基于前向线性预测理论的混合电力滤波器谐波电流预测方法

A Harmonic Current Forecasting Method for Hybrid Active Power Filters Based on the Forward Linear Prediction Theory

对谐波电流进行预测是提高混合电力滤波器补偿效果的有效方法。提出基于前向线性预测理论的混合电力滤波器谐波电流预测方法。推导出最佳预测系数正则方程、谐波电流最小预测误差和阶更新方程,给出1～3阶滤波器的预测系数和最小预测误差值。该方法能根据当前电网或负载的谐波电流和历史经验数据,计算出下一时刻的谐波电流,通过自适应预测控制策略实现最小误差补偿。仿真和实验结果表明,该方法具有预测精度高、快速跟踪能力强和补偿效果好等特点。

It is considered as an effective method to improve the effects of hybrid active power filters on harmonic current prediction.A harmonic current forecasting method for hybrid active power filter based on the forward linear prediction theory was proposed in this paper.The optimal forecasting coefficient canonical equation,minimum forecasting errors and the rank update equation were deduced.As an example for low order filters,the forecasting coefficient and minimum forecasting errors of one-three order filters were given.According to harmonic currents and historical data of the on-line grid structure and loads,this method forecast harmonic currents of the next time,and could realize minimum errors compensation by the adaptive forecasting control.The simulation results show that the method has high forecasting precision,good fast tracking performance and compensation effects.