Current sensorless deadbeat predictive control method for permanent magnet synchronous motor

Abstract

A deadbeat predictive control based current sensorless control method is proposed as a backup method in this paper, by replacing the current feedback with the predicted current at one sampling period ahead, the feedback current information is not needed, thus current sensors can be omitted. Noticing that parameter mismatch will largely influence the control performance of predictive control, q axis inductance is identified online by recursive least square (RLS) method when the motor is functioning with healthy current sensors. Effectiveness of the proposed methods is verified through experiment on a 750W PMSM servo system and it shows that favorable speed response can be achieved.