关键词: 改进型无差拍控制, 拉格朗日插值法, 重复预测, 自适应前向线性电流预测, 复合预测

Abstract: Aiming at the problems of time delay and low prediction accuracy of traditional deadbeat control, a deadbeat current control based on composite prediction is proposed. This method first analyzes the reasons for the time delay and low prediction accuracy. Then, when the load is stable, Lagrange interpolation method and repeated prediction are combined to improve the current prediction accuracy. When the load changes, adaptive forward linear current prediction is introduced to shorten the system oscillation time and enhance the dynamic response ability. Finally, the two methods are combined for control. The inverter output current is predicted one beat in advance and the sampling reference current is predicted two beats in advance, which solves the problem of time delay. The simulation results show that this method reduces the harmonic content by 92% and the oscillation time by 50% compared with that before compensation. It has higher prediction accuracy and dynamic response ability, and better control effect.

Key words: improved deadbeat control, Lagrange polynomial, repeated prediction, adaptive forward linear current prediction, compound prediction