Flow distribution imaging and sensing for leaks in pipelines using decimated signal diagonalization

Abstract

Control and monitoring of pipelines, dedicated to liquids transportation and distribution, are generally performed by active and passive techniques. With active techniques, we intend the possibility of using external sources to target pipelines and/or liquid inside the infrastructure. The used radiation can be partly reflected by the liquid and pipeline. Passive techniques are related to spontaneous emissions from liquids within the pipelines. However, capturing water displacements and fluctuations within the pipeline, by means of sensors and transducers, is somehow a passive technique. This latter is here illustrated thanks to pressure sensors mounted on an experimental pipeline. The basic idea is to detect leaks but also to monitor water flow regime by means of imaging without using a camera. That is possible thanks to the algorithm here developed and based on 2D/3D Decimated Signal Diagonalization (DSD). It is a technique used in nuclear magnetic resonance, and susceptible to bring to excellent results if water flow within a pipeline is considered as blood flow inside an artery or a vein. The approach has been applied to an experimental hydraulic circuit.