The traditional model predictive control (MPC) of permanent magnet synchronous motor (PMSM) applies only one voltage vector through ergodic calculation in a control cycle, which makes the control effect is not ideal, and the algorithm has a large amount of calculation. In order to solve this problem, an optimization method is proposed based on the existed two MPC methods, which achieves good dynamic and static performance while reducing switching frequency. This method reduces calculation by judging the sector. Runge-Kutta discrete method with higher precision is used in the discrete process. Lagrange multiplier method is used to calculate duty ratio. Finally, the effectiveness of the proposed method is verified by comparison with the existed two dual-vector MPC methods through model simulation.