Injection-Based Self-Sensing of Rotor Displacement in Combined Winding Bearingless Motors

This paper investigates rotor displacement self-sensing (sensorless) algorithms for surface permanent magnet bearingless motors (SPMBMs). The techniques are derived for and applied to multi-phase combined (single) windings where each phase contributes to both torque and force production. First, a non-linear analytic model based on modified winding function theory is presented for the SPMBM which includes the effects of time-varying rotor displacements and rotor rotation. Then, the state-of-the-art high-frequency (HF) injection methods for rotor displacement self-sensing are reviewed and applied to the model. It is shown that these methods apply well at low motor loading, but do not extend to the full-load/high-speed region where bearingless motors are desired to operate. Analytical equations are derived which predict the maximum load and rotational speed permissible for state-of-art self-sensing methods. An improved method is proposed for current component frequency separation which can effectively isolate the HF response. Simulations validate the analytically-predicted motor load limitations and the proposed algorithm. Realistic HF injection self-sensing is simulated for a SPMBM at 90 kRPM with 10 kW of mechanical output power. The results herein motivate displacement self-sensing as a practical alternative to physical gap sensors.