Torque-Ripple Mitigation for Brushless DC Machine Drive System Using One-Cycle Average Torque Control

This paper investigates the torque-ripple mitigation method for a brushless dc machine caused by nonideal trapezoidal back electromotive force (EMF) using a novel one-cycle average torque control algorithm. This algorithm can ensure that the average torque follows the reference in each switching cycle. The feedback average torque is computed using the energy flowing into the system in each control cycle. No sophisticated observer is needed since neither the back EMF nor accurate rotor position information is required. No current sensor is needed to measure the phase current. Only two sensors are used to obtain the dc-bus voltage and current information to calculate the input energy. Simulations and experimental results are demonstrated to verify that the proposed method is able to reduce the major lower harmonics of the machine torque ripple by more than 70%.