A New Predictive Torque Control Based Torque Sharing Function for Switched Reluctance Motors

The torque ripple of switched reluctance motors(SRMs) is large due to its doubly salient structure, which has to be suppressed to extend its application. Torque Sharing Function (TSF) is introduced to reduce torque ripple in SRMs. In torque hysteresis loop controller, there are errors in torque observation as the actual torque hysteresis band is larger than simulation, which leads to significant torque ripple. In order to control phase torque more flexibly, phase torque reference value is converted to current value by an inverse T − i − θ look-up table. However, the transformation is complicated. This paper proposes a novel TSF which combines torque sharing method with predictive control method. First, torque reference is separated to three phases by a sinusoidal outline. Second, predictive torque control method is utilized to track the phase torque reference. The simulation model has been built in MATLAB/Simulink to verify the demonstrated performance of proposed method.