Model predictive torque control of induction motor drives with reduced torque ripple

Abstract

Model predictive torque control (MPTC) is emerging as a high-performance control strategy for induction motor (IM) drives, due to its intuitive nature, flexibility to incorporate constraints and quick dynamic response. However, the implementation of MPTC requires high computational ability and the use of single voltage vector during one control period fails to reduce the torque ripple to the minimal value. This study proposes an improved MPTC for IM drives with reduced torque ripple and low complexity. On the basis of the relationship between stator current and stator flux, the complicated current prediction for each voltage vector is eliminated, reducing the control complexity significantly. Torque ripple reduction is achieved by allocating only a fraction of control period to the active vector selected from conventional MPTC, whereas the rest of time is allocated for a null vector. Two kinds of methods for optimising the duty ratio of the active vector are proposed and evaluated in detail. Presented experimental results prove that, compared with conventional MPTC, the proposed MPTC achieves better steady-state performance by reducing the torque ripple significantly. Meanwhile, the quick dynamic response of conventional MPTC is reserved.