Oil Pipeline Safety Monitoring Method based on Vibration Signal Analysis and Recognition

Author

Yan, Hu ; Shi, Guangshun ; Hao, Shangqing ; Wang, Qingren

Author_Institution

Inst. of Machine Intell., Nankai Univ., Tianjin, China

Volume

3

fYear

2009

fDate

19-21 May 2009

Firstpage

200

Lastpage

206

Abstract

An analytically tractable model is presented to describe oil and gas pipeline safety monitoring system. The basic idea is that it uses fiber sensor to collect signals produced by soil vibration around pipelines, and then focuses on intelligent processing and smart recognition of soil vibration signals. Finally, we implement the developed model and it is practically used in Jiangsu, China. Experiment results show that the system can fully satisfy the real time requirement. Further more, the alarm rate is higher than 98% and the recognition rate is 95.3% for five different kinds of human activities (ramming, picking, drilling, steel pipe knocking, forklift working), much better than other results reported yet.

Keywords

learning (artificial intelligence); monitoring; neural nets; oil technology; pipelines; safety; signal classification; Jiangsu China; artificial neural network; fiber sensor; gas pipeline safety monitoring system; incremental learning; intelligent processing; oil pipeline safety monitoring method; signal classification; signal recognition; soil vibration; vibration signal analysis; Intelligent sensors; Monitoring; Optical fiber sensors; Petroleum; Pipelines; Real time systems; Safety; Signal analysis; Signal processing; Soil; Safety of oil and gas pipelines; artificial neural network; fiber sensor; incremental learning; soil vibration signals;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems, 2009. GCIS '09. WRI Global Congress on

Conference_Location

Xiamen

Print_ISBN

978-0-7695-3571-5

Type

conf

DOI

10.1109/GCIS.2009.353

Filename

5209171