Leak diagnosis in gas pipelines based on Symlets wavelet transformation

A novel pipeline leakage diagnosis method based on wireless sensor networks is presented to improve the accuracy of leakage localization, the detecting range and the real-time performance. In order to eliminate noise and the frequency dispersion, the method uses the distributed data aggregation technique combined with the Symlets wavelet transformation to convert original signals to single-mode acoustic emission signals in source nodes, and in sink nodes, it divides the single-mode signals into couples from ordinary nodes. The leak position is obtained by the cross-correlation time differences of arrival (TDOA) localization and the weighted average method. The simulation results show that the localization accuracy can be improved obviously by aggregating the data from multi-nodes. The localization error of the method is less than 5%.