Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/109460
Title: | Laboratory validation of in-pipe pulsed thermography in the rapid assessment of external pipe wall thinning in buried metallic utilities | Authors: | Yu, S Chung, WWS Lau, TCW Lai, WWL Sham, JFC Ho, CY |
Issue Date: | Apr-2023 | Source: | NDT and E international, Apr. 2023, v. 135, 102791 | Abstract: | This study characterized the in-pipe thermal signature of external pipewall thinning in steel pipes, a common problem that is caused by external corrosion in hostile underground environment. A model system was prepared to imitate the underground environment by milling several holes of various sizes and residual thicknesses into a mild steel plate. Wall thinning was investigated using active infrared thermography. The non-defective side of the steel plate was heated to 27.4 °C through the application of a thermal energy pulse while the ambient temperature was 22°C. Thermograms were captured inside the pipe at a frequency of 0.02 seconds for 5 min. The images of the thinned surface were processed in two steps. First, the peak contrast time algorithm was used to estimate the residual thickness. Second, Gaussian adaptive thresholding was used to estimate the size of the holes. The maximum observable defects had a diameter of 5 mm and a residual thickness of 3 mm. The type of defect interface (steel–sand or steel–air) had no significant effect on the estimation of residual thickness or size. This study developed a rapid approach in classifying defect's residual thickness by only utilizing two well-known parameters from infrared images – defect's peak thermal contrast and estimated area. Thus, the feasibility of non-destructive, in-pipe, quantitative IR thermographic analysis of buried metal pipelines is demonstrated. | Keywords: | Adaptive thresholding Infrared thermography Peak contrast time Peak thermal contrast Residual thickness estimation Size estimation |
Publisher: | Elsevier Ltd | Journal: | NDT and E international | ISSN: | 0963-8695 | EISSN: | 1879-1174 | DOI: | 10.1016/j.ndteint.2023.102791 |
Appears in Collections: | Journal/Magazine Article |
Show full item record
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.