Model Predictive Current Control Strategy with Reduced Computation Burden

Abstract

In this paper, three model predictive current control (MPCC) schemes for permanent magnet synchronous motors (PMSM) are studied. The first control scheme is the traditional optimal duty cycle model predictive current control (ODC-MPCC). In this scheme, according to the principle of minimizing the cost function, the optimal voltage vector is selected from the six basic voltage vectors which are optimized simultaneously with the duty, and then, the optimal voltage vector and its duty are applied to the inverter. In order to reduce the computational burden of ODC-MPCC, a second control scheme is proposed. This scheme optimizes the voltage vector control set, reducing the number of candidate voltage vectors from 6 to 2. Finally, according to the principle of minimizing the cost function, the optimal voltage vector is found from the two voltage vectors, and the optimal voltage vector and its duty cycle are applied to the inverter. In addition, in order to further improve the steady-state performance, another vector selection method is introduced. In the combination of voltage vectors, the third control scheme extends the combination of voltage vectors in the second control scheme. The simulation results show that the second control scheme not only reduces the computational burden of the first control scheme but also obtains steady-state performance and dynamic performance equivalent to the first control scheme. The third control scheme obtains better steady-state performance without significantly increasing the computational burden and has dynamic performance comparable to the first and second control schemes.