Speed sensorless model predictive current control with ability to start a free running induction motor

© The Institution of Engineering and Technology. In this study, a speed-sensorless finite control set-model predictive current control method is proposed based on an adaptive full order observer. The control system features simplicity and low cost because of no requirement for speed measurement, modulator and tuning of weighting factors. In most sensorless based schemes, the motor is assumed to be started from standstill. There is limited research considering starting a free running motor with unknown rotational direction and speed. To start a free-running induction motor (IM), the feedback gain matrix is designed to guarantee the convergence of estimated speed to actual speed even with incorrect initial value. To improve the efficiency, amplitude of flux is adjusted along with load condition. The presented results show that the proposed method is able to smoothly start an IM with unknown initial speed and work well over a wide speed range. The effectiveness of the proposed method is verified by both simulation and experimental tests on a two-level inverter fed IM drive platform