摘要
为了解决配电网线路数量巨大、线路结构复杂,故障几率高的问题,利用高压脉冲并基于神经网络来研究配电网多端故障定位系统。文中在待测电缆上输入高压脉冲获得暂态行波特征,分析获得的暂态行波特征实现对架空线和电缆故障的定位,最后利用神经网络进一步分析确定故障的种类,以此进行故障维护和预防的决策制定。针对故障定位测距,采用BP神经网络算法能够有较高的识别精度,误差最大不超过0.4km。针对故障诊断,从上表可以看出测试误差小于0.58,该神经网络能够明显区分三种故障。
In order to solve the problem that the number of distribution network lines is huge,the line structure is complex,and the probability of failure is high. Distribution network multi-terminal fault location system is designed by high voltage pulse and design based on neural network. In this paper,the high-voltage pulse is input on the cable to be tested to obtain transient traveling wave characteristics. Analysis of the obtained transient traveling wave characteristics to achieve the positioning of overhead lines and cable faults. Finally,the neural network is used to further analyze and determine the type of fault,so as to make decision- making for fault maintenance and prevention. For fault location and ranging,BP neural network algorithm can be used to achieve higher recognition accuracy,and the error is no more than 0.4Km. For fault diagnosis,it can be seen from the above table that the test error is less than 0.58,and the neural network can clearly distinguish three types of faults. of the user.
作者
简学军
傅兴强
王秋影
JIAN Xue-jun;FU Xing-qiang;WANG Qiu-ying(Qujing Power Supply Bureau,Yunnan Power Grid Company Limited Liability,Qujing 655000,China)
出处
《电子设计工程》
2019年第14期30-34,共5页
Electronic Design Engineering
关键词
架空线电缆
故障定位
神经网络
识别精度
overhead cable
fault location
neural networks
recognition accuracy