Machine Learning-Based Non-Destructive Evaluation of HDPE Lining Integrity from Ultrasonic Signal Features.
e-Journal of Nondestructive Testing · 2026-07 · 31 (7)
DOI: 10.58286/33256 ↗Kurz gefasst
Fachartikel in der Zeitschrift e-Journal of Nondestructive Testing, 2026-07. Verfasst von Thiago t. m. Neves und 7 weiteren Autor:innen.
Abstract (Originalsprache)
A recurring issue in the oil industry is internal corrosion within oil, water, and gas flow lines. In the offshore sector, one strategy to reduce maintenance costs is the use of carbon steel pipes internally coated with HDPE. However, dissolved gases such as CO₂, H₂S, and some hydrocarbons can permeate the coating, causing disbondment and liner displacement. This study develops an algorithm to classify coating integrity based on A-Scan signals from a non-destructive ultrasonic inspection tool. Signal analysis revealed variations linked to integrity conditions, and the proposed approach helps prevent unplanned downtime and reduce maintenance costs by enabling the timely replacement of damaged sections.
Zitieren
APA
Neves, T. T. M., Grosso, M., Siqueira, N. C. D., Althoff, H., Braga, D., Casacão, J., Camerini, C., & Pereira, G. R. (2026). Machine Learning-Based Non-Destructive Evaluation of HDPE Lining Integrity from Ultrasonic Signal Features. e-Journal of Nondestructive Testing, 31(7). https://doi.org/10.58286/33256
Vancouver
Neves TTM, Grosso M, Siqueira NCD, Althoff H, Braga D, Casacão J, et al. Machine Learning-Based Non-Destructive Evaluation of HDPE Lining Integrity from Ultrasonic Signal Features. e-Journal of Nondestructive Testing. 2026;31(7). doi:10.58286/33256
BibTeX
@article{Neves2026,
author = {Neves, Thiago t. m. and Grosso, Marcella and Siqueira, Natalie Chaves de and Althoff, Heloisa and Braga, Daniel and Casacão, João and Camerini, Cesar and Pereira, Gabriela Ribeiro},
title = {Machine Learning-Based Non-Destructive Evaluation of HDPE Lining Integrity from Ultrasonic Signal Features},
journal = {e-Journal of Nondestructive Testing},
year = {2026},
volume = {31},
number = {7},
doi = {10.58286/33256}
}
EUVIMED