Stiffened panels are widely used in a range of engineering domains to enhance load-bearing capacity, reduce buckling, and improve overall structural integrity. Stiffener disbonding is a common failure mode, making it essential to detect these disbonds before they reach a critical level that could lead to structural failure. In this work, a transformer architecture has been implemented to detect the presence and extent of disbonding between the stiffener and the host aluminium plate. The research uses a dataset of Lamb waves obtained from an experimental setup with four types of specimens, ranging from no disbonding to disbonding of varying widths induced by Teflon strips. The problem is framed as a multi-class classification task with labels representing different disbonding levels. The transformer model utilizes a self-attention mechanism to analyze the Lamb wave responses and ascertain the presence and extent of disbanding. Operating in encoder-only mode, the tailored transformer accepts Lamb wave responses as input and outputs a class probability score for each of the four predefined disbonding levels using a softmax activation function. The developed model has been able to predict the state of the stiffened panels with an accuracy of 100%. To demonstrate the effectiveness of the transformer model, it has been compared to the convolutional neural network (CNN) model. When tested on an unseen noisy dataset, the transformer model achieved an accuracy of 100%, outperforming the CNN model, which achieved an accuracy of 89%.
AzimiMEslamlouADPekcanG (2020) Data-driven structural health monitoring and damage detection through deep learning: State-of-the-art review. Sensors20(10): 2778.
2.
BalasubramaniamKSikdarSWandowskiT, et al. (2022) Ultrasonic guided wave-based debond identification in a GFRP plate with l-stiffener. Smart Materials and Structures31(1): 015023.
3.
BhojanapalliSChakrabartiAGlasnerD, et al. (2021) Understanding robustness of transformers for image classification. In: Proceedings of the IEEE/CVF international conference on computer vision, pp.10231–10241.
4.
BijudasCRMitraMMujumdarPM (2013) Time reversed lamb wave for damage detection in a stiffened aluminum plate. Smart Materials and Structures22(10): 105026.
5.
BroerAGalanopoulosGBenedictusR, et al. (2022) Fusion-based damage diagnostics for stiffened composite panels. Structural Health Monitoring21(2): 613–639.
6.
DevlinJChangMWLeeK, et al. (2018) Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
7.
DingYJiaMMiaoQ, et al. (2022) A novel time–frequency transformer based on self–attention mechanism and its application in fault diagnosis of rolling bearings. Mechanical Systems and Signal Processing168: 108616.
8.
DongLXuSXuB (2018) Speech-transformer: a no-recurrence sequence-to-sequence model for speech recognition. In: 2018 IEEE international conference on acoustics, speech and signal processing (ICASSP), pp.5884–5888. IEEE.
9.
DosovitskiyABeyerLKolesnikovA, et al. (2020) An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929.
10.
FlahMNunezIBen ChaabeneW, et al. (2021) Machine learning algorithms in civil structural health monitoring: A systematic review. Archives of Computational Methods in Engineering28: 2621–2643.
11.
GolubMVShpakANMuellerI, et al. (2021) Lamb wave scattering analysis for interface damage detection between a surface-mounted block and elastic plate. Sensors21(3): 860.
12.
HavivARamOPressO, et al. (2022) Transformer language models without positional encodings still learn positional information. arXiv preprint arXiv:2203.16634.
13.
KazemnejadAPadhiINatesan RamamurthyK, et al. (2023) The impact of positional encoding on length generalization in transformers. Advances in Neural Information Processing Systems36: 24892–24928.
14.
LiSJinXXuanY, et al. (2019a) Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting. Advances in Neural Information Processing Systems32.
15.
LiuHZhangY (2020) Deep learning based crack damage detection technique for thin plate structures using guided lamb wave signals. Smart Materials and Structures29(1): 015032.
16.
LiuMChenSWongZZ, et al. (2020) In situ disbond detection in adhesive bonded multi-layer metallic joint using time-of-flight variation of guided wave. Ultrasonics102: 106062.
17.
LiXZhangWDingQ (2019b) Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism. Signal Processing161: 136–154.
18.
LongZZhangXZhangL, et al. (2021) Motor fault diagnosis using attention mechanism and improved adaboost driven by multi-sensor information. Measurement170: 108718.
19.
LuYYeLSuZ, et al. (2009) Artificial neural network (ANN)-based crack identification in aluminum plates with lamb wave signals. Journal of Intelligent Material Systems and Structures20(1): 39–49.
20.
MandalDDBanerjeeS (2019) Identification of breathing type disbonds in stiffened panels using non-linear lamb waves and built-in circular pwt array. Mechanical Systems and Signal Processing117: 33–51.
21.
MarksRClarkeAFeatherstonC, et al. (2016) Lamb wave interaction with adhesively bonded stiffeners and disbonds using 3d vibrometry. Applied Sciences6(1): 12.
22.
MelvilleJAlguriKSDeemerC, et al. (2018) Structural damage detection using deep learning of ultrasonic guided waves. AIP Conference Proceedings1949: 230004.
23.
MemmoloVMonacoEBoffaND, et al. (2018) Guided wave propagation and scattering for structural health monitoring of stiffened composites. Composite Structures184: 568–580.
24.
MilanoskiDGalanopoulosGBroerA, et al. (2021) A strain-based health indicator for the shm of skin-to-stringer disbond growth of composite stiffened panels in fatigue. In: European workshop on structural health monitoring: Special collection of 2020 papers-volume 1, pp.626–635. Springer.
25.
MitraMGopalakrishnanS (2016) Guided wave based structural health monitoring: A review. Smart Materials and Structures25(5): 053001.
26.
MollJKatholJFritzenC-P, et al. (2019) Open guided waves: Online platform for ultrasonic guided wave measurements. Structural Health Monitoring18(5–6): 1903–1914.
27.
NevesACGonzálezILeanderJ, et al. (2017) Structural health monitoring of bridges: A model-free ann-based approach to damage detection. Journal of Civil Structural Health Monitoring7(5): 689–702.
28.
NickWAsameneKBullockG, et al. (2015) A study of machine learning techniques for detecting and classifying structural damage. International Journal of Machine Learning and Computing5(4): 313–318.
29.
PandaRSRajagopalPBalasubramaniamK (2018) Rapid guided wave inspection of complex stiffened composite structural components using non-contact air-coupled ultrasound. Composite Structures206: 247–260.
30.
PandeyPRaiAMitraM (2022) Explainable 1-d convolutional neural network for damage detection using lamb wave. Mechanical Systems and Signal Processing164: 108220.
31.
PrajapatiKKRaiAMitraM (2022) Lamb wave-based damage detection using artificial neural network and automated feature extraction. Transactions of the Indian National Academy of Engineering7: 1009–1016.
32.
PyzikPZiaja-SujdakASpytekJ, et al. (2021) Detection of disbonds in adhesively bonded aluminum plates using laser-generated shear acoustic waves. Photoacoustics21: 100226.
33.
RaghavanA (2007) Guided-wave structural health monitoring. PhD Thesis.
34.
RaiAMitraM (2021) Lamb wave based damage detection in metallic plates using multi-headed 1-dimensional convolutional neural network. Smart Materials and Structures30(3): 035010.
35.
RamadasCBalasubramaniamKJoshiM, et al. (2011) Interaction of lamb mode (ao) with structural discontinuity and generation of “turning modes” in a t-joint. Ultrasonics51(5): 586–595.
36.
RautelaMGopalakrishnanS (2021) Ultrasonic guided wave based structural damage detection and localization using model assisted convolutional and recurrent neural networks. Expert Systems with Applications167: 114189.
37.
RothWGiurgiutiuV (2017) Structural health monitoring of an adhesive disbond through electromechanical impedance spectroscopy. International Journal of Adhesion and Adhesives73: 109–117.
38.
ShangZSunLXiaY, et al. (2021) Vibration-based damage detection for bridges by deep convolutional denoising autoencoder. Structural Health Monitoring20(4): 1880–1903.
39.
StaszewskiW (2005) Ultrasonic/guided waves for structural health monitoring. In: Key Engineering Materials, vol. 293. Trans Tech Publ, pp.49–62.
40.
SuCJiangMLvS, et al. (2019) Improved damage localization and quantification of cfrp using lamb waves and convolution neural network. IEEE Sensors Journal19(14): 5784–5791.
41.
TuliSCasaleGJenningsNR (2022) Tranad: Deep transformer networks for anomaly detection in multivariate time series data. arXiv preprint arXiv:2201.07284.
42.
VaswaniAShazeerNParmarN, et al. (2017) Attention is all you need. Advances in Neural Information Processing Systems30.
43.
VaswaniAShazeerNParmarN, et al. (2023) Attention is all you need. Advances in Neural Information Processing Systems30.
44.
WangQLiBXiaoT, et al. (2019) ‘Learning deep transformer models for machine translation’, arXiv preprint arXiv:1906.01787 .
45.
WuHXuJWangJ, et al. (2021a) Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. Advances in Neural Information Processing Systems34: 22419–22430.
46.
WuJXuXLiuC, et al. (2021b) Lamb wave-based damage detection of composite structures using deep convolutional neural network and continuous wavelet transform. Composite Structures276: 114590.
47.
XuJWuHWangJ, et al. (2021) Anomaly transformer: Time series anomaly detection with association discrepancy. arXiv preprint arXiv:2110.02642.
48.
YangCHHTsaiYYChenPY (2021) Voice2series: Reprogramming acoustic models for time series classification. In: International conference on machine learning, PMLR, pp.11808–11819.
49.
YangZYangHTianT, et al. (2023) A review on guided-ultrasonic-wave-based structural health monitoring: From fundamental theory to machine learning techniques. Ultrasonics133: 107014.
50.
YelveNPMitraMMujumdarPM (2015) Detection of stiffener disbonding in a stiffened aluminium panel using nonlinear lamb wave. Applied Acoustics89: 267–272.
51.
YeXJinTYunC (2019) A review on deep learning-based structural health monitoring of civil infrastructures. Smart Structures and Systems24(5): 567–585.
52.
YuYWangCGuX, et al. (2019) A novel deep learning-based method for damage identification of smart building structures. Structural Health Monitoring18(1): 143–163.
53.
ZarouchasDSAlderliestenRC (2015) The effect of disbonds on stability aspects of adhesively bonded aluminum panels during compression loading. Thin-Walled Structures96: 372–382.
54.
ZerveasGJayaramanSPatelD, et al. (2021) A transformer-based framework for multivariate time series representation learning. In: Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, pp.2114–2124.
55.
ZhangHSongHLiS, et al. (2024) A survey of controllable text generation using transformer-based pre-trained language models. ACM Computing Surveys56: 1–37.
56.
ZhengKLiZMaZ, et al. (2019) Damage detection method based on lamb waves for stiffened composite panels. Composite Structures225: 111137.
57.
ZhouTMaZWenQ, et al. (2022) Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting. In: International conference on machine learning, PMLR, pp.27268–27286.
58.
ZinnoRHaghshenasSSGuidoG, et al. (2022) Artificial intelligence and structural health monitoring of bridges: A review of the state-of-the-art. IEEE Access10: 88058–88078.
