On offset-free continuous model predictive current control of permanent magnet synchronous motors

In this work, an offset-free continuous control set model predictive current control (CCS-MPCC) strategy for synchronous machines based on a slack formulation of the Primal-Dual Interior-Point method is proposed. A horizon of two steps is achieved within 100 µs sampling period. To account for robustness against model mismatch and uncertainty, an incremental formulation of the MPC problem is used to ensure zero steady-state tracking error. The proposed controller is compared with the state of the art Field Oriented Control with PI controllers (FOC-PI), with the Deadbeat Model Predictive Current Control (DB-MPCC), and with the latter controller combined with discrete integrators in the feedback loop (DB-MPCC-I). Experimental results on a 0.5 kW PMSM prove that the proposed CCS-MPCC has outperformed the state of the art control techniques typically used to control electrical machines.