Yong Liu’s research while affiliated with The University of Sheffield and other places

A winding copper loss can be significantly increased due to skin and proximity eddy-current effects. The skin and proximity losses due to fundamental frequency current have been investigated in literature, but the influence of pulsewidth modulation (PWM) on the skin and proximity losses has not been reported. In this paper, a 2-D finite element method is employed to analyze the skin and proximity losses in a permanent magnet brushless ac machine, in which significant proximity loss exists due to high frequency current ripples induced by the PWM, as confirmed by both theoretical calculation and experiment. The analyses should be generally applicable to other machines.

This paper describes an improved implementation of direct torque control (DTC) to a permanent-magnet brushless dc (BLDC) drive. The commutation torque ripple, which occurs every 60deg elec. in a conventional three-phase BLDC machine with two-phase 120deg elec. conduction, is reduced by employing a hybrid two- and three-phase switching mode during the commutation periods. It is based on the criterion of minimizing the error between the commanded torque and the estimated torque and does not require knowledge of the conduction duration of the three-phase switching mode. It adaptively adjusts the phase-current waveform to maintain constant electromagnetic torque so that commutation torque ripple, which would have resulted with conventional DTC, particularly at high rotational speeds, is effectively eliminated, as confirmed by both simulations and measurements.

A key issue in the direct torque control of permanent-magnet brushless dc motors is the estimation of the instantaneous electromagnetic torque, while sensorless control is often advantageous. A sliding-mode observer is employed to estimate the nonsinusoidal back-electromotive-force waveform, and a simplified extended Kalman filter is used to estimate the rotor speed. Both are combined to calculate the instantaneous electromagnetic torque. The effectiveness of this approach is validated by simulations and measurements

The application of direct torque control (DTC) to brushless ac drives has been investigated extensively. This paper describes its application to brushless dc drives, and highlights the essential differences in its implementation, as regards torque estimation and the representation of the inverter voltage space vectors. Simulated and experimental results are presented, and it is shown that, compared with conventional current control, DTC results in reduced torque ripple and a faster dynamic response.

A simplified extended Kalman filter (EKF) based sensorless direct torque control technique for a permanent magnet brushless AC drive is proposed. Its performance is compared with that obtained with other sensorless methods for estimating the rotor speed and position from a stator flux-linkage. Since the EKF has an inherently adaptive filtering capability and does not introduce phase delay, the technique provides better speed estimates. In addition, the technique is easy to implement and requires minimal computation.