Extended Prediction Horizon Multi-objective FSMPC without Weighting Factors

Extended prediction horizon FS-MPC shows its merits of a low current harmonic distortion per switching frequency. However, the weighting factors are hard to be optimized in an extended prediction horizon FS-MPC with more involved control objectives. A multi-objective FS-MPC with a longer prediction horizon without the involvement of any weighting factors is proposed in this paper. The descriptions of control plant and FS-MPC with a prediction horizon of 3 are presented. The control objectives, torque, flux and switching frequency are optimized in the cost function. Therefore, there are two weighting factors to be optimized. To cope with the abovementioned issue, a ranking evaluation approach is proposed for weighting factor optimization in the proposed FSMPC method with an extended horizon to obtain two possible vectors, which are optimized in the next prediction horizon. The vector with a minimal ranking value is selected as the optimal. The simulation results of the proposed method on a pair of 2. 2kW induction machine drives are evaluated and compared.     «

Extended prediction horizon FS-MPC shows its merits of a low current harmonic distortion per switching frequency. However, the weighting factors are hard to be optimized in an extended prediction horizon FS-MPC with more involved control objectives. A multi-objective FS-MPC with a longer prediction horizon without the involvement of any weighting factors is proposed in this paper. The descriptions of control plant and FS-MPC with a prediction horizon of 3 are presented. The control objectives, t...     »