基于HHT和RBF神经网络的串联故障电弧识别

Series Fault Arcs Identification Based on HHT and RBF Neural Network

针对串联故障电弧的识别,提出一种基于HHT(Hilbert-Huang transform,希尔伯特-黄变换)和RBF(radial basis function,径向基函数)神经网络相结合的识别方法。通过对不同负载下正常电流和故障电流信号的EMD分解,得到一系列的IMF分量;提取相关IMF分量的能量熵作为RBF神经网络输入的特征向量,训练RBF神经网络。实验数据分析表明,经过训练的神经网络能够有效地识别串联故障电弧。

A recognition method based on I-IHT （Hilbert-Huang transform） and RBF （radial basis function） neural network is proposed for the identification of series fault arcs. A series of IMF components are obtained by decomposing the normal current and fault current signals under different loads. The energy entropy of the relevant IMF component is extracted as the eigenvector of the RBF neural network to train the RBF neural network. The experimental data show that the trained neural network can effectively identify the series fault arcs.