Pipe system diagnosis and leak detection by unsteady-state tests. 2. Wavelet analysis

✍ Scribed by Marco Ferrante; Bruno Brunone

• Publisher: Elsevier Science
• Year: 2003
• Tongue: English
• Weight: 996 KB
• Volume: 26
• Category: Article
• ISSN: 0309-1708
• DOI: 10.1016/s0309-1708(02)00102-1

✦ Synopsis

As shown in Part 1, the analysis of a pressure signal in the frequency domain can give items of information for leak detection even though, moving from time to frequency domain, interesting information about the arrival time of pressure waves are lost. To overcome this limitation and to widen the applicability of transient-test based methodologies for leak detection, the wavelet transform of laboratory experimental data is used. Such data consist of pressure time histories acquired at one measurement section during transients in a single pipe system. The wavelet transform is used to detect local singularities in the pressure time history due to the presence of a leak. The discontinuity occurrence in time reveals the arrival time of the leak reflected pressure wave and is the basis for leak location.