|本期目录/Table of Contents|

 基于EEMD和二次相关法的管道泄漏定位检测(PDF)

《纳米技术与精密工程》[ISSN:1672-6030/CN:12-1351/O3]

期数:
2017年5期
页码:
372-377
栏目:
精密测量
出版日期:
2017-09-15

文章信息/Info

Title:
 Pipeline Leak Location Detection Based on EEMD and Second Correlation
文章编号:
1672-6030(2017)05-0372-06
作者:
 李健 封超
 (天津大学精密仪器与光电子工程学院,天津300072)
Author(s):
 Li Jian Feng Chao
 (School of Precision Instrument and Opto-Electronics Engineering, Tianjin University, Tianjin 300072, China)
关键词:
 管道泄漏 总体平均经验模态分解 泄漏定位 二次相关
Keywords:
 pipeline leak ensemble empirical mode decomposition leak location second correlation
分类号:
TE832
DOI:
10.13494/j.npe.20150118
文献标识码:
A
摘要:
 针对当前管道泄漏信号噪声大、定位误差大的问题,提出一种提高定位精度的新方法.泄漏信号经过总体平均经验模态分解(EEMD)之后,可以得到不同尺度的固有模态函数 (IMF) 分量,这些分量与原信号的相关系数可以作为信号重构的主要依据.这种自适应的降噪方法,不仅提高了重构信号的信噪比,还有效去除了模态混叠的干扰.然后,利用二次相关运算对两路降噪后的泄漏信号进行延时估计,再根据泄漏定位模型计算泄漏位置.最后,采用直接相关方法、基于经验模态分解(EMD)的相关方法以及本文提出的EEMD相关数据处理方法,分别对同组实验数据进行处理,对比定位误差.实验结果表明,EEMD相关方法相比前两种方法,有效抑制了模态混叠,提高了定位精度.
Abstract:
 Regarding the big noise and large location error of pipeline leak signal, a new method is proposed to improve the location accuracy of pipeline leak detection in this paper. First, the signal was decomposed by ensemble empirical mode decomposition (EEMD) and then several intrinsic mode function (IMF) components were obtained. The correlation coefficients between the signal and the IMF components provided the basis for reconstructing the signal. This adaptive method suppressed the phenomenon of mode mixing and improved the signal-to-noise ratio of the reconstructed signal. Then, second correlation was adopted to estimate time delay of two reconstructed leak signals. Combining the wave velocity and pipeline length, the leak location was calculated. Finally, experiment was conducted to demonstrate this method. Compared with direct correlation and the correlation based on empirical mode decomposition (EMD), the method proposed in this paper achieves a higher accuracy.

参考文献/References

备注/Memo

备注/Memo:
收稿日期: 2017-01-20.
作者简介: 李健(1973—),男,博士,副教授.
通讯作者: 封超,feng1chao@126.com.
更新日期/Last Update: 2017-09-27