Leak detection in pipelines using Hilbert transform

Water leakage has been a subject of increasing concern in recent decades due to the significant economic and environmental cost. Time delay estimation (TDE) plays a key role in the detection and location of water leaks using acoustic signals. A variety of TDE algorithms, such as cross-correlation based time delay estimators, have been proposed and implemented. This paper focuses on the use of Hilbert transform of pressure signals in combination with the cross-correlation process. The new analysis of the proposed method enables better understanding of the detectability of the leak by pressure measurements in the presence of interfering noise. Its performance in relation to leak detection is further compared to the well-established cross-correlation methods for TDE in detecting a leak from an actual PVC water pipe.