A pipeline leak detection method based on wavelet packet and BP neural network

Author

Guanhua, Chen ; Jianwei, Li ; Zongjian, Zhang ; Jian, Guan

Author_Institution

Sch. of Mech. & Electron. Control Eng., Beijing Jiaotong Univ., Beijing, China

fYear

2010

fDate

26-28 June 2010

Firstpage

5822

Lastpage

5825

Abstract

According to the problems of pipeline leak detection existed in oil transport, a testing and positioning method based on wavelet packet and BP neural network is proposed. A Db4 wavelet packet is used to filter the noises of pressure signals, and extract its energy signals as the input vectors to train the BP neural network, which can be used to identify the pipeline´s working state and accurately locate the leak points of the pipe. The simulation experiment with water pipe shows that this method is accurate, reliable, and has strong adaptability.

Keywords

backpropagation; mechanical engineering computing; neural nets; pipes; wavelet transforms; BP neural network; oil transport; pipeline leak detection; pressure signal; wavelet packet; Control engineering; Electronic equipment testing; Filters; Leak detection; Neural networks; Petroleum; Pipelines; Signal processing; Wavelet analysis; Wavelet packets; BP neural network; leak detection; oil pipeline; wavelet packet;

fLanguage

English

Publisher

ieee

Conference_Titel

Mechanic Automation and Control Engineering (MACE), 2010 International Conference on

Conference_Location

Wuhan

Print_ISBN

978-1-4244-7737-1

Type

conf

DOI

10.1109/MACE.2010.5535353

Filename

5535353