[1]单锐,代海波,刘文..基于小波和神经网络模型的邮电业务总量预测[J].郑州大学学报(工学版),2013,34(03):94-97,120.[doi:10.3969/j.issn.1671-6833.2013.03.023]
　LIAO Xiao-hui,LIANC Heng-na,DING Qian.Research of Power Cable Fault Location Based on Wavelet Transform[J].Journal of Zhengzhou University (Engineering Science),2013,34(03):94-97,120.[doi:10.3969/j.issn.1671-6833.2013.03.023]
点击复制

基于小波和神经网络模型的邮电业务总量预测()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
34卷
期数:
2013年03期
页码:
94-97,120
栏目:
出版日期:
2013-05-31

文章信息/Info

Title:
Research of Power Cable Fault Location Based on Wavelet Transform
作者:
单锐; 代海波; 刘文.
燕山大学理学院,河北秦皇岛,066004, 燕山大学理学院,河北秦皇岛,066004, 燕山大学理学院,河北秦皇岛,066004
Author(s):
LIAO Xiao-hui; LIANC Heng-na; DING Qian
School of Electrical Engineering,Zhengzhou University,Zhengzhou 450001,China
关键词:
BP神经网络模型;  极大重叠离散小波变换;  小波神经网络;  邮电业务总量
Keywords:
power cable;  wavelet transform ;  modulus maximam; traveling wave fault location
分类号:
TM743
DOI:
10.3969/j.issn.1671-6833.2013.03.023
文献标志码:
A
摘要:
引入极大重叠离散小波变换的概念,利用极大重叠离散小波变换的多分辨分析特性,对邮电业务总量序列进行分解.然后对分离得到的光滑项和细节项两部分利用小波神经网络模型进行建模和预测,最后再重构得到邮电业务总量序列的预测值.数据测试结果表明:本文方法可实现多步预测,且对邮电业务总量的预测精度比单纯的用小波神经网络模型或BP神经网络模型高.
Abstract:
With wide application of power cable in transmission and distribution systems, the demand for cablefault location accuracy is improved. In order to locate cable fault accurately,the wavelet transform is applied indetecting the traveling wave signal of power cable. According to the prineiple of singularity detection of sig-nals,the traveling wave starting pulse and reflection pulse time point is determined by searching modulus max-imum,and then locating by single terminal traveling wave method for the online fault location.The experimen-tal results show that the traveling wave singularity point can be detected by using localized time - frequencycharacteristic of wavelet transform,thereby the accurate time of pulse arrival is obtained. This method is notaffected by fault type and the range error is small,which can achieve a higher fault location accuracy.
更新日期/Last Update: 1900-01-01