计及参数误差的永磁同步电机最优虚拟矢量预测电流控制

Limited by the number of voltage vectors, there are large current ripples in the finite control set predictive control with vector duration optimization, and the accuracy of current prediction is highly dependent on precise motor parameters. However, these motor parameters may not match with their actual values because of magnetic saturation, and parameter errors during operation reduce current control accuracy. Therefore, an improved predictive current control strategy based on the optimal virtual voltage vector is proposed to improve current control performance. Voltage control set is extended to the hexagonal region by virtual voltage vectors, which are composed of basic voltage vectors. To reduce prediction calculations, the sector of the optimal virtual vector is determined by the deadbeat control principle and the basic voltage vectors of the sector are selected for calculating the virtual voltage vector. Then, a current cost function with feedback correction of prediction errors is designed. The durations of basic vectors are calculated based on the cost function, and the optimal virtual vector is established to improve current control accuracy. The proposed method significantly reduces current ripples and calculations, and also improves parameter robustness. Simulations and experiments were carried out on a permanent magnet synchronous motor drive system with a two-level inverter. The results verify the effectiveness of the proposed method.