Improved Parameter Estimation Method for Flux Saturation Model of Synchronous Reluctance Machines

Abstract

This paper proposes a method to improve the accuracy of parameter estimation for the magnetic flux saturation model of a synchronous reluctance motor. The selected flux saturation model includes an arctangent function representing the nonlinear relationship between current and flux. The parameter estimation algorithm integrates the arctangent model with respect to current and transforms the model into a polynomial to which Least Squares Method (LSM) can be applied. For the estimation, measured currents and calculated flux are used and obtained during hysteresis voltage injection test. Because the locus of the current and flux shows the hysteresis band, estimation errors may occur when using the integral. This paper proposed an initialization method for each integration interval to mitigate the error caused by the hysteresis band. The proposed method was validated on an experimental set of 1.5 kW SynRM.