EUVIMEDEUVIMED entdecken
Crossref

Magnetic flux leakage defect size estimation method based on physics-informed neural network

Yi Xiong, Shuai Liu, Litao Hou, Taotao Zhou iD

Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · 08.01.2024 · 382 (2264)

DOI: 10.1098/rsta.2022.0387 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, online erschienen am 20.11.2023. Verfasst von Yi Xiong und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Safety and Ocean Engineering, China University of Petroleum-Beijing, Beijing, People's Republic of China. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 30 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Young Elite Scientist Sponsorship Program by Beijing Association for Science and Technology, Science Foundation of China University of Petroleum, Beijing.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Magnetic flux leakage (MFL) is a magnetic method of non-destructive testing for in-pipe defect detection and sizing. Despite the fact that recent developments in machine learning have revolutionized disciplines like MFL defect size estimation, the most current methods for quantifying pipeline defects are primarily data-driven, which may violate the underlying physical knowledge. This paper proposes a physics-informed neural network-based method for MFL defect size estimation. The training process of neural network is guided by the MFL data and the physical constraints that is mathematically represented by the magnetic dipole model. We use synthetic MFL data produced by a virtual MFL testing of pipeline defects to validate the proposed method through a comparison to purely data-driven neural networks and support vector machines. The findings imply that the physics-informed strategy can both improve predictive accuracy and mitigate physical violations in MFL testing, providing us with a better knowledge of how neural networks perform in defect size estimation. This article is part of the theme issue ‘Physics-informed machine learning and its structural integrity applications (Part 2)’.

Schlagwörter

Stichwörter der Autor:innen

defect sizein-pipe inspectionmagnetic dipole modelmagnetic flux leakagephysics-informed neural network

Autor:innen und Institutionen

  • College of Safety and Ocean Engineering, China University of Petroleum-Beijing, Beijing, People's Republic of China.
  • Key Laboratory of Oil and Gas Safety and Emergency Technology, Ministry of Emergency Management, Beijing, People's Republic of China.
  • Dushanzi Oil and Gas Branch, West Pipeline, PipeChina, Karamay, People's Republic of China.

ORCID-Profile: Taotao Zhou

Förderung

  • National Natural Science Foundation of China
  • Young Elite Scientist Sponsorship Program by Beijing Association for Science and Technology
  • Science Foundation of China University of Petroleum, Beijing

Zitieren

APA

Xiong, Y., Liu, S., Hou, L., & Zhou, T. (2024). Magnetic flux leakage defect size estimation method based on physics-informed neural network. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 382(2264). https://doi.org/10.1098/rsta.2022.0387

Vancouver

Xiong Y, Liu S, Hou L, Zhou T. Magnetic flux leakage defect size estimation method based on physics-informed neural network. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2024;382(2264). doi:10.1098/rsta.2022.0387

BibTeX
@article{Xiong2024,
  author = {Xiong, Yi and Liu, Shuai and Hou, Litao and Zhou, Taotao},
  title = {Magnetic flux leakage defect size estimation method based on physics-informed neural network},
  journal = {Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences},
  year = {2024},
  volume = {382},
  number = {2264},
  doi = {10.1098/rsta.2022.0387}
}
Kostenlos starten →