Virtual-Vector-Based Model Predictive Control With Deadbeat Solution for Symmetrical Six-Phase Induction Motors

Model predictive control (MPC) has recently been proposed as an interesting alternative for multiphase electric drive regulation because it easily exploits the inherent advantages of multiphase machines. Nevertheless, as the number of phases increase, the employment of MPC becomes more difficult. A key advantage is that the redundancy offered by the extra phases allows greater optimization of the control actions. Symmetrical six-phase induction machines (S6PIM) present advantages over asymmetrical counterparts in terms of control, namely the fact that certain voltage vectors (VVs) offer high torque/flux generation while simultaneously providing null injection of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$x-y$</tex> components. However, the application of a single switching state during the entire sampling period inevitably impacts the torque and currents ripples, deteriorating the power quality. Aiming to overcome such shortcomings, this work proposes a three virtual-vectors-based MPC with Deadbeat control (3VVV-DBMPC). Thanks to the Deadbeat feature, only two control actions are evaluated per sampling time (Ts), resulting in a reduced computational burden. Besides, the inclusion of virtual voltage vectors (VVV) in the control set minimizes harmonic and torque content. The proposed method's effectiveness is demonstrated by experimental data.