Monitoring the state of concrete structures and assessing their performance are significant tasks for civil engineering. This study proposes a combined technique of novel concrete implantable bar (CIB) transducers and enhanced long short-term memory (LSTM) networks for monitoring and assessing reinforced concrete (RC) beam state over the whole loading process. The CIB can be installed on the inspected structure in an implantable manner. It contains an array of piezoceramic sensing units that can generate and receive ultrasonic waves from a concrete medium. To enhance the LSTM network’s ability to learn very long-series data, a time-shift energy (TSE) strategy was developed. Compared with another existing convolutional neural network (CNN)–LSTM network, the proposed TSE–LSTM network is highlighted to fully consider the ultrasound propagation characteristics in concrete when extracting sample features, instead of using simple convolution operation. A numerical study was conducted to investigate the sensitivity of the TSE feature to different concrete damage levels through mesoscale finite element models. The results provided the best parameter settings of the TSE. Eventually, to validate the feasibility of the proposed technique, an RC beam four-point bending test was carried out, in which two CIBs were implanted into the specimen for emitting and collecting ultrasonic waves in different damage states to establish a dataset. Two schemes including a classification model for predicting RC beam stress stages and another regression model for predicting the carried forces were separately investigated. The experimental results showed that the TSE–LSTM networks can successfully predict the signature of three stages of RC beams and can, in general, predict their carried forces. The comparison to the results obtained by CNN–LSTM networks further highlighted the stability and accuracy of the proposed one in learning long ultrasound series. The combined technique of CIB transducers and TSE–LSTM networks shows a promising application for monitoring and assessing RC structures.
TsangouriEKaraiskosGAggelisDG, et al. Crack sealing and damage recovery monitoring of a concrete healing system using embedded piezoelectric transducers. Struct Health Monit Int J2015; 14(5): 462–474.
2.
SunXTZhangMHGaoWH, et al. A novel method for steel bar all-stage pitting corrosion monitoring using the feature-level fusion of ultrasonic direct waves and coda waves. Struct Health Monit Int J2023; 22(1): 714–729.
3.
YangXZhangMHChenHB, et al. A novel damage assessment method for RC beam using force-hammer excitation and piezoelectric sensing technology. Struct Control Health Monit2023; 2023: 4365213.
4.
TrtnikGTurkGKavcicF, et al. Possibilities of using the ultrasonic wave transmission method to estimate initial setting time of cement paste. Cem Concr Res2008; 38(11): 1336–1342.
5.
OzkanIYaylaZ.Evaluation of correlation between physical properties and ultrasonic pulse velocity of fired clay samples. Ultrasonics2016; 66: 4–10.
6.
Solis-CarcanoRMorenoEI.Evaluation of concrete made with crushed limestone aggregate based on ultrasonic pulse velocity. Constr Build Mater2008; 22(6): 1225–1231.
7.
JiangJLZhangDWGongFY, et al. Prediction of ultrasonic pulse velocity for cement, mortar, and concrete through a multiscale homogenization approach. Materials2022; 15(9): 3241.
8.
TrtnikGKavcicFTurkG.Prediction of concrete strength using ultrasonic pulse velocity and artificial neural networks. Ultrasonics2009; 49(1): 53–60.
9.
HongSUChoYSKimSH, et al. Estimation of compressive strength of concrete structures using the ultrasonic pulse velocity method and spectral analysis of surface wave method. Mater Res Innovations2015; 19: 1289–1294.
10.
DelgadilloHHKernBLoenderslootR, et al. A methodology based on pulse-velocity measurements to quantify the chemical degradation levels in thin mortar specimens. J Nondestr Eval2018; 37(4): 79.
11.
HwangEKimGChoeG, et al. Evaluation of concrete degradation depending on heating conditions by ultrasonic pulse velocity. Constr Build Mater2018; 171: 511–520.
12.
