随着现代电力系统的飞速发展，配电网的结构也越来越复杂，其故障测距方法已无法满足系统运行的稳定性及可靠性。相比于传统的测距方法，行波法不受系统运行方式、过渡电阻等因素的影响，使其受到了广泛的关注，成为故障测距技术的研究热点。本课题主要采用快速独立分量法对配电网故障行波测距分析，进行测距新方法的研究。 针对配电网系统的结构，研究了配电网线路模型，分析了配电线路的故障暂态行波传输特性。故障后的行波信号是具有非高斯性、突变性、非平稳性的混合信号，其可以等效为线性数学模型。根据配电线路的行波提取要求，分析了峭度和负熵的独立分量法（FastICA）对非高斯性质的混合信号和故障电流信号分离特点；负熵算法能够可靠分离交流故障暂态电流信号，可用于对配电线路的故障行波特性分析，建立将故障电流行波信号为信号源的FastICA故障测距算法。 采用单通道的故障测距方法可以节约测距成本，在此基础上建立了基于负熵的单通道FastICA最优目标函数。通过对混合电流信号的线性等效，提取含有故障特征的电流行波信号，判断故障电流行波波头极性是否一致并捕获前两个最大波头的时间，利用单端法测距公式计算距离。 利用Matlab/simulink构建了配电系统模型；对不同支路、不同位置故障点和不同接地等运行方式故障测距进行仿真，结果表明单通道FastICA测距方法准确、可靠。

The more rapid development of modern power systems, more complex of distribution network structure. The fault measuring location methods has been unable to meet the requirement of stability and reliability operation of the system. Compared to traditional methods ,traveling wave mode received widespread attention without the impact from system operation mode, transition resistance and other factors, become a hot topic on fault location technology. Therefore, this paper will focus on the use of fast independent component method for distribution network Traveling Wave Fault Location research. According to the distribution network structure of the system, to study the distribution network line model, and analyze the fault transient traveling wave transmission characteristics of distribution lines. Traveling wave signal is a mixed-signal with the nature of non-Gaussian, mutagenic, non-stationary after fault, it can be equivalent to a linear mathematical model. Separated comparison of mixed-signal of non-Gaussian nature and fault current signal through kurtosis and independent component method (FastICA) of negative entropy, AC transient current signal can be reliably separated by negative entropy algorithm. Therefore, this study proposed using of fault current traveling wave signal as the signal source FastICA fault location algorithms. Fault location method using single-channel can save the cost, to establish a single channel FastICA optimal objective function on negative entropy. extraction electric wave signal with features of fault With the rapid development of modern power systems, distribution network structure is more complex. The fault location methods have been unable to meet the stability and reliability of system operation. Compared to traditional methods ,traveling wave mode without using ranging impact system operation, transition resistance and other factors, it received widespread attention, become a hot topic fault location technology. Therefore, this paper will focus on the use of fast independent component method for distribution network Traveling Wave Fault Location research.