Transient-based leak detection in the frequency domain considering fluid–structure interaction and viscoelasticity. (15th May 2021)

• Record Type: Journal Article
• Authors: Keramat, Alireza; Karney, Bryan; Ghidaoui, Mohamed S.; Wang, Xun
• Is Part Of: Mechanical systems and signal processing. Volume 153(2021)
• Publication Date: 2021-05-15
• Subjects: Leak detection -- Waterhammer -- Fluid-structure interaction -- Maximum likelihood estimation -- Viscoelasticity; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2020.107500
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: Transients including FSI and pipe-wall viscoelasticity in a leaky pipe is simulated. The contribution of each item i.e. FSI, VE or leak on frequency spectra is presented. The Matched Field Processing technique based on the new model is implemented. Several cases illustrating the increased accuracy due to the FSI model are discussed. Abstract: Waterhammer induced fluid–structure interaction (FSI) of a leaky viscoelastic (VE) pipe is simulated in the frequency-domain. The developed model is then applied to assess the accuracy of leak detection when both FSI and VE phenomena are represented. FSI arises due to the coupling of wall stress and pressure waves, which can significantly distort the response spectra. The frequency response of the transient waves is derived using the transfer matrix method, which is then exploited for localization using a maximum likelihood estimate. Several numerical cases including one or two leaks are studied while Poisson and junction coupling both with and without VE wall behaviour. The results demonstrate that including FSI effects can allow leaks to be more accurately located. Not surprisingly, the importance of FSI mechanisms in leak detection is shown to be more crucial in VE pipes than in elastic pipes because of their larger Poisson effects.