Online monitoring of bolt torque is critical to ensure the safe operation of bolted structures. Guided waves have been intensively explored for bolt loosening monitoring. Nevertheless, guided waves are excessively sensitive to fluctuation of ambient temperature. As a result of the complexity of wave transmitting across a bolted joint, it is highly challenging to compensate for the effect of temperature. To this end, an attention-based multi-task network is developed towards accurate detection of bolt loosening in multi-bolt connections over a wide range of temperature variation. By integrating improved attention gate modules in a modified U-Net architecture, an attention U-Net is configured for temperature compensation. A two-layer convolutional subnetwork is connected in series behind the attention U-Net to identify bolt loosening. Experimental validation is carried out on a bolt jointed lap plate simulating a real aircraft structure. The results have proved that the developed multi-task network achieves temperature compensation and accurate bolt loosening identification. To further understand the multi-task network, the Integrated Gradients method and a simplified structure of the bolt lap plate are used to interpret the developed network. It is proved that the A0 mode is sensitive to bolt loosening, while the S0 mode is not.
WangFSongG. 1D-TICapsNet: an audio signal processing algorithm for bolt early looseness detection. Structural Health Monitoring. Epub ahead of print 15 December 2020. DOI: 10.1177/1475921720976989.
2.
ZhangZLiuMSuZ, et al. Quantitative evaluation of residual torque of a loose bolt based on wave energy dissipation and vibro-acoustic modulation: A comparative study. Journal of Sound and Vibration 2016; 383: 156–170. DOI: 10.1016/j.jsv.2016.07.001.
3.
ZhangZXiaoYSuZ, et al. Continuous monitoring of tightening condition of single-lap bolted composite joints using intrinsic mode functions of acoustic emission signals: a proof-of-concept study. Structural Health Monitoring2018; 18: 1219–1234. DOI: 10.1177/1475921718790768.
4.
ZhuYLiFHuY. The contact characteristics analysis for rod fastening rotors using ultrasonic guided waves. Measurement2020; 151: 107149. DOI: 10.1016/j.measurement.2019.107149.
5.
HongMMaoZToddMD, et al. Uncertainty quantification for acoustic nonlinearity parameter in Lamb wave-based prediction of barely visible impact damage in composites. Mechanical Systems and Signal Processing 2017; 82: 448–460. DOI: 10.1016/j.ymssp.2016.05.035.
6.
WangKLiuMSuZ, et al. Analytical insight into “breathing” crack-induced acoustic nonlinearity with an application to quantitative evaluation of contact cracks. Ultrasonics 2018; 88: 157–167. DOI: 10.1016/j.ultras.2018.03.008.
7.
DuFXuCRenH, et al. Structural health monitoring of bolted joints using guided waves: a review. In: MagdAbdel WahabYunLai ZhouMaiaNMM (eds.) Structural health monitoring from sensing to processing. London, UK: IntechOpen, 2018, pp.163–180.
8.
YangJChangF-K. Detection of bolt loosening in C–C composite thermal protection panels: I. Diagnostic principle. Smart Materials and Structures2006; 15: 581–590. DOI: 10.1088/0964-1726/15/2/041.
9.
WangTSongGWangZ, et al. Proof-of-concept study of monitoring bolt connection status using a piezoelectric based active sensing method. Smart Materials and Structures2013; 22: 087001.
10.
HaynesCYeagerMToddM, et al. Monitoring bolt torque levels through signal processing of full-field ultrasonic data. In: Health Monitoring of Structural and Biological Systems2014, p.906428. San Diego, CA: International Society for Optics and Photonics.
11.
MascarenasDMacknellyDMullinsJ, et al. Dynamic characterization of satellite assembly for responsive space applications. Measurement Science and Technology2013; 24: 075101.
12.
MontoyaADoyleDMajiA, et al. Ultrasonic evaluation of bolted connections in satellites. Research in Nondestructive Evaluation2014; 25: 44–62.
13.
JiangTWuQWangL, et al. Monitoring of bolt looseness-induced damage in steel truss arch structure using piezoceramic transducers. Ieee Sens J2018; 18: 6677–6685.
14.
DuFXuCZhangJ. A bolt preload monitoring method based on the refocusing capability of virtual time reversal. Struct Control Health Monitor2019; 26(8): e2370. DOI: 10.1002/stc.2370.
15.
WangFHoSCMSongG. Monitoring of early looseness of multi-bolt connection: a new entropy-based active sensing method without saturation. Smart Mater Struct2019; 28: 10LT01. DOI: 10.1088/1361-665X/ab3a08.
16.
