The Development of an Intelligent Sensor for the Monitoring of Pipeline System Integrity

Abstract

Abstract This paper presents experimental studies that integrate and develop a prototype sensor to monitor, analyse and communicate any changes in oil/gas/water pipeline performance in such a way that reduces pipeline failures and the associated disruption to supply, damage to the environment and cost to the company. In the oil steel pipeline systems industry alone, the financial implications of pipeline failures have been estimated at £2.5 billion year. Further, current methods of detection rely on periodic inspection which misses more occurrences of failure than it identifies. What is needed is a semi-continuous monitoring system that not only recognises imminent pipeline failure characteristics but also automatically transmits a warning to a central control centre. This paper reports the findings of a laboratory based test programme to evaluate the potential for vibration sound emission detection to form the basis of such an early warning system. Attenuation and waveform analysis were used to identify theoretical changes in the flow characteristics of a liquid medium as it passed along a damaged section of pipe. The effectiveness of piezoelectric sensors to detect the changed flow characteristics were evaluated for a range of artificially induced defects. The flow characteristics were analysed (using spectrum analysis of the signal across a number of frequency bands) and compared to the theoretical predictions. The results show that a piezoelectric vibration sound emission sensor can detect changes in the flow characteristics of water and diesel oil to a level of accuracy that could form the basis of an integrated wireless sensor device. Further qualitative diagnosis theory can form the basis of an algorithm which could be used to associate the changed flow characteristics with underlying pipeline defects without third party intervention.