【摘要】 在传统谐波注入法基础上提出一种神经网络自适应谐波电流抑制方法 ,根据自适应噪声抵消技术运用人工神经网络的自适应和自学习特性检测出谐波电流并注入电力系统 ,达到抑制谐波的目的。通过对一典型正弦电流的仿真研究结果表明 ,该方法是可行和有效的 ,它不但有较高的检测精度 ,而且能跟踪检测 ,根据环境的变化能自适应地调整神经网络的权值 ,以便正确地检测出线路的谐波电流更多还原

【Abstract】 An adaptive approach to detection of the harmonic currents based on the classical way of harmonic injection and artificial neural network is presented in this paper. The ANN has teo orthogonal inputs and its weights are adjusted through BP algorithms, its output can vary adaptively with the input. The results obtained from simulation on a typical sine wave show that the approach is feasible and effective. It can detect and track the harmonic currents accurately, for it can adjust the weights of artificial neural network with setting.更多还原