Spectral based pipeline leak detection using a single spatial measurement. (December 2021)

• Record Type: Journal Article
• Authors: Keramat, Alireza; Duan, Huan-Feng
• Is Part Of: Mechanical systems and signal processing. Volume 161(2021)
• Publication Date: 2021-12
• Subjects: Leak detection -- Water hammer -- Signal processing -- Maximum likelihood estimation -- Matched field processing; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2021.107940
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A new Matched-Field Processing (MFP) technique for leak detection is developed. The new method can localize leaks by domain enumeration. The new method is more robust to noisy measurements than the conventional one. The conventional and the proposed MFP model are closely investigated. Several leak detection problems including two laboratory tests are discussed. Abstract: This study offers a vast improvement to the established Matched Field Processing (MFP) method for leak identification in terms of its versatility to pinpoint leaks with a single spatial measurement. All the previous schemes require at least two measurement signals of pressure head as well as transient signal of the upstream flow rate so as to render a solution to the leak identification problem. This study reformulates the one-dimensional identification procedure so that localization can be accomplished using a single spatial measurement. The numerical results convincingly demonstrate that the new method outweighs the conventional one as it provides a smoother objective function and hence robust to identify leaks using observations of greater noise levels. Checked against two recent laboratory experiments, the proposed method gave satisfactory localization to both of which. The new scheme can provide an improved (both in terms of efficiency and accuracy) initial estimate to the multidimensional optimizations required to localize multiple leaks.