ParvasiSMHoSCMKongQ, et al. Real time bolt preload monitoring using piezoceramic transducers and time reversal technique—a numerical study with experimental verification. Smart Mater Struct2016; 25: 085015.
17.
MitraMGopalakrishnanS. Guided wave based structural health monitoring: a review. Smart Materials and Structures2016; 25: 053001. DOI: 10.1088/0964-1726/25/5/053001.
18.
LuYMichaelsJE. A methodology for structural health monitoring with diffuse ultrasonic waves in the presence of temperature variations. Ultrasonics2005; 43: 717–731. 2005/07/05. DOI: 10.1016/j.ultras.2005.05.001.
19.
KonstantinidisGDrinkwaterBWWilcoxPD. The temperature stability of guided wave structural health monitoring systems. Smart Mater Struct2006; 15: 967–976. DOI: 10.1088/0964-1726/15/4/010.
20.
AnY-KSohnH. Integrated impedance and guided wave based damage detection. Mech Syst Signal Process2012; 28: 50–62. DOI: 10.1016/j.ymssp.2011.11.016.
21.
ClarkeTSimonettiFCawleyP. Guided wave health monitoring of complex structures by sparse array systems: Influence of temperature changes on performance. J Sound Vibrat2010; 329: 2306–2322.
22.
CroxfordAJWilcoxPD, et al. Strategies for overcoming the effect of temperature on guided wave structural health monitoring. Health Monitor Struct Biolog Syst2007; 6532: 590–599.
23.
MarianiSHeinleinSCawleyP. Compensation for temperature-dependent phase and velocity of guided wave signals in baseline subtraction for structural health monitoring. Struct Health Monitor2019; 19(1): 147592171983515. DOI: 10.1177/1475921719835155.
24.
YueNAliabadiMH. A scalable data-driven approach to temperature baseline reconstruction for guided wave structural health monitoring of anisotropic carbon-fibre-reinforced polymer structures. Struct Health Monitor2019; 19(5): 1487–1506. DOI: 10.1177/1475921719887109.
25.
GorginRLuoYWuZ. Environmental and operational conditions effects on Lamb wave based structural health monitoring systems: a review. Ultrasonics2020; 105: 106114. DOI: 10.1016/j.ultras.2020.106114.
26.
NazarkoPZiemianskiL. Force identification in bolts of flange connections for structural health monitoring and failure prevention. Procedia Struct Integrity2017; 5: 460–467. DOI: 10.1016/j.prostr.2017.07.142.
27.
LiangDYuanS-f. Decision fusion system for bolted joint monitoring. Shock and Vibration2015; 2015: 11.
28.
WangFChenZSongG. Monitoring of multi-bolt connection looseness using entropy-based active sensing and genetic algorithm-based least square support vector machine. Mech Syst Signal Process2020; 136: 106507. DOI: 10.1016/j.ymssp.2019.106507.
29.
WangFChenZSongG. Smart crawfish: a concept of underwater multi-bolt looseness identification using entropy-enhanced active sensing and ensemble learning. Mech Syst Sig Process2021; 149: 107186. DOI: 10.1016/j.ymssp.2020.107186.
30.
YingYGarrettJHOppenheimIJ, et al. Toward data-driven structural health monitoring: application of machine learning and signal processing to damage detection. J Comput Civil Eng2013; 27: 667–680. DOI: 10.1061/(asce)cp.1943-5487.0000258.
31.
WangPZhouWBaoY, et al. Ice monitoring of a full-scale wind turbine blade using ultrasonic guided waves under varying temperature conditions. Struct Control Health Monitor2018; 25: e2138. DOI: 10.1002/stc.2138.
32.
LeCunYBengioYHintonG. Deep learning. Nature2015; 521: 436–444. DOI: 10.1038/nature14539.
33.
ZhuJChenNShenC. A new deep transfer learning method for bearing fault diagnosis under different working conditions. IEEE Sens J2020; 20: 8394–8402. DOI: 10.1109/jsen.2019.2936932.
34.
LiJWangYZiY, et al. Whitening-net: a generalized network to diagnose the faults among different machines and conditions. IEEE transactions on neural networks and learning systems. Epub ahead of print 22April2021. DOI: 10.1109/TNNLS.2021.3071564.
35.
BukhshZAJansenNSaeedA. Damage detection using in-domain and cross-domain transfer learning. Neural Computing and Applications2021; 33: 16921–16936. DOI: 10.1007/s00521-021-06279-x.
36.
IyerSVelmuruganTGandomiAH, et al. Structural health monitoring of railway tracks using IoT-based multi-robot system. Neural Computing and Applications2020; 33: 5897–5915. DOI: 10.1007/s00521-020-05366-9.
