中文摘要：

将高阶多分辨率奇异熵用于挖掘输电线路故障特征,提出了一种利用电流故障分量的高阶多分辨率奇异熵值进行故障选相的方法,该方法基本不受故障类型、故障位置、过渡电阻和系统摆开角的影响。故障选相过程中,对电流故障分量数据进行小波变换,计算不同尺度下小波变换系数的高阶累积量,并对以高阶累积量为元素的状态矩阵进行奇异值分解,最后计算出奇异值的信息熵,通过比较熵值与设定阈值的大小关系即可判断出故障相别。通过与基于故障电流分量奇异熵值的选相结果和基于全相故障电流高阶多分辨率奇异熵值的选相结果进行比较,表明所提选相方法的效果较好。

英文摘要：

A faulty phase selection method based on the high-order multi-resolution singular entropy of full-phase fault current is presented for the transmission lines,which is not affected by the fault type,fault location,transition resistance and system angular difference.The wavelet transform is applied to the fault current components to compute the high-order cumulants with the wavelet coefficients of different scales,which are used to form the state matrix.The singular values can be then obtained by applying singular value decomposition to it and their high-order multi-resolution information entropies are calculated,which are used to select the faulty phase by comparing them with the threshold set.The comparison of faulty phase selection results between the method based on the singular entropy of fault current component and that based on the high-order multi-resolution singular entropy of full-phase fault current show the later is better.