A model based approach for pipeline monitoring and leak locating

Abstract

A fluid flowing through a pipeline is a complex phenomenon, depending on many physical factors. Developing efficient methods for system monitoring and leak locating start with a good understanding of the pipeline. The modeling technique becomes a very useful tool, allowing a separate analysis of the different hypostases describing the process. This paper investigates the problem of using a reduced lumped parameters model for the pipeline system. Both static and dynamic regimes are discussed. The final goal is oriented on testing the acoustic methods on the resulted model, given the fact that the pipeline is a multipath, noise-correlated environment. An equivalent laboratory experimental model is used for comparison.