Pipeline Leak Detection Based On Fiber Optic Early-Warning System

With the rapid development of our economy, the demand for energy is increasing. Pipeline transportation is playing an important role in eco-nomic and social development. However, because of its long distance, al-ong the way, the environment is mostly desert, swamp and other bad host-ile regions, so it is difficult to execute manual inspection. At the same ti-me, pipeline leaks occur due to natural causes including the aging of long transport, erosion and human factors such as steal. These can cause subs-tantial economic loss, and even environmental pollution. Therefore, it is a major problem in pipeline transportation field. Energy consumption in C-hina is on an increasing trends, the 12th five-year plan will further streng-then the construction of pipeline transportation. Together with pipeline tr-ansportation, security is being taken seriously. Only safe transportation c-an take pipeline transportation to a new level.This topic came from the pipeline transportation safety system requ-irements, implemented the optical early warning system. The system can monitor along the pipeline for real-time. When the stolen event happens, the system will alarm and locate the position at the first time to prevent t- he occurrence of pipeline leakage. This thesis followed the functional re-quirements of the system, completed the software and hardware design i-ncluding:building optical system based on Mach-Zehnder, laser circuit d-esign, optical receiver circuit design, debugging, data acquisition based on PCI1714UL, location and simulation. The features are in line with the design requirements under test.The system was tested with a 500m long pipeline, which is mined 0.5m deep. When the output power is 2mW and wavelength is 1550nm, the function of alarm and positioning is completed and the positioning ac-curacy is less than 25 meters. Recognition method based on Hmm is used to classify the danger signal. Simulation results on the collected signal sh-ow that this method can reduce the false alarm and improve system perfo-rmance.