Pipeline Leak Detection Using the Matched-Field Processing Method

A transient-based leakage detection method using a matched-field processing (MFP) scheme is formulated and applied. It is found to be efficient, robust, accurate, and provides a unique solution. Its inherent efficiency stems from the fact that the MFP method decouples the search for the leak location and the leak size. Its robustness is demonstrated by showing that the MFP method is able to estimate the location and size of leak in the presence of (1) noise even for signal-to-noise ratio (SNR) as low as SNR=−3 dB, and (2) uncertainty in the fluid–pipe system wave speed. Another positive attribute of the MFP method is that it is able to use all available frequencies, rather than just resonant frequencies, and does not need to identify resonant frequencies. As a result, the MFP method provides precise localization estimates even in noisy environments. It is also shown that a unique identification is achieved by the MFP method provided that the pressure signal is measured at two locations along the pipe. For the case of multiple leaks, MFP identifies each of the leaks provided that the distance between the leaks is of the same order or larger than the shortest probing half-wavelength.