3D modeling of large elongated structures for non-destructive testing and inspection

Authors: Hesabi, Somayeh
Advisor: Laurendeau, Denis
Abstract: According to a Central Intelligence Agency (CIA) report ¹ presented in a flagship NDT journal ², there were a total of 3.3 million km of pipelines present in 120 countries in the world in 2014. This means that pipelines play an important role in the energy infrastructure in order to safely transport liquid or natural gas. Although pipelines are the most efficient and reliable way to carry various liquids ranging from water to oil, they are vulnerable to external and internal damages. Fortunately, a periodic inspection of pipelines can increase their functionality and decrease the environmental disasters as well as economic losses caused by potential spills, explosions or other malfunctions. In this context of the exploitation of pipelines and other similar elongated structures and considering the benefits of 3D sensors which allow us to create an accurate digital replica of the surface of physical objects in addition to the advantages of Non-Destructive Testing (NDT) technology which provides the ability of under-surface monitoring, our research proposes a solution to build a 3D model of pipeline or other elongated structures to monitor their status. For this purpose, we first measure the geometry of the pipeline by handheld 3D scanners and construct the 3D model of the structure. Then, the information of subsurface defects that is estimated efficiently by approaches developed by other team members using infrared thermography is integrated to the reconstructed 3D model. The manuscript investigates different challenges related to the 3D modeling of large elongated structures with high accuracy and repeatability for quality control purposes as well as for long-term maintenance. 1. The World Factbook, updated 18 May 2015. 2. Materials Evaluation (M.E.), vol. 73, no. 7, July 2015
Document Type: Thèse de doctorat
Issue Date: 2017
Open Access Date: 24 April 2018
Permalink: http://hdl.handle.net/20.500.11794/27640
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
33142.pdf50.45 MBAdobe PDFView/Open
All documents in CorpusUL are protected by Copyright Act of Canada.