LiuDWGongCTangY, et al. Evaluation of corrosion damage in sulfate-attacked concrete by CT, ultrasonic pulse velocity testing and AHP methods. Sensors2022; 22(8): 3037.
13.
MukhtiJARoblesKPVLeeKH, et al. Evaluation of early concrete damage caused by chloride-induced steel corrosion using a deep learning approach based on RNN for ultrasonic pulse waves. Materials2023; 16(9): 3502.
14.
ZhutovskySKovlerK.Application of ultrasonic pulse velocity for assessment of thermal expansion coefficient of concrete at early age. Mater Struct2017; 50(1): 5.
15.
YangCLChenJ.Fully noncontact nonlinear ultrasonic characterization of thermal damage in concrete and correlation with microscopic evidence of material cracking. Cem Concr Res2019; 123: 105797.
16.
ChoYSHanSWJangHS, et al. Estimating crack depth of concrete structures using ultrasonic pulse velocity. In: International conference on function materials and nanotechnology (FMN 2012), Zhengzhou, China, 2012, pp. 546–+.
17.
YangJWSuYYHeY, et al. Machine learning-enabled resolution-lossless tomography for composite structures with a restricted sensing capability. Ultrasonics2022; 125: 106801.
18.
SchabowiczK.Ultrasonic tomography - the latest nondestructive technique for testing concrete members - description, test methodology, application example. Arch Civ Mech Eng2014; 14(2): 295–303.
19.
KeeSHZhuJY.Using piezoelectric sensors for ultrasonic pulse velocity measurements in concrete. Smart Mater Struct2013; 22(11): 115016.
20.
YangZQChenQJLiXQ, et al. Crack identification in concrete structures using implantable sensors. Measurement2022; 202: 111780.
21.
ZhangMXLiMCZhangJR, et al. Onset detection of ultrasonic signals for the testing of concrete foundation piles by coupled continuous wavelet transform and machine learning algorithms. Adv Eng Inform2020; 43: 101034.
22.
YangZQGaoWHChenL, et al. A novel electromechanical impedance-based method for non-destructive evaluation of concrete fiber content. Constr Build Mater2022; 351: 128972.
23.
KarbassiAMohebiBRezaeeS, et al. Damage prediction for regular reinforced concrete buildings using the decision tree algorithm. Comput Struct2014; 130: 46–56.
24.
PrayogoDChengMYWuYW, et al. Combining machine learning models via adaptive ensemble weighting for prediction of shear capacity of reinforced-concrete deep beams. Eng Comput2020; 36(3): 1135–1153.
25.
HalbouniAGunawanTSHabaebiMH, et al. Machine learning and deep learning approaches for cybersecurity: a review. IEEE Access2022; 10: 19572–19585.
26.
FanLZhangTZhaoX, et al. Deep topology network: a framework based on feedback adjustment learning rate for image classification. Adv Eng Inform2019; 42: 100935.
27.
SingaravelSSuykensJGeyerP.Deep convolutional learning for general early design stage prediction models. Adv Eng Inform2019; 42: 100982.
28.
DasAKLeungCKYWanKT.Application of deep convolutional neural networks for automated and rapid identification and computation of crack statistics of thin cracks in strain hardening cementitious composites (SHCCs). Cem Concr Compos2021; 122: 104159.
29.
OhBKParkYParkHS.Seismic response prediction method for building structures using convolutional neural network. Struct Control Health Monit2020; 27(5): e2519.
30.
ChowJKSuZWuJ, et al. Anomaly detection of defects on concrete structures with the convolutional autoencoder. Adv Eng Inform2020; 45: 101105.
31.
LaxmanKCTabassumNAiL, et al. Automated crack detection and crack depth prediction for reinforced concrete structures using deep learning. Constr Build Mater2023; 370: 130709.
32.
WangWShiPDengL, et al. Residual strength evaluation of corroded textile-reinforced concrete by the deep learning-based method. Materials2020; 13(14): 3226.
33.
RazaviSVJumaatMZEi-ShafieAH, et al. Load-deflection analysis prediction of CFRP strengthened RC slab using RNN. Adv Concr Constr2015; 3(2): 91–102.
34.
FeiYFLiaoWJZhangS, et al. Integrated schematic design method for shear wall structures: a practical application of generative adversarial networks. Buildings2022; 12(9): 1295.
35.
HaileMAZhuEHsuC, et al. Deep machine learning for detection of acoustic wave reflections. Struct Health Monit Int J2020; 19(5): 1340–1350.
36.
CaiJNZhangYXCaiHB.Two-step long short-term memory method for identifying construction activities through positional and attentional cues. Autom Constr2019; 106: 102886.
37.
RanjbarIToufighV.Deep long short-term memory (LSTM) networks for ultrasonic-based distributed damage assessment in concrete. Cem Concr Res2022; 162: 107003.
38.
Nguyen-LeDHTaoQBNguyenVH, et al. A data-driven approach based on long short-term memory and hidden markov model for crack propagation prediction. Eng Fract Mechs2020; 235: 107085.
39.
YessoufouFZhuJS.Classification and regression-based convolutional neural network and long short-term memory configuration for bridge damage identification using long-term monitoring vibration data. Struct Health Monit Int J2023; 22(6): 4027–4054.
40.
YangDSGuCSZhuYT, et al. A concrete dam deformation prediction method based on LSTM with attention mechanism. IEEE Access2020; 8: 185177–185186.
41.
ZhangYZhongWLiYL, et al. A deep learning prediction model of DenseNet-LSTM for concrete gravity dam deformation based on feature selection. Eng Struct2023; 295: 116827.
42.
ZhangQYBarriKBabanajadSK, et al. Real-time detection of cracks on concrete bridge decks using deep learning in the frequency domain. Engineering2021; 7(12): 1786–1796.
43.
YueZXDingYLZhaoHW, et al. Mechanics-guided optimization of an LSTM network for real-time modeling of temperature-induced deflection of a cable-stayed bridge. Eng Struct2022; 252: 113619.
44.
AhmedBMangalathuSJeonJS.Seismic damage state predictions of reinforced concrete structures using stacked long short-term memory neural networks. J Build Eng2022; 46: 103737.
GersFASchmidhuberJCumminsF.Learning to forget: continual prediction with LSTM. Neural Comput2000; 12(10): 2451–2471.
47.
QianHJLiYYangJF, et al. Image-based microstructure classification of mortar and paste using convolutional neural networks and transfer learning. Cem Concr Compos2022; 129: 104496.
48.
YuanCWangSYQiYZ, et al. Automated structural bolt looseness detection using deep learning-based prediction model. Struct Control Health Monit2022; 29(3): e2899.
49.
KongQZRobertRHSilvaP, et al. Cyclic crack monitoring of a reinforced concrete column under simulated pseudo-dynamic loading using piezoceramic-based smart aggregates. Appl Sci2006; 6(11): 341.
50.
SiYZhangZSKongCQ, et al. Looseness condition feature extraction of viscoelastic sandwich structure using dual-tree complex wavelet packet-based deep autoencoder network. Struct Health Monit Int J2020; 19(3): 873–884.
51.
YangZQChenQJZhangMH, et al. Time-reversal imaging for concrete hole defect based on implantable sensing technology. Mech Syst Signal Process2023; 186: 109903.
52.
XuBChenHBMoYL, et al. Dominance of debonding defect of CFST on PZT sensor response considering the meso-scale structure of concrete with multi-scale simulation. Mech Syst Signal Process2018; 107: 515–528.
53.
RamosASimoesJA.Tetrahedral versus hexahedral finite elements in numerical modelling of the proximal femur. Med Eng Phys2006; 28(9): 916–924.
54.
AlsumaidaeeYAMYawCTKohSP, et al. Detection of corona faults in switchgear by using 1D-CNN, LSTM, and 1D-CNN-LSTM methods. Sensors2023; 23(6): 3108.
