Title :
On using double power spectral density information for leak detection
Author_Institution :
Appl. Electron. Dept., Politeh. Univ., Timisoara, Romania
Abstract :
This paper studies the possibility of detecting and locating a leak in a pipeline through the power spectral density (PSD) information of the acoustic signals received at some arbitrary, fixed points. For the ideal case of a lossless line, a theoretical formula and a derived pattern are deduced, for the double PSD of the received signals. Following these results an algorithm is proposed, based on matching the sequences of the acquired signals in the installation and the theoretical patterns. An experimental model and an equivalent SimscapeTM simulation model of a pipeline are used for the algorithm verification.
Keywords :
acoustic signal detection; condition monitoring; leak detection; pipelines; PSD information; Simscape simulation; acoustic signals; double power spectral density information; pipeline leak detection; signal acquisition; signal matching; signal sequencing; Acoustics; Mathematical model; Noise; Pattern matching; Pipelines; Signal processing algorithms; Transient analysis; leak detection; pipeline modeling; pipeline monitoring; spectral estimation;
Conference_Titel :
Industrial Technology (ICIT), 2013 IEEE International Conference on
Conference_Location :
Cape Town
Print_ISBN :
978-1-4673-4567-5
Electronic_ISBN :
978-1-4673-4568-2
DOI :
10.1109/ICIT.2013.6505837