Skip to main content

Ultrasound classification of interacting flaws using finite element simulations and convolutional neural network

Engineering with Computers (2022)Cite this article

Abstract

Interacting flaws refer to conditions when two flaws are in close proximity and have the potential to interact with each other to significantly reduce the integrity of a structure. Accurate detection and classification of non-visible interacting and single flaws using ultrasound time signals continue to be a significant challenge. Machine learning-based flaw detection and classification systems are promising, but have been unable to be implemented as they lack training data that are expected to be obtained from a large set of well-labeled field data or experiments. Cracks and corrosion wall loss are two flaw types of primary concern in metallic structures, and are the focus of this study. We present an approach that utilizes finite element simulation data to train an ultrasound time signal-based convolutional neural network (CNN). No flaw, single crack, single wall loss corrosion, two cracks, and combined crack with corrosion are the five categories comprising single and interacting flaws considered in this work. A dataset containing 2000 numerical ultrasound NDT signals created through finite element simulations was used to train an optimal CNN architecture. A validation study was conducted using 13 3D metal printed steel specimens containing a variety of interacting and single flaws. Twenty-five measurements considering precise and offset transducer placements were used for the validation study. The simulation-trained CNN showed 100% accuracy in classifying all categories of flaws from the independent experimental ultrasound NDT signals. The results are promising as the classification of non-visible interacting flaws that has traditionally been a very difficult problem could be addressed using the methodology presented here.

This is a preview of subscription content, access via your institution.

Access options

Buy single article

Instant access to the full article PDF.

USD 39.95

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Notes

  1. This is not a strict limit but an assumption in our studies. We find that the neural network classification accuracy can stay high even for larger offset signals.

References

  1. Okodi A, Li Y, Cheng J, Kainat M, Yoosef-Ghodsi N, Adeeb S (2021) Effect of location of crack in dent on burst pressure of pipeline with combined dent and crack defects. J Pipeline Sci Eng 1(2):252–263. https://doi.org/10.1016/j.jpse.2021.05.003

    Article  Google Scholar 

  2. Benjamin AC, Freire JLF, Vieira RD, Cunha DJ (2016) Interaction of corrosion defects in pipelines - Part 1: Fundamentals. Int J Press Vessels Pip 144:56–62. https://doi.org/10.1016/j.ijpvp.2016.05.007

    Article  Google Scholar 

  3. Coules HE (2018) On predicting the interaction of crack-like defects in ductile fracture. Int J Press Vessels Pip 162(February):98–101. https://doi.org/10.1016/j.ijpvp.2018.03.006

    Article  Google Scholar 

  4. Quickel GT, Beavers JA (2016) Pipeline failures resulting from interacting integrity threats. Proc Bienn Int Pipeline Conf IPC 1:1–15. https://doi.org/10.1115/IPC2016-64436

    Article  Google Scholar 

  5. Xie M, Wang Y, Xiong W, Zhao J, Pei X (2022) A crack propagation method for pipelines with interacting corrosion and crack defects. Sensors. https://doi.org/10.3390/s22030986

    Article  Google Scholar 

  6. Ariffin MZ, Zhang YM, Xiao ZM (2017) Elastic-plastic fracture response of multiple 3-D interacting cracks in offshore pipelines subjected to large plastic strains. Eng Fail Anal 76:61–79. https://doi.org/10.1016/j.engfailanal.2017.02.003

    Article  Google Scholar 

  7. Hasegawa K, Saito K, Iwamatsu F, Miyazaki K (2009) Prediction of fully plastic collapse stresses for pipes with two circumferential flaws. J Press Vessel Technol Trans ASME 131(2):1–6. https://doi.org/10.1115/1.3066967

    Article  Google Scholar 

  8. Yao Y, Tung S-TE, Glisic B (2014) Crack detection and characterization techniques: an overview. Struct Control Health Monit 21(12):1387–1413. https://doi.org/10.1002/stc.1655

    Article  Google Scholar 

  9. Carvalho AA, Rebello JM, Souza MP, Sagrilo LV, Soares SD (2008) Reliability of non-destructive test techniques in the inspection of pipelines used in the oil industry. Int J Press Vessels Pip 85(11):745–751. https://doi.org/10.1016/j.ijpvp.2008.05.001

