Fully convolutional network-based ultrasonic inversion for multi-layered bonded composites

Fully convolutional network-based ultrasonic inversion for multi-layered bonded composites

Mason Doust, Zhifei Xiao, Huadong Mo, Jing Rao

Abstract. Ultrasonic methods are widely used for the detection and characterisation of defects in multi-layered bonded composites. However, quantitative reconstruction of defects, such as disbonds, which can affect adhesive bond integrity and severely reduce the strength of assemblies, remains challenging. In this work, a supervised full convolutional network (FCN)-based ultrasonic method is used to quantitatively reconstruct defects hidden in multi-layered bonded composites. This proposed method consists of a training process and a predicting process. In the training process, the FCN builds a non-linear mapping from the ultrasound data to the corresponding longitudinal (L-wave) velocity model. In the predicting process, the network obtained from the training process is used to directly reconstruct the L-wave velocity models from the new measured ultrasonic data of adhesively bonded composites. The simulation results show that the FCN-based ultrasonic inversion method has the ability to achieve the accurate quantitative reconstruction of ultrasonic L-wave velocity models of the high contrast defects, which has potential in online detection of multi-layered bonded composites.

Keywords
Quantitative Reconstruction, Multi-Layered Bonded Composites, Deep Learning-Based Inversion, Defect Detection, Non-Destructive Evaluation

Published online 3/30/2023, 7 pages
Copyright © 2023 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA

Citation: Mason Doust, Zhifei Xiao, Huadong Mo, Jing Rao, Fully convolutional network-based ultrasonic inversion for multi-layered bonded composites, Materials Research Proceedings, Vol. 27, pp 315-321, 2023

DOI: https://doi.org/10.21741/9781644902455-41

The article was published as article 41 of the book Structural Health Monitoring

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

References
[1] P. Pyzik, A. Ziaja-Sujdak, J. Spytek, et al., Detection of disbonds in adhesively bonded aluminum plates using laser-generated shear acoustic waves, Photoacoustics, 21 (2021) 100226. https://doi.org/10.1016/j.pacs.2020.100226
[2] D. W. Yan, S. A. Neild, B. W, Drinkwater. Modelling and measurement of the nonlinear behaviour of kissing bonds in adhesive joints, NDT & E Int., 47 (2012) 18–25. https://doi.org/10.1016/j.ndteint.2011.12.003
[3] K. S. Zhang, S. C. Li, Z. G. Zhou, Detection of disbonds in multi-layer bonded structures using the laser ultrasonic pulse-echo mode, Ultrasonics, 94 (2019) 411–418. https://doi.org/10.1016/j.ultras.2018.06.005
[4] M. Y. Y. Hung, H. M. Shang, L. X. Yang, Unified approach for holography and shearography in surface deformation measurement and nondestructive testing, Opt. Eng., 42 (5) (2000) 1197–1207. https://doi.org/10.1117/1.1567263
[5] R. J. Freemantle, R. E. Challis, Ultrasonic compression wave NDT of adhesively bonded automotive structures, UTonline Appl. Workshop, 2(5) (1997).
[6] J. M. Allin, P. Cawley, M. J. S. Lowe, Adhesive disbond detection of automotive components using first mode ultrasonic resonance, NDT & E Int., 36 (2003) 503–514. https://doi.org/10.1016/S0963-8695(03)00045-8
[7] S. M. Deregowski, Common-offset migrations and velocity analysis, First Break, 8 (6) (1990) 224–234. https://doi.org/10.3997/1365-2397.1990011
[8] A. Tarantola, Inversion of seismic reflection data in the acoustic approximation, Geophysics, 49 (8) (1986) 1259–1266. https://doi.org/10.1190/1.1441754
[9] Y. Kim, N. Nakata, Geophysical inversion versus machine learning in inverse problems, The Leading Edge, 37 (2018) 894–901. https://doi.org/10.1190/tle37120894.1
[10] W. Rawat, Z. Wang, Deep convolutional neural networks for image classification: A comprehensive review, Neural Comput., 29 (2017) 2352–2449. https://doi.org/10.1162/neco_a_00990
[11] A. Voulodimos, N. Doulamis, A. Doulamis, E. Protopapadakis, Deep learning for computer vision: a brief review, Comput. Intell. Neurosci, (2018) 7068349. https://doi.org/10.1155/2018/7068349
[12] M. Ghosh, H. Mukherjee, et al., Identifying the presence of graphical texts in scene images using cnn, In Proc. Int. Conf. Doc. Anal. Recognit. (ICDAR), 1 (2019) 86–91. https://doi.org/10.1109/ICDARW.2019.00020
[13] K. H. Jin, M. T. McCann, E. Froustey, M. Unser, Deep convolutional neural network for inverse problems in imaging, IEEE Trans. Image Process., 26 (9) (2017) 4509–4522. https://doi.org/10.1109/TIP.2017.2713099
[14] S. C. Li, B. Liu, et al., Deep-learning inversion of seismic data, IEEE Trans. Geosci. Remote Sens., 58 (3) (2020) 2135–2149. https://doi.org/10.1109/TGRS.2019.2953473
[15] X. F. Jiang, Y. W. Wang, W. B. Liu, et al., Capsnet, cnn, fcn: Comparative performance evaluation for image classification, Int. J. Mach. Learn. Comput., 9(6) (2019) 840–848. https://doi.org/10.18178/ijmlc.2019.9.6.881
[16] J. Rao, F. S. Yang, et al., Quantitative reconstruction of defects in multi-layered bonded composites using fully convolutional network-based ultrasonic inversion, submitted to J. Sound Vib.
[17] J. Rao, F. S. Yang, et al., Quantitative reconstruction of defects in multi-layered bonded composites using fully convolutional network-based ultrasonic inversion. arXiv preprint arXiv:2109.07284, 2021.