Objective Hydraulic concrete produces acoustic emission phenomena due to cracking and other damage. It is important to quickly and accurately locate the damage source based on acoustic emission signals for real-time monitoring of the health status of hydraulic buildings. The traditional iterative localization method is greatly affected by the initial value of iteration and the iteration process. An inappropriate initial value of iteration often leads to unstable or divergent iterations, ultimately resulting in poor localization accuracy. Therefore, the selection of the initial value of iteration is critical in the traditional iterative localization method. In addition, traditional iterative localization methods are heavily influenced by the number of sensors and environmental noise, which reduces localization stability and efficiency. The rapid development of deep learning in recent years provides new approaches for acoustic emission localization. Deep learning demonstrates strong feature extraction capability and generalization ability. In response to the limitations of traditional iterative localization methods, a convolutional neural network-based acoustic emission localization model is constructed, which improves the efficiency, stability, and accuracy of acoustic emission localization to a certain extent. Methods The absolute propagation time of acoustic emission was generally difficult to obtain, but the arrival time difference between the sensors contained sufficient information that was utilized to localize the acoustic emission source position. This study constructed a convolutional neural network-based acoustic emission localization model using a cylindrical concrete specimen as the experimental object, with the 3D coordinates of eight sensors and the propagation time difference as input, and the 3D coordinate position of the acoustic emission source as output. In addition, the effect of the number of convolutional layers and the size of the convolutional kernel on the localization accuracy was analyzed, and the optimal convolutional neural network structure was obtained. At the same time, the traditional iterative localization method was compared, and the advantages of the constructed localization model in terms of localization accuracy and efficiency were analyzed to verify the localization performance of the constructed localization model. Results and Discussions For the case of 8 sensors, the optimal number of convolutional layers was 4, and the optimal convolutional kernel size was 3×1. In the X, Y, and Z directions, the root mean square errors (RMSE) of the localization model were 0.865 2, 0.826 6, and 0.722 1 mm respectively, the mean absolute errors (MAE) were 0.532 2, 0.617 3, and 0.473 3 mm respectively, the mean absolute percentage errors (MAPE) were 0.222 0, 0.510 1, and 0.051 0 respectively, and the coefficients of determination (R2) were 0.994 2, 0.993 8, and 0.999 3 respectively, which were close to 1. Most of the localization errors were distributed near 0, which basically conformed to the standard normal distribution. The localization accuracy of the localization model in the depth direction was higher than that in the horizontal direction. Compared to the traditional iterative localization method, the localization efficiency was stable, and it showed clear advantages in processing many localization tasks, while the localization error was reduced by about 5%. The damage location detected by the localization model basically matched the real crack location. Conclusions The proposed localization model shows good localization efficiency and accuracy. Compared to the traditional iterative localization method, it is not influenced by the initial iteration value or the iterative process, and it exhibits the advantages of stability and high efficiency. In addition, it maintains stable performance for new data points and shows strong applicability, making it a reliable reference for early warning of damage evolution based on nondestructive testing, with potential application to the damage detection of other materials in the future.
KangYumei, LiuJianpo, LiHaibin,et al.An AE source location combination algorithm based on least square method[J].Journal of Northeastern University(Natural Science),2010,31(11):1648‒1651.
WangZonglian, RenHuilan, NingJianguo.Acoustic emission source location based on wavelet transform denois-ing[J].Journal of Vibration and Shock,2018,37(4):226‒232.
LiJian, GaoYongtao, XieYuling,et al.Improvement of microseism locating based on simplex method without velocity measuring[J].Chinese Journal of Rock Mechanics and Engineering,2014,33(7):1336‒1346.
HuangXiaohong