    Article  Google Scholar 

  10. Drinkwater BW, Wilcox PD (2006) Ultrasonic arrays for non-destructive evaluation: a review. NDT and E Int 39(7):525–541. https://doi.org/10.1016/j.ndteint.2006.03.006

    Article  Google Scholar 

  11. Zhang J, Drinkwater BW, Wilcox PD, Hunter AJ (2010) Defect detection using ultrasonic arrays: the multi-mode total focusing method. NDT and E Int 43(2):123–133. https://doi.org/10.1016/j.ndteint.2009.10.001

    Article  Google Scholar 

  12. Zhang E, Dao M, Karniadakis GE, Suresh S (2022) Analyses of internal structures and defects in materials using physics-informed neural networks. Sci Adv. https://doi.org/10.1126/sciadv.abk0644

    Article  Google Scholar 

  13. Jin H, Jiao T, Clifton RJ, Kim K-S (2022) Dynamic fracture of a bicontinuously nanostructured copolymer: a deep-learning analysis of big-data-generating experiment. J Mech Phys Solids 164:104898. https://doi.org/10.1016/j.jmps.2022.104898

    Article  Google Scholar 

  14. Duan J, Asteris PG, Nguyen H, Bui X-N, Moayedi H (2021) Novel artificial intelligence technique to predict compressive strength of recycled aggregate concrete using ica-xgboost model. Eng Comput 37:3329–3346. https://doi.org/10.1007/s00366-020-01003-0

    Article  Google Scholar 

  15. Yin M, Ban E, Rego BV, Zhang E, Cavinato C, Humphrey JD, Em Karniadakis G (2022) Simulating progressive intramural damage leading to aortic dissection using DeepONet: an operator-regression neural network. J R Soc Interface 19:187. https://doi.org/10.1098/rsif.2021.0670

    Article  Google Scholar 

  16. Mishra M, Bhatia AS, Maity D (2021) A comparative study of regression, neural network and neuro-fuzzy inference system for determining the compressive strength of brick-mortar masonry by fusing nondestructive testing data. Eng Comput 37:77–91. https://doi.org/10.1007/s00366-019-00810-4

    Article  Google Scholar 

  17. Liu X, Athanasiou CE, Padture NP, Sheldon BW, Gao H (2020) A machine learning approach to fracture mechanics problems. Acta Mater 190:105–112. https://doi.org/10.1016/j.actamat.2020.03.016

    Article  Google Scholar 

  18. Goswami S, Yin M, Yu Y, Karniadakis GE (2022) A physics-informed variational deeponet for predicting crack path in quasi-brittle materials. Comput Methods Appl Mech Eng 391:114587. https://doi.org/10.1016/j.cma.2022.114587

    MathSciNet  Article  MATH  Google Scholar 

  19. He Y, Zhang L, Chen Z, Li CY (2022) A framework of structural damage detection for civil structures using a combined multi-scale convolutional neural network and echo state network. Eng Comput. https://doi.org/10.1007/s00366-021-01584-4

    Article  Google Scholar 

  20. Mozaffar M, Bostanabad R, Chen W, Ehmann K, Cao J, Bessa MA (2019) Deep learning predicts path-dependent plasticity. Proc Natl Acad Sci USA 116(52):26414–26420. https://doi.org/10.1073/pnas.1911815116

    Article  Google Scholar 

  21. Sambath S, Nagaraj P, Selvakumar N (2011) Automatic defect classification in ultrasonic NDT using artificial intelligence. J Nondestr Eval 30(1):20–28. https://doi.org/10.1007/s10921-010-0086-0

    Article  Google Scholar 

  22. Yang P, Li Q (2014) Wavelet transform-based feature extraction for ultrasonic flaw signal classification. Neural Comput Appl 24(3–4):817–826. https://doi.org/10.1007/s00521-012-1305-7

    Article  Google Scholar 

  23. Liu J, Xu G, Ren L, Qian Z, Ren L (2017) Defect intelligent identification in resistance spot welding ultrasonic detection based on wavelet packet and neural network. Int J Adv Manuf Technol 90(9–12):2581–2588. https://doi.org/10.1007/s00170-016-9588-y

