Improved Optimization Process for Nonlinear Model Predictive Control of PMSM

Model-based predictive control (MPC) is one of the most efficient techniques that is widely used in industrial applications. In such controllers, increasing the prediction horizon results in better selection of the optimal control signal sequence. On the other hand, increasing the prediction horizon increase the computational time of the optimization process which make it impossible to be implemented in real time. In order to solve this problem, this paper presents an improved strategy in the field of nonlinear MPC (NMPC) of the permanent magnet synchronous motor (PMSM). The proposed method applies a sequence of reduction weighting coefficients in the cost function, over the prediction horizon. By using the proposed strategy, NMPC give a more accurate response with less number of prediction horizon. This means the computational time is reduced. It also suggests using an incremental algorithm to reduce the computational time. Performance of the proposed Nonlinear MPC (NMPC) scheme is compared with the previous NMPC methods via simulations performed by MATLAB/Simulink software, in permanent magnet synchronous motor drive system. The results show that the use of proposed structure not only lowers prediction horizon and hence computational time, but also it improves speed tracking performance and reduces electromagnetic torque ripple. In addition, using the incremental algorithm also reduces the computational time which makes it suitable for realtime applications.