Improved Finite-Control-Set Model Predictive Control of BLDC Motors driven by Three-Phase-Four-Leg Inverters

This paper presents an improved finite-control-set model predictive control (FCS-MPC) of brushless DC (BLDC) motors driven by three-phase-four-leg inverters. In the conventional MPC of BLDC motors based on three-phase-four-leg drivers, more harmonic currents can be introduced which will deteriorate the steady-state performance of the system. In this paper the concept of duty cycle control is employed to MPC for motor drives. This novel method allocates an active voltage vector in a fraction of one control period and a null vector in the rest of the period, the duration of the active voltage vector is determined by minimizing the cost function to optimize the current flowing into the motor. In addition, the fixed switching frequency of the inverter can be realized by appropriately inserting the action time of zero vector in one cycle through vector modulation. Compared with the conventional method, the simulation results show that the proposed method can further reduce the steady-state torque and reactive power ripples.