    Article  Google Scholar 

  24. Meng M, Chua YJ, Wouterson E, Ong CPK (2017) Ultrasonic signal classification and imaging system for composite materials via deep convolutional neural networks. Neurocomputing 257:128–135. https://doi.org/10.1016/j.neucom.2016.11.066

    Article  Google Scholar 

  25. Ye R, Pan CS, Chang M, Yu Q (2018) Intelligent defect classification system based on deep learning. Adv Mech Eng 10(3):1–7. https://doi.org/10.1177/1687814018766682

    Article  Google Scholar 

  26. Meijer D, Scholten L, Clemens F, Knobbe A (2019) A defect classification methodology for sewer image sets with convolutional neural networks. Autom Constr 104:281–298. https://doi.org/10.1016/j.autcon.2019.04.013

    Article  Google Scholar 

  27. Wang M, Cheng JC (2020) A unified convolutional neural network integrated with conditional random field for pipe defect segmentation. Comput-Aided Civil Infrastruct Eng 35(2):162–177. https://doi.org/10.1111/mice.12481

    Article  Google Scholar 

  28. Han X, Zhao Z, Chen L, Hu X, Tian Y, Zhai C, Wang L, Huang X (2022) Structural damage-causing concrete cracking detection based on a deep-learning method. Constr Build Mater 337(2022):127562. https://doi.org/10.1016/j.conbuildmat.2022.127562

    Article  Google Scholar 

  29. Ji X, Yan Q, Huang D, Wu B, Xu X, Zhang A, Liao G, Zhou J, Wu M (2020) Filtered selective search and evenly distributed convolutional neural networks for casting defects recognition. J Mater Process Technol. https://doi.org/10.1016/j.jmatprotec.2021.117064

    Article  Google Scholar 

  30. Hu Y, Wang J, Zhu Y, Wang Z, Chen D, Zhang J, Ding H (2021) Automatic defect detection from X-ray scans for aluminum conductor composite core wire based on classification neutral network. NDT and E Int 124:102549. https://doi.org/10.1016/j.ndteint.2021.102549

    Article  Google Scholar 

  31. Yang L, Wang H, Huo B, Li F, Liu Y (2021) An automatic welding defect location algorithm based on deep learning. NDT and E Int 120(March):102435. https://doi.org/10.1016/j.ndteint.2021.102435

    Article  Google Scholar 

  32. Kiranyaz S, Ince T, Abdeljaber O, Avci O, Gabbouj M (2019) 1-D convolutional neural networks for signal processing applications. IEEE Int Conf Acoust Speech Signal Process. https://doi.org/10.1109/ICASSP.2019.8682194

    Article  Google Scholar 

  33. Kiranyaz S, Avci O, Abdeljaber O, Ince T, Gabbouj M, Inman DJ (2021) 1D convolutional neural networks and applications: a survey. Mech Syst Signal Process 151:107398. https://doi.org/10.1016/j.ymssp.2020.107398

    Article  Google Scholar 

  34. Janssens O, Slavkovikj V, Vervisch B, Stockman K, Loccufier M, Verstockt S, Van de Walle R, Van Hoecke S (2016) Convolutional neural network based fault detection for rotating machinery. J Sound Vib 377:331–345. https://doi.org/10.1016/j.jsv.2016.05.027

    Article  Google Scholar 

  35. Zhang W, Peng G, Li C, Chen Y, Zhang Z (2022) A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors (Switzerland). https://doi.org/10.3390/s17020425

    Article  Google Scholar 

  36. Zhang W, Li C, Peng G, Chen Y, Zhang Z (2018) A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load. Mech Syst Signal Process 100:439–453. https://doi.org/10.1016/j.ymssp.2017.06.022

    Article  Google Scholar 

  37. Abdeljaber O, Avci O, Kiranyaz S, Gabbouj M, Inman DJ (2017) Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks. J Sound Vib 388:154–170. https://doi.org/10.1016/j.jsv.2016.10.043

    Article  Google Scholar 

  38. Munir N, Kim HJ, Park J, Song SJ, Kang SS (2019) Convolutional neural network for ultrasonic weldment flaw classification in noisy conditions. Ultrasonics 94(2018):74–81. https://doi.org/10.1016/j.ultras.2018.12.001

