电力电缆故障测距近年来是我国的热门研究课题。随着社会的飞速发展，保障线路 的安全运行，快速、准确的实现电力线路故障检测与定位，具有显著的经济效益与社会 效益。 本文综述了电力电缆线路故障测距的国内外现状。以小波理论、信息熵理论为基础， 对电缆故障的精确定位方法、系统建模方面进行了研究。文中先对故障信号进行消噪、 相模变换、归一化的预处理，再针对单端法进行故障测距时，难以判断检测到的波头来 自故障点反射波还是对端母线反射波，研究了利用小波包熵函数确定中间熵阈值对故障 所在区域进行判别的方法，给出了熵与线路长度关系曲线图，最后本文提出的一种小波 差分算法实现故障定位功能。 小波差分算法主要思想包含如下几个方面：首先对含有故障的线路的故障信号进行 小波模极大值分析，确定故障行波初始浪涌所在序列，然后对初始浪涌后的极大值点按 从大到小排序，将所获得的模极大值进行差分，选择适当的阈值，取相对位置较大的模 极大值点为故障点对应位置，并采用不受波速影响的单端测距算法实现了电缆故障定 位。这样既能保证故障信号极大值的准确获取，又能消除其他干扰信号的影响，使得故 障测距精度有了较大提高。 论文最后用实际故障信号的数据对所提出的电缆故障测距方法进行了验证。验证结 果表明，该故障测距方法及软件设计正确，实现了电缆故障定位，并提高了故障测距精 度。

In recent years, power cable fault location is a hot research topic in China. With the rapid development of the community, fast and accurate implementation of power line fault detection and location has significa nt economic and socia l benefits to ensure the safe operation of the line . The power cable line fault dia gnosis of home and abroad is reviewed in this paper. Based on wavelet theory and information entropy theory, the precise method of cable fault location and system modeling aspects is studied. Firstly, pretreatment of fault signa ls is done by noise cancella tion, phase-mode transformation and norma lization. Secondly, aiming at the difficulty to determine whether the eflected wave is from the point of failure or from the other end of the cable, the wavelet packet entropy method is used to discrimina te the fault regions, where the midd le entropy threshold is the midd le fault point, the curve graph of the relationship between the wavelet packet entropy and line length is given. Finally, a wavelet difference algorithm is proposed to achieve fault location function. The ma in idea of wavelet difference algorithm includes the following operations. Firstly, process the traveling wave signa l of fault transmission line with modulus ma xima of wavelet analysis, determine the initia l burst sequences of fault traveling wave, array the great value points after the initia l burst by descend ing order, process these modulus maximum values with differential action. Then, the relative ly larger position of the modulus ma ximum point is correspond ing to the failure location after select ing the appropriate threshold, the one end of cable fault location algorithm which is not impacted by velocity, is used to calculate the fault location .This method will not only ensure accuracy acquisition of fault signa l ma ximums, but also eliminate the impact of other interfering signa ls, the fault location accuracy has been greatly enhanced. Finally, the proposed method of cable fault location has been tested by the data from the real fault cable. Experimental results show that the software design of fault location is correct, the cable fault location is achieved, and the accuracy of fault location is improved.