While it is difficult to detect target signals, on-line monitoring remains the most reliable method to protect underground pipelines from the third party damage. To distinguish signals from a variety of white noises during remote pipelines monitoring, it is necessary to filter signals from unanticipated noise and to find co-relation among signals. In this paper, an on-line monitoring system based on local area network (LAN) is developed not only to detect a damage occurrence but also to find its location along the pipelines. For this purpose, an acoustic wave propagation model is generated along the gas pipelines, and signals based on the model are compared with experimental data. In modeling the wave propagation, auto power spectral density (APSD) and cross power spectral density (CPSD) processing methods are applied to distinguish damage signals from noises. To locate the damage point on the pipelines, signals from three different piezoelectric accelerometers are monitored, and compared with those from the proposed model. From the experiments, the number of sensors should be determined in accordance with the attenuation coefficient of signal. Also, the proposed scheme which combines APSD and CPSD processing methods is proved to be powerful in practical applications. In signal analyses, the power spectrum density assessment, especially cross method, is useful with satisfactory results. As a consequence of adopting such methods, most of impact signals are detected with dominant frequencies that are focused to 200-500 Hz simulating breakers and hammer drills which are third party damage sources. For most low frequency excavation machines (except those such as drills which are being operated at relatively high frequency band), the power spectrum density detection method is proved to be useful and powerful in practical situations.
Protection of underground gas pipelines from the third party damage by on-line monitoring using piezoelectric accelerometer
Schutz von unterirdischen Gasrohrleitungen gegen Haftpflichtschäden durch Online-Überwachung mit Hilfe piezoelektrischer Beschleunigungsmesser
2005
9 Seiten, 10 Bilder, 1 Tabelle, 16 Quellen
Aufsatz (Konferenz)
Englisch