Author(s): Chen Qiu-Ru; Zhang Jiangjiang; Duan Huan-Feng; Che Tong-Chuan
Linked Author(s): Huan-Feng Duan
Keywords: Leak localization Water pipes Data assimilation Transient waves Super-resolution
Abstract: Unknown leaks in urban water supply pipes have resulted in significant non-revenue water. Inverse transient analysis (ITA) is a classical and popular transient wave-based leak localization method. However, it suffers from low computational efficiency and is unable to quantify uncertainty. To overcome these challenges, this study proposes a transient wave-based data assimilation (DA) framework for pipe leak localization. The proposed framework is compared with a classical ITA method for multiple leak localization in single water pipes. It demonstrates that the proposed DA framework is more accurate and efficient than the classical ITA method. Moreover, one local updating-based DA algorithm can super-resolve two close leaks spaced smaller than a minimum half-wavelength. This validates the strength of the proposed transient wave-based DA framework for leak localization and shows its potential for identifying other pipe anomalies.
Year: 2025