Multiscalar Model-Based Predictive Torque Control Without Weighting Factors and Current Sensors for Induction Motor Drives
Anxin Yang, Ziguang Lu
Abstract
This article focuses on eliminating the weighting factor in traditional predictive torque control (PTC) and considers the control scheme without current sensors. A current sensorless direct PTC method without the weighting factor for induction motor (IM) drives is proposed. Based on the multiscalar model, the controller directly predicts the torque and its dual quantity, instead of predicting the stator flux and stator current in advance and then indirectly calculating the torque like PTC. Since the torque and its dual quantity (the inner and external products of the flux and stator current) have the same units and dimensions, the design of the cost function does not require a weighting factor. Furthermore, an adaptive virtual current observer is used to reconstruct the stator currents. The estimated currents replace the measured currents to realize current sensorless, which can avoid the measurement noise caused by the current sensors, thus reducing the current and torque ripples. Simulation and experimental results illustrate the effectiveness of the proposed approach under different working conditions.