An MRAS-based Sensorless Control Algorithm for Permanent Magnet Brushless AC Machines

This paper presents a sensorless control scheme for Surface-Mounted Permanent Magnet Brushless AC Machines (PMBACMs), which estimates both electric rotor speed and position through a Model Reference Adaptive Systems (MRAS) approach. In particular, starting from the PMBACM sampled-data model, a priori and a posteriori quantities are introduced, based on which the adaptive law is set in accordance with the Popov’s Hyperstability Criterion. Additionally, the issues arising from the duality of the solution computed by the MRAS-based algorithm are overcome by the support of Hall sensors. The effectiveness of the proposed MRAS-based Hall-assisted sensorless algorithm is verified through numeric simulations, which also regard the employment of a double MRAS-based algorithm to enhance rotor speed and position estimation further.