37.
YangSHuangY. Damage identification method of prestressed concrete beam bridge based on convolutional neural network. Neural Computing and Applications2020; 33: 535–545. DOI: 10.1007/s00521-020-05052-w.
38.
ZhangYSunXLohKJ, et al. Autonomous bolt loosening detection using deep learning. Structural Health Monitoring2019; 19: 105–122. DOI: 10.1177/1475921719837509.
39.
WangF. A novel autonomous strategy for multi-bolt looseness detection using smart glove and Siamese double-path CapsNet. Struct Health Monitor. Epub ahead of print 2021. DOI: 10.1177/14759217211054575.
40.
ZhangTBiswalSWangY. SHMnet: condition assessment of bolted connection with beyond human-level performance. Struct Health Monitor2019; 19: 1188–1201. DOI: 10.1177/1475921719881237.
41.
ZhangBHongXLiuY. Multi-task deep transfer learning method for guided wave based integrated health monitoring using piezoelectric transducers. IEEE Sens J2020; 20(23): 14391–14400. DOI: 10.1109/jsen.2020.3009194.
42.
VaswaniAShazeerNParmarN, et al. Attention is all you need. In: Proceedings of the 31st international conference on neural information processing systems, New York, USA, 2017, pp.6000–6010.
43.
JetleySLordNALeeN, et al. Learn to pay attention. In: International conference on learning representations, Massachusetts, USA, 2018.
44.
FuJLiuJTianH, et al. Dual attention network for scene segmentation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition2019, pp.3146–3154. Washington, DC: IEEE.
45.
OktayOSchlemperJFolgocLL, et al. Attention u-net: learning where to look for the pancreas. In: Conference on Medical Imaging with Deep Learning, 2018, Amsterdam, The Netherlands, pp. 1–10.
46.
AbrahamNKhanNM. A novel focal tversky loss function with improved attention u-net for lesion segmentation. In: 2019 IEEE 16th international symposium on biomedical imaging (ISBI 2019)2019, pp.683–687. Washington, DC: IEEE.
47.
ZhangHGoodfellowIMetaxasD, et al. Self-attention generative adversarial networks. In: International conference on machine learning2019, pp.7354–7363. Princeton, NJ: PMLR.
48.
OskoueiRHKeikhosravyMSoutisC. Estimating clamping pressure distribution and stiffness in aircraft bolted joints by finite-element analysis. Proceed Instit Mech Eng G J Aerospace Eng2009; 223: 863–871. DOI: 10.1243/09544100jaero596.
49.
HudaFKajiwaraIHosoyaN, et al. Bolt loosening analysis and diagnosis by non-contact laser excitation vibration tests. Mech Syst Signal Process2013; 40: 589–604. DOI: 10.1016/j.ymssp.2013.05.023.
50.
KedraRRuckaM. Research on assessment of bolted joint state using elastic wave propagation. J Phys Conf Ser2015; 628: 012025.
51.
AmeriniFMeoM. Structural health monitoring of bolted joints using linear and nonlinear acoustic/ultrasound methods. Struct Health Monitor2011; 10: 659–672.
52.
CroxfordAJMollJWilcoxPD, et al. Efficient temperature compensation strategies for guided wave structural health monitoring. Ultrasonics2010; 50: 517–528. DOI: 10.1016/j.ultras.2009.11.002.
53.
CroxfordAJWilcoxPDDrinkwaterBW, et al. Strategies for guided-wave structural health monitoring. Proceed Royal Soc A Math Phys Eng Sci2007; 463: 2961–2981. DOI: 10.1098/rspa.2007.0048.
54.
DuFWuSXuC, et al. Electromechanical impedance temperature compensation and bolt loosening monitoring based on modified Unet and multitask learning. IEEE Sens J. Epub ahead of print 6December2021. DOI: 10.1109/JSEN.2021.3132943.
55.
FangYMaZZhangZ, et al. Dynamic multi-task learning with convolutional neural network. In: Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, California, USA, 2017, pp. 1668–1674.
56.
GuoSYanZZhangK, et al. Toward convolutional blind denoising of real photographs. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Washington, DC, USA, 2019, pp.1712–1722.
57.
SundararajanMTalyAYanQ. Axiomatic attribution for deep networks. In: DoinaPYee WhyeT (eds.). Proceedings of the 34th international conference on machine learning. PMLR, 2017, pp.3319–3328.
58.
AnconaMCeoliniEÖztireliC, et al. Towards better understanding of gradient-based attribution methods for Deep Neural Networks. In: 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30-May 3, 2018, Conference Track Proceedings2018, OpenReview. net.
