论文部分内容阅读
提出了一种应用于有源滤波器系统的神经网络控制器，能为该系统提供合适的开关控制信号，可应用于单相、三相三线制、三线四线制非线性负载电流波形畸变的抑制。进化算法和反向传播被用于神经网络的训练。仿真结果表明了所提出的神经网络控制器的有效性。
A neural network controller for active filter system is proposed, which can provide suitable switch control signals for this system. It can be applied to single-phase, three-phase three-wire system and three-wire four-wire nonlinear load current waveform distortion inhibition. Evolutionary algorithms and back propagation are used in neural network training. The simulation results show the effectiveness of the proposed neural network controller.