Nonlinear State Estimation for Induction and Permanent Magnet Motors

Abstract

This paper presents a nonlinear state estimation for induction and permanent magnet synchronous motors (PMSM) using a recently developed cubature information filter (CIF). The CIF is based on extended information filter and cubature Kalman filter. One of the important properties of the CIF is its ability to deal with highly nonlinear systems without evaluating the Jacobians of state and measurement dynamics. CIF belongs to information filters family. The update step in CIF is computationally inexpensive than its counterpart covariance filters. In this paper, the speed of an induction motor and the speed and rotor position of a PMSM are estimated using CIF. Monte-Carlo simulations have been performed to validate the CIF for these electric motors and is compared with results from extended information and unscented information filters.