Browse by author
Lookup NU author(s): Emeritus Professor Gui Yun Tian
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
© 2025. Large Diameter Pipeline (LDP) serves as an efficient means of transporting liquid or gaseous substances, such as oil or natural gas. However, the presence of environmentally corrosive materials (such as soil) and corrosive impurities (such as H2S) in oil or gas can accelerate the propagation of fatigue defects and corrosive pitting, thereby posing a threat to the integrity of the pipeline. To enable online and quantitative detection of small-scale cracks in the inner wall of LDP, a new passive probe based on the motion-induced eddy current technique is designed. The interactions between the designed probe and various defects in the pipeline are then explored through a series of experiments. Based on these findings, a quantitative detection approach is proposed to classify and size multiple defects on the pipeline using a convolutional neural network. Finally, the performance of the proposed probe and detection approach is verified. The experimental investigations revealed that the relative error remains consistently below 10 % for the multiple cracks in the inner wall of LDP, even when the width of the fatigue defect is smaller than 0.1 mm as well as when the diameter of the corrosive pitting is smaller than 1 mm. The designed probe and the proposed approach are not only capable of online and quantitative evaluation of the different cracks in the inner wall of the LDP through a single measurement but also offer the potential for monitoring the health conditions of other running metal components, such as the high-speed track and bearing.
Author(s): Yu Y, Yang W, Lu Y, Li R, Tian G
Publication type: Article
Publication status: Published
Journal: Measurement
Year: 2026
Volume: 259
Issue: Part B
Print publication date: 01/02/2026
Online publication date: 12/11/2025
Acceptance date: 11/11/2025
ISSN (print): 0263-2241
ISSN (electronic): 1873-412X
Publisher: Elsevier BV
URL: https://doi.org/10.1016/j.measurement.2025.119710
DOI: 10.1016/j.measurement.2025.119710
Altmetrics provided by Altmetric
