included in this structure. It is shown that discretization errors of the proposed implementation are small and that there is more freedom to choose an observer gain, even if the simple forward Euler discretization is used.