This study proposes a hybrid current controller for a permanent-magnet synchronous motor fed by a two-level inverter. It combines the advantages of model predictive control (MPC) and field-oriented control (FOC), which results in fast dynamics and a zero steady-state error. In addition, space vector pulse width modulation is used to ensure a constant switching frequency and simplify the switching signal generation for the two-level inverter. Two optimal switching methods were proposed to maintain a smooth transition between FOC and MPC and eliminate any undesired bumps or oscillations. These two methods are based on hysteresis average current ripple and artificial neural networks. The proposed switching methods (PSMs) were compared with two recent conventional switching methods (CSMs). Simulation and experimental results were given to validate the effectiveness of the PSMs over the CSMs. The results show that the PSMs have a robust and smooth transition between MPC and FOC in various setups with faster dynamics, minimal overshoot, and zero steady-state error.
