Finite element simulation of guided waves to detect product theft from pipelines
Proceedings Publication Date
The huge economic and environmental costs, as well as the injuries and fatalities associated with pipeline failures are daunting. Product theft (hot tap) and intentional attack (vandalism) are among the major causes of pipeline failure. The existing pipeline monitoring/inspection techniques mainly serve as reactive measures to detect leakage resulting from damage. Guided waves (GWs) are finding more applications for the real time structural health monitoring (SHM) of pipelines and other long, slender structures. GW offers the advantages of long range examination of a structure and rapid detection of damage. As an example stress wave signals generated through physical attack on a pipeline propagates in the form of GW. These signals can be detected by a GW sensor to provide information about the source and location of the interference. Deliberately excited GWs can be used to detect the presence of additional features such as small branch introduced to initiate a product theft. FE analysis is conducted on a 12 in (305 mm) diameter steel pipe with 12 mm wall thickness. A frequency sweep was conducted for fundamental longitudinal L(0,1) and torsional T(0,1) GW modes propagating along a pipeline. It was observed that a low frequency (below 5 kHz) tone burst excitation modified by a Hanning window produces a GW with low attenuation and dispersion. For example, at 2.5 kHz centre frequency, a phase velocity of 5100 m/s and attenuation of 0.00034 m-1 were obtained from the simulation results. At this attenuation rate, the GW signal would theoretically retain more than 10 % of its original energy after propagation through a distance of 8 km. The sensitivity of the signals at this frequency was tested with detection of 2 in (50 mm) branch pipe attached along the 12 in pipeline. Future work will use laboratory tests to validate the FE results.