基于行波特征分类的有源配电网故障定位

随着分布式电源(distributed generator,DG)的接入,配电网的潮流方向和结构发生改变,许多传统配电网的故障定位方法已不再适用。单相接地故障是配电网常见故障且可能带来二次故障乃至断电等危害,从线模行波小波特征值与含DG的配电线路故障区段之间的关系入手,通过线性判别分析(linear discriminant analysis,LDA)降维挑选出最优故障特征,再利用机器学习中与该模型契合最好的基于核分布的贝叶斯构造分类模型,实现单相接地故障定位新方法。构建含DG的IEEE 33节点模型对有源配电网不同区段的故障进行实验,得出最优三维特征样本的定位准确率为97.9%,结果表明该方法能实现故障的准确定位。

Fault location of active distribution network based on traveling wave feature classification

With the access of distributed generator (DG), the power flow direction and structure of distribution networks have changed. Therefore, many traditional fault location methods are no longer applicable. Single-phase-to-ground fault is a common fault in distribution networks, which may bring secondary fault and even blackout. Based on the relationship between the wavelet eigenvalues of traveling waves and the fault section of distribution lines with DG, the optimal fault features were selected by dimension reduction of linear discriminant analysis (LDA), and then the Bayesian construction classification model based on kernel distribution was used to realize a new method of single-phase-to-ground fault location. The IEEE 33 bus model with DG was constructed to test the faults in different sections of active distribution networks. The location accuracy of the optimal three-dimensional feature sample reached 97.9%, demonstrating that the proposed method can achieve accurate fault location.