A Novel Stator Current Estimation Method Based on Embedded Linear Hall Sensors in FSCW-SPMSM

Embedded magnetic encoder (EME) has emerged as a cost-effective and compact solution for rotor position detection in permanent magnet synchronous motors (PMSMs), integrating linear Hall sensors within the motor cavity to detect permanent magnet fields (PMF) and extract rotor position. While conventional EME system prioritize suppressing armature reaction field (ARF) disturbances for high-precision position estimation under load, this article innovatively exploits ARF as information carriers, proposing a linear Hall-based current estimation methodology that transforms interference into a sensing resource. Through decoupled analysis of d/q-axis current-induced amplitude-phase modulation effects on magnetic flux density (MFD) signals at both slot-opening types in fractional-slot concentrated winding permanent magnet synchronous machines (FSCW-PMSMs), this method establishes a relationship between orthogonal currents and electromagnetic signatures. Capitalizing on the inherent immunity of in-phase slot-openings to ARF, it achieves effective separation of PMF interference during loaded operation. Furthermore, it comprehensively accounts for nonideal factors such as Hall sensor misalignment and magnetic saturation, and the algorithm directly quantifies d/q-axis current variations, achieving online stator current monitoring and redundant backup capabilities. The proposed method is validated via simulations and experiments, demonstrating its effectiveness in current estimation.