并联混合型有源滤波器电流跟踪的迭代学习控制

在并联混合型有源电力滤波器控制中,由于传统PI控制只能对直流给定信号进行无差调节,对周期量谐波滤波效果不够理想。该文根据稳定性负载电网谐波具有重复周期性特点,提出了PI型迭代学习控制算法,并引入遗忘因子学习律,增强系统鲁棒性。同时通过模糊推理构造参考电流误差的D型学习律前馈环节,提高了系统的跟踪精度。该控制方法计算量小,易于实现。仿真分析与现场运行结果表明了该控制方法的有效性。

Iterative Learning Control for Current-tracking of Shunt Hybrid Active Power Filter

Due to the traditional PI controller can only deal with given direct current signal,and can not control cycle harmonic signal well,the performance of shunt hybrid active power filter used traditional PI controller is not satisfied.According to the character of re-period of current harmonics in steady-load power system,this paper presents an algorithm of PI-type iterative learning control,in the same of using a forgetting factor to enhance systemic robust;and constructing a feed forward based D-type learning law of inferenced current error by fuzzy reasoning to improve its dynamic performance.The proposed algorithm is small computation capacity and easy to realize.Simulation and engineering results confirm the value of the algorithm.