    Article  Google Scholar 

  39. Munir N, Park J, Kim HJ, Song SJ, Kang SS (2020) Performance enhancement of convolutional neural network for ultrasonic flaw classification by adopting autoencoder. NDT and E Int 111:102218. https://doi.org/10.1016/j.ndteint.2020.102218

    Article  Google Scholar 

  40. Niu S, Srivastava V (2022) Simulation trained CNN for accurate embedded crack length, location, and orientation prediction from ultrasound measurements. Int J Solids Struct 242:111521. https://doi.org/10.1016/j.ijsolstr.2022.111521

    Article  Google Scholar 

  41. Srivastava V, Chester SA, Anand L (2010) Thermally actuated shape-memory polymers: experiments, theory, and numerical simulations. J Mech Phys Solids 58:1100–1124. https://doi.org/10.1016/j.jmps.2010.04.004

    Article  MATH  Google Scholar 

  42. Srivastava V, Chester SA, Ames NM, Anand L (2010) A thermo-mechanically-coupled large-deformation theory for amorphous polymers in a temperature range which spans their glass transition. Int J Plast 26(8):1138–1182. https://doi.org/10.1016/j.ijplas.2010.01.004

    Article  MATH  Google Scholar 

  43. Kothari M, Niu S, Srivastava V (2019) A thermo-mechanically coupled finite strain model for phase-transitioning austenitic steels in ambient to cryogenic temperature range. J Mech Phys Solids 133:103729. https://doi.org/10.1016/j.jmps.2019.103729

    MathSciNet  Article  Google Scholar 

  44. Bai Y, Kaiser NJ, Coulombe KL, Srivastava V (2021) A continuum model and simulations for large deformation of anisotropic fiber-matrix composites for cardiac tissue engineering. J Mech Behav Biomed Mater 121(May):104627. https://doi.org/10.1016/j.jmbbm.2021.104627

    Article  Google Scholar 

  45. Zhong J, Srivastava V (2021) A higher-order morphoelastic beam model for tubes and filaments subjected to biological growth. Int J Solids Struct 233(January):111235. https://doi.org/10.1016/j.ijsolstr.2021.111235

    Article  Google Scholar 

  46. Kim J, Mailand E, Ang I, Sakar MS, Bouklas N (2021) A model for 3D deformation and reconstruction of contractile microtissues. Soft Matter 17(45):10198–10209. https://doi.org/10.1039/d0sm01182g

    Article  Google Scholar 

  47. Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958

    MathSciNet  MATH  Google Scholar 

  48. Diederik P. Kingma, Jimmy Ba, Adam: A method for stochastic optimization, Proceedings at the 3rd International Conference for Learning Representations, San Diego, 2015. https://doi.org/10.48550/arXiv.1412.6980

  49. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: machine learning in python. J Mach Learn Res 12:2825–2830

    MathSciNet  MATH  Google Scholar 

Download references

Acknowledgements

The authors would like to thank Dr. Rainer Hebert, professor, and director of Pratt and Whitney Additive Manufacturing Center at the University of Connecticut for kindly providing the 3D-printed metal specimens. The authors would like to thank the U.S. Department of Transportation Pipeline and Hazardous Materials Safety Administration (USDOT PHMSA) for the financial support under Grant No. 693JK31950001CAAP and 693JK32050001CAAP. The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any agency of the U.S. government.

Author information

Authors and Affiliations

  1. School of Engineering, Brown University, Providence, RI, 02912, USA

    Sijun Niu & Vikas Srivastava

  2. Center for Biomedical Engineering, Brown University, Providence, RI, 02912, USA

    Vikas Srivastava

Authors
  1. Sijun Niu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Vikas Srivastava
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Vikas Srivastava.

Ethics declarations

Conflict of interest

The authors declare that there is no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Niu, S., Srivastava, V. Ultrasound classification of interacting flaws using finite element simulations and convolutional neural network. Engineering with Computers (2022). https://doi.org/10.1007/s00366-022-01681-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00366-022-01681-y

Keywords

  • Ultrasound
  • Non-destructive evaluation
  • NDT
  • Finite element method
  • Neural networks
  • Flaw classification
  • Crack
  • Corrosion
  • Pipelines
  • Inline Inspection