On using a simplified model for leak detection improving in fluid filled pipelines

Abstract

A major difficulty of the acoustic techniques for pipeline leak detection and location is the small signal-to-noise ratio (SNR) of the received data, due to a wide range of perturbing factors. In order to choose the optimum method is necessary a good understanding of the given system. This paper uses a simplified pipeline model, analog with a transmission line, as a tool for simulating the acoustic signals, in order to study and develop an improved method for leak locating in a multipath and a correlated noise environment. It is shown that the results obtained by averaging the output of different algorithms, parallel implemented, lead to an improved overall robust behavior. Also, the effect of the pipeline end opening is studied. An improved accuracy is found when the opening is brought near to a matched output load impedance.