并联有源电力滤波器的人工神经网络控制方法

An Artificial Neural Network Method Used for Control of Shunt-wound Active Power Filters

提出并分析了一种适用于控制并联有源电力滤波器的人工神经网络方法 .该方法采用两个不同的人工神经网络 ,其中的多元自适应线性神经元网络用以获取欲补偿的三相有害电流 ,而其中的三层前馈神经网络用以控制滤波器的输出电流 .仿真结果显示 :电力系统经并联有源电力滤波器补偿后 。

An artificial neural network method for control of shunt active power filters is presented and analyzed. The method contains two different artificial neural networks. One is multi adaptive linear artificial neural network that is used to tackle the three phase deleterious currents to require compensating, and another is three layer feed forward artificial neural network that is adopted to control the output currents of the proposed filter. Simulation results indicate a remarkable improvement in source current waveforms of power distribution system through compensation of the proposed filter.