Multi-Objective Predictive Control of Two-Motor Drive System with Five-Leg Inverter

Dual-motor drive systems have been widely favored due to many advantages, including reduced dimensions and cost. In this paper, a model-based predictive control (MPC) method for a dual-motor drive system fed by a five-leg inverter (FLI) is introduced. Among the advantages of the MPC method, the independent and fast tracking of reference control variables and the elimination of cascade control structures dependent on the modulator for multi-motor systems can be mentioned. PI loops have disadvantages such as delayed time response, as well as design limitations of PI coefficients due to the FLI structure. In this paper, using the predictive control method based on the multi-objective model proportional to speed and current (MOMPC), the PI control loops have been removed. One of the challenges of this structure is how to allocate the DC link voltage to the motors. For this purpose, by defining the duty cycle corresponding to the steady-state voltage of the motors, the DC link voltage of the inverter is shared between the motors. In addition, by using this method, the control objectives of two motors become independent from each other, which reduce the torque ripple and current ripple of the motors.