High definition electrical capacitance tomography for pipeline inspection

M. Evangelidis, Lu Ma, M. Soleimani

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)

Abstract

Pipelines made of dielectric materials such as Polyethylene (PE) are becoming increasingly popular. With no suitable inspection technique for dielectric pipes, there is an urgent need to develop new technology for their inspection. This paper presents a novel pipe inspection technique using Electrical Capacitance Tomography (ECT) imaging. Traditionally ECT is used for industrial process tomography as a low resolution but fast tomographic imaging technique. Typically commercial ECT can provide a resolution of approximately 10 percent of the imaging region. In this paper a limited region tomography technique is developed take into account prior knowledge about the geometry of the pipe. This has significantly enhanced the imaging resolution of the ECT system, making it a viable pipe inspection solution. The experimental results in this study demonstrate an interior wall loss area as small as 0.195 percent of the ECT cross sectional imaging region is repeatable and can be reliably detected. A narrowband pass filter method (NPFM) is used as a means to limit the region for the ECT algorithm. This results in an unprecedented resolution, making ECT a viable non-destructive evaluation (NDE) technique for plastic pipes. The NDE application of the ECT for PE pipes is demonstrated in this paper with several experimental results. A wall loss of depth of 1.5mm could be detected for an ECT sensor array of 150mm in diameter, showing a high resolution and high definition ECT (HD-ECT) imaging that has not been reported before
Original languageEnglish
Pages (from-to)1-15
Number of pages15
JournalProgress In Electromagnetics Research (PIER)
Volume141
Publication statusPublished - 2013

Fingerprint Dive into the research topics of 'High definition electrical capacitance tomography for pipeline inspection'. Together they form a unique fingerprint.

Cite this