Identification of parameters of induction motor with error of speed sensor

Abstract

This paper proposes a new approach to parametric identification of induction motor with error of speed sensor. The ordinary least squares (OLS) estimates are shown to be biased. Generalized total least squares (GTLS) with a singular covariance matrix of the error are propose for estimation of K parameters of induction motor. Identification of parameters takes place in 2 steps. At the first step, two K parameters are estimated using GTLS with a singular error matrix. At the second step, the remaining three K parameters are estimated. An equivalent augmented system is used to solve a biased normal system. Based on the simulation results, a comparison of the accuracy of the estimates of the proposed algorithm and ordinary least squares estimates is given.