Full-Cycle Iterative Observer : A Comprehensive Approach for Position Estimation in Sensorless Predictive Control of SynRM

Abstract

This article presents a new iteration algorithm for the position estimation of synchronous reluctance motors (SynRMs) by unifying the active flux observer and Newton-Raphson (NR) algorithm. Thus, a full-cycle iterative observer (FCIO) is derived, in which the active flux is estimated in each iteration of the NR technique. Hence, the dependency of the active flux to the position is considered in the NR formula. This guarantees the accurate convergence of the position. Furthermore, the stator flux observer is enhanced with nonlinear feedback to attain a robust sensorless control scheme, compensating for the stator resistance mismatch. The experimental results confirm the superior performance and improved robustness of the proposed sensorless technique compared to the standard NR-based sensorless control method.