Title :
TDR-based pipe leakage detection and location using Bayesian inference
Author :
Sihyeong Woo, Taejin Kim ; Youn, Byeng D. ; Huh, Young Cheol
Author_Institution :
Department of Mechanical and Aerospace Engineering Seoul National University Seoul 151-742, Korea
Abstract :
TDR is used to detect and locate the water leakages at pipe joint. A wire attached on the pipe surface is soaked by water when the leakage occurs and it affects the characteristic impedance of the wet part, leading to the change of the reflected signal. To infer the leakage from the TDR signal, we used the finite difference time domain based forward model which provides output of the TDR signal given the configuration of the transmission line. Then by solving the inverse problem using Bayesian inference, the locations and the number of leakages are found. The suggested method is validated for multiple leakages which has complex signal to interpret.
Keywords :
Bayes methods; Detectors; Impedance; Power cables; Transmission line measurements; Wires; Bayesian inference; Leakage detection; Pipe; Time domain reflectrometry (TDR);
Conference_Titel :
Prognostics and Health Management (PHM), 2015 IEEE Conference on
Conference_Location :
Austin, TX, USA
DOI :
10.1109/ICPHM.2015.7245056