Because hammering sound tests are inexpensive and can be performed easily, they are commonly used as an inspection method for examining the presence of defect areas (voids or peelings) in aged concrete structures. However, the evaluation of the health of concrete using hammering sounds depends on the subjective experience of the inspector. Therefore, there is a demand to develop a highly reliable and objective diagnostic method that is accurate and efficient. In this study, we used a convolutional autoencoder (CAE) to develop a diagnostic method that could assist the inspectors with quantitative diagnostic results of tapping sound when detecting defect areas in concrete. In particular, we verified the anomaly detection accuracy of hammering sound data of actual bridges that have deteriorated over time using the proposed CAE model.
Ministry of Land, Transport, Infrastructure and Tourism. Regarding the spalling accident occurred on railway structures. Procedia Eng2014; 171: 926–933.
2.
Abdel-QaderIAbudayyehOKellyME.Analysis of edge-detection techniques for crack identification in bridges. J Comput Civ Eng2003; 17(4): 255–263.
3.
NishikawaTYoshidaJSugiyamaT, et al. Concrete crack detection by multiple sequential image filtering. Comput-Aided Civ Inf Eng2012; 27(1): 29–47.
4.
SuzukiTOgataHTakadaR, et al. Use of acoustic emission and X-ray computed tomography for damage evaluation of freeze-thawed concrete. Constr Build Mater2010; 24(12): 2347–2352.
5.
MarzecITejchmanJSkarżyńskiŁ.Fracture evolution in concrete compressive fatigue experiments based on X-ray micro-CT images. Int J Fatigue2019; 122: 256–272.
6.
RumbayanRWasherGA.Modeling of environmental effects on thermal detection of subsurface damage in concrete. J Nondestruct Eval2014; 25(4): 235–252.
7.
ShamFChenNLongL.Surface crack detection by flash thermography on concrete surface. Insight: Non-Destr Test Cond Monit2008; 50(5): 240–243.
8.
TranQHHanDKangC, et al. Effects of ambient temperature and relative humidity on subsurface defect detection in concrete structures by active thermal imaging. Sensors2017; 17(8): 1718.
9.
SelleckSFLandisENPetersonML, et al. Ultrasonic investigation of concrete with distributed damage. Mater J1998; 95(1): 27–36.
10.
SudaTTabataAKawakamiJ, et al. Development of an impact sound diagnosis system for tunnel concrete lining. In: ShirlawJNZhaoJKrishnanR (eds.) Tunnelling and underground space technology. Underground space for sustainable urban development. Proceedings of the 30th ITA-Aites World tunnel congress, Singapore, 22–27 May 2004, vol. 19, no. 4–5, pp. 328–329. https://www.nlb.gov.sg/biblio/12445230
11.
SonodaYFukuiY.A basic study on hammering tests of deteriorated concrete structures. Proceedings of 35th Conference on our world in concrete and structures, Singapore, 2010.
12.
InoueFDoiSIshizakiT, et al. Study on automated inspection robot and quantitative detection of outer tile wall exfoliation by wavelet analysis. In: Institute of Control, Robotics, and Systems, Institute of Electrical and Electronics Engineers, Korean Research Council. ICCAS 2010. Seaside, CA: IEEE, 2010, pp. 994–999. https://www.worldcat.org/title/1058325348
13.
TsunoKKishidaK.Evaluation of spalling of concrete pieces from tunnel lining employing joint shear model. Tunn Undergr Space Technol2020; 103: 103456.
14.
SonodaYWatanabeT.Quantitative evaluation on the defect distinction performance of the rotary hammering test. Kozo Kogaku Ronbunshu A (J Struct Eng A)2013; 59: 682–6922.
15.
GrigorescuSTrasneaBCociasT, et al. A survey of deep learning techniques for autonomous driving. J Field Robot2020; 37(3): 362–386.
LitjensGSánchezCITimofeevaN, et al. Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis. Sci Rep2016; 6(1): 1–11.
18.
LeCunY.Generalization and network design strategies. In: PfeiferRSchreterZFogelmanFSteelsL. (eds.) Connectionism in perspective. Switzerland: Elseiver, 1989, pp. 143–155.
19.
KrizhevskyASutskeverIHintonGE.Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst2012; 25: 1097–1105.
20.
BaoYTangZLiH, et al. Computer vision and deep learning–based data anomaly detection method for structural health monitoring. Struct Health Monit2019; 18(2): 401–421.
21.
TangZChenZBaoY, et al. Convolutional neural network-based data anomaly detection method using multiple information for structural health monitoring. Struct Control Health Monit2019; 26(1): e2296.
ChaYJWangZ.Unsupervised novelty detection-based structural damage localization using a density peaks-based fast clustering algorithm. Struct Health Monit2018; 17(2): 313–324.
24.
WangZChaYJ.Unsupervised deep learning approach using a deep auto-encoder with a one-class support vector machine to detect damage. Struct Health Monit2021; 20(1): 406–425.
25.
VincentPLarochelleHBengioY, et al. Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th international conference on Machine learning, Helsinki, Finland, 2008, pp.1096–1103. New York: Association for Computing Machinery.
26.
VincentPLarochelleHLajoieI, et al. Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J Mach Learn Res2010; 11(12): 3371–3408.
27.
ChowJKSuZWuJ, et al. Anomaly detection of defects on concrete structures with the convolutional autoencoder. Adv Eng Inform2020; 45: 101105.
28.
MasciJMeierUDCireşanD, et al. Stacked convolutional auto-encoders for hierarchical feature extraction. In: Proceedings of 21th international conference on artificial neural networks, Berlin, Heidelberg, 2011, vol. I, pp.52–59. Heidelberg: Springer.
29.
BloomfieldP.Fourier analysis of time series: an introduction. New York, NY: John Wiley & Sons, 2004.
