四桥臂APF虚拟矢量模型预测电流控制

Four-legs APF Virtual Vector Model Predictive Current Control

模型预测电流控制(MPCC)由于具有良好的动态性能而被应用于有源电力滤波器(APF)中。针对传统MPCC跟踪误差大的问题,应用一种虚拟矢量MPCC,并在此基础上进行优化。为减小开关损耗,对脉冲宽度调制过程中构成虚拟矢量的基矢量的作用顺序和其中零矢量的选取进行优化,离线计算出基矢量的作用时间。为减少滚动优化次数和程序运行时间,提出一种矢量区域划分方式,寻优前对参考电压矢量进行预测并对其所在区域进行判断。为解决控制过程中直流侧电压变化引起的问题,对直流侧电压进行拉格朗日插值预测。最后通过仿真和实验验证了该控制方法的有效性。

Model predictive current control(MPCC)is used in active power filter(APF)because of its excellent dynamic performance.The traditional MPCC has big tracking error,so a virtual vector MPCC is used and optimized based on it.For decreasing switching loss,the order of base vectors that form virtual vectors in pulse width modulation and the choose of zero vector are optimized,the active time of base vectors is counted offline.For decreasing the number of rolling optimization and the running time of program,a way to divide the area of vectors is proposed,the reference voltage vector is predicted and the area of vector is defined before optimizing.The Lagrange’s interpolation is used to predict direct current side voltage to solve the problem caused by its change in control process.Finally,the effectiveness of the control way is verified by simulation and experiment.