Composite pipelines, widely applied in the oil and gas industry due to their high strength-to-weight ratio and corrosion resistance, are nonetheless vulnerable to damage under harsh operating conditions, making structural health monitoring (SHM) crucial for enhancing safety and preventing potential failures. However, the complex service environment and anisotropic properties of composite pipelines pose challenges for traditional SHM methods, which often struggle to extract underlying features and identify structural defects. To address these challenges, a new graph deep learning method, termed physical-guided graph deep learning (PGGDL), which leverages physical rules to enhance defect detection accuracy in composite pipeline health monitoring, is proposed in this study. The PGGDL method constructs graph data by integrating signals based on physical rules, including the sensor arrangement and the guided wave propagation mechanism, enabling efficient data fusion. Meanwhile, the PGGDL constructs a temporal model and a graph attention network to capture both node-level and global spatial features. Furthermore, an adaptive learning rate strategy is designed to dynamically adjust the learning rate, improving training efficiency, achieving faster convergence, and enhancing model stability. Four experimental cases are conducted to demonstrate that the PGGDL outperforms traditional deep learning models in accurately detecting and localizing pipeline defects with limited data.
FuHSharif KhodaeiZAliabadiMHF. An event-triggered energy-efficient wireless structural health monitoring system for impact detection in composite airframes. IEEE Internet Things J2019; 6(1): 1183–1192.
2.
SilvaWLopesRZscherpelU, et al. X-ray imaging techniques for inspection of composite pipelines. Micron2021; 145: 103033.
3.
CoshamAHopkinsPMacdonaldKA.Best practice for the assessment of defects in pipelines – corrosion. Eng Fail Anal2007; 14(7): 1245–1265.
4.
VishnuvardhanSMurthyARChoudharyA.A review on pipeline failures, defects in pipelines and their assessment and fatigue life prediction methods. Int J Press Ves Pip2023; 201: 104853.
5.
NiPMangalathuS.Fragility analysis of gray iron pipelines subjected to tunneling induced ground settlement. Tunn Undergr Space Technol2018; 76: 133–144.
6.
SharmaVBSinghKGuptaR, et al. Review of structural health monitoring techniques in pipeline and wind turbine industries. Appl Syst Innov2021; 4(3): 59.
7.
LeeHYangJSohnH.Baseline-free pipeline monitoring using optical fiber–guided laser ultrasonics. Struct Health Monit2012; 11(6): 684–695.
8.
XuXLiFHuY, et al. A joint sparse representation-based approach for anomaly rectification in extended-duration SHM systems with sensor networks. Mech Syst Signal Process2025; 231: 112666.
9.
ZangXXuZDLuH, et al. Ultrasonic guided wave techniques and applications in pipeline defect detection: a review. Int J Press Vessel Pip2023; 206: 105033.
10.
HouEZYRostamiJ. Novel design of an effective pneumatic magnetostrictive patch transducer based on the ultrasonic guided wave for application of fast pipe health inspection. Meas Sci Technol2021; 32(10): 105904.
11.
WilleyCLSimonettiFNagyPB, et al. Guided wave tomography of pipes with high-order helical modes. NDT & E Int2014; 65: 8–21.
12.
LiuKWuZJiangY, et al. Guided waves based diagnostic imaging of circumferential cracks in small-diameter pipe. Ultrasonics2016; 65: 34–42.
13.
BangSSLeeYHShinYJ.Defect detection in pipelines via guided wave-based time–frequency-domain reflectometry. IEEE Trans Instrum Meas2021; 70: 1–11.
14.
ChenQCaoJZhuS.Data-driven monitoring and predictive maintenance for engineering structures: technologies, implementation challenges, and future directions. IEEE Internet Things J2023; 10(16): 14527–14551.
15.
ChenZMauricioALiW, et al. A deep learning method for bearing fault diagnosis based on cyclic spectral coherence and convolutional neural networks. Mech Syst Signal Process2020; 140: 106683.
16.
EwaldVGrovesRMBenedictusR.DeepSHM: a deep learning approach for structural health monitoring based on guided Lamb wave technique. Sens Smart Struct Technol Civ Mech Aerosp Syst2019; 10970: 84–99.
17.
ArafinPBillahAMIssaA.Deep learning-based concrete defects classification and detection using semantic segmentation. Struct Health Monit2024; 23(1): 383–409.
18.
MaQDuGYuZ, et al. Classification of damage types in liquid-filled buried pipes based on deep learning. Meas Sci Technol2022; 34(2): 025010.
19.
WangXLinMLiJ, et al. Ultrasonic guided wave imaging with deep learning: applications in corrosion mapping. Mech Syst Signal Process2022; 169: 108761.
20.
LiuHZhangY.Deep learning based crack damage detection technique for thin plate structures using guided lamb wave signals. Smart Mater Struct2019; 29(1): 015032.
21.
ZhangXTangLChenJ.Fault diagnosis for electro-mechanical actuators based on STL-HSTA-GRU and SM. IEEE Trans Instrum Meas2021; 70: 1–16.
22.
ScarselliFGoriMTsoiAC, et al. The graph neural network model. IEEE Trans Neural Netw2009; 20(1): 61–80.
23.
KipfTNWellingM.Semi-supervised classification with graph convolutional networks. In: International Conference on Learning Representations (ICLR), 2017.
24.
VeličkovićPCucurullGCasanovaA, et al. Graph attention networks. In: International Conference on Learning Representations (ICLR), 2018.
25.
XuKHuWLeskovecJ, et al. How powerful are graph neural networks? In: International Conference on Learning Representations (ICLR), 2019.
26.
ChenZWangNSunH.Graph oscillators: physics-guided graph modeling of mass–spring–damper systems for trajectory prediction and damage localization. Mech Syst Signal Process2024; 212: 111297.
27.
BloemheuvelSvan den HoogenJAtzmuellerM.A computational framework for modeling complex sensor network data using graph signal processing and graph neural networks in Structural Health Monitoring. Appl Netw Sci2021;6:97.
28.
YuXTangBZhangK.Fault diagnosis of wind turbine gearbox using a novel method of fast deep graph convolutional networks. IEEE Trans Instrum Meas2021; 70: 1–14.
29.
JianXXiaYDuthéG, et al. Using graph neural networks and frequency domain data for automated operational modal analysis of populations of structures. arXiv.2407.06492, 2024.
30.
YangZLiuZZhouJ, et al. A graph neural network (GNN) method for assigning gas calorific values to natural gas pipeline networks. Energy2023; 278: 127875.
31.
YacoutiMShakibaM.Integrated convolutional and graph neural networks for predicting mechanical fields in composite microstructures. Compos Part A Appl Sci Manuf2025; 190: 108618.
32.
ZhaoWLiMHarleyJB, et al. Reconstruction of Lamb wave dispersion curves by sparse representation with continuity constraints. J Acoust Soc Am2017; 141(2): 749.
33.
HuYCuiFLiF, et al. Sparse wavenumber analysis of guided wave based on hybrid Lasso regression in composite laminates. Struct Health Monit2021; 21(4): 1367–1378.
34.
SimonyanKZissermanA.Very deep convolutional networks for large-scale image recognition. In: International Conference on Learning Representations (ICLR), 2015.
35.
HeKZhangXRenS, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
36.
DefferrardMBressonXVandergheynstP.Convolutional neural networks on graphs with fast localized spectral filtering. In: Proceedings of the 30th international conference on neural information processing systems (NIPS), Barcelona, Spain, 5–10 December 2016. Red Hook, NY: Curran Associates Inc.
