An adaptive method for water pipeline leak localization

DOI

10.2991/wartia-18.2018.24How to use a DOI?

Keywords

Leak localization, neural network, generalized cross correlation, adaptive filter.

Abstract

A common method for leak localization in water pipelines is the use of cross correlation on two measured acoustic signals located on either side of a leak. However, in practice, leak signals are inevitably corrupted with non-leak sounds and noise. This is due to the complicated pipeline environment that makes time delay estimation between the sensors inaccurate. In addition, time delay errors further affect leak localization accuracy, and then reduce the reliability of the system. In this paper, a new adaptive leak detection method that combines BP neural networks with GCC (generalized cross correlation) is proposed. The new proposed method can discriminate the leak signal from non-leak acoustic sources and noise, and improve leak localization accuracy. Simulation results indicates that the proposed leak detection method can be used to localize a leak in a buried water pipeline and achieved a correct detection rate of 92.5%.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Shun Na
AU  - Jinhong Liu
AU  - Qi Li
AU  - Yang Liu
AU  - Yong Tie
PY  - 2018/09
DA  - 2018/09
TI  - An adaptive method for water pipeline leak localization
BT  - Proceedings of the 4th Workshop on Advanced Research and Technology in Industry (WARTIA 2018)
PB  - Atlantis Press
SP  - 146
EP  - 153
SN  - 2352-5401
UR  - https://doi.org/10.2991/wartia-18.2018.24
DO  - 10.2991/wartia-18.2018.24
ID  - Na2018/09
ER  -