论文部分内容阅读
介绍了一种用于有源电力滤波器(Active Power Filter,APF)的神经网络自适应谐波电流检测方法。该方法利用了自适应噪声抵消技术(Adaptive Noise Canceling Technology,ANCT),神经网络采用了线性神经元,权值的调整采用了改进型变步长LMS算法,解决了负载电流突变时跟踪效果差的问题。应用Matlab对该方法进行了仿真研究,仿真结果表明无论负载电流突然增加还是减小,本方法均能快速地检测出谐波电流,且稳态精度高,实现简单,自适应能力强。
A neural network adaptive harmonic current detection method for active power filter (APF) is introduced. The method uses adaptive noise canceling technology (ANCT), the neural network adopts linear neurons, and the modified variable step size LMS algorithm is adopted to adjust the weight. The method solves the problem that the tracking effect is poor when the load current suddenly changes problem. The method is simulated by Matlab, and the simulation results show that this method can detect the harmonic current rapidly with a sudden increase or decrease of the load current, and has high steady-state accuracy, simple implementation and strong self-adaptability.