To address the problems of large harmonic currents and poor robustness in conventional virtual-voltage-vector-based model predictive control for dual three-phase permanent magnet synchronous motors(DTP-PMSM)

a duty-ratio-modulation-based model-free predictive control method for DTP-PMSM is proposed. Firstly

to address the problem that the predictive control performance of DTP-PMSM is easily affected by motor parameters

an extended state observer-based ultra-local model was established. Virtual voltage vectors synthesized from three adjacent large vectors were selected as the candidate set according to the volt-second balance principle to suppress harmonic currents. To further reduce torque ripple and computational burden

the sector of the optimal voltage vector was obtained based on the deadbeat principle. Meainwhile

the duty cycle and application time of the optimal voltage vector were calculated through the cost function. Moreover

the PWM switching sequence was optimized to eliminate the state conflict in virtual voltage vector synthesis. Finally

effectiveness of the proposed method was verified on the experimental platform for three-phase permanent magnet synchronous motors that had been constructed. Experimental results indicate that

compared with conventional model predictive control

the developed method significantly reduces harmonic currents and exhibits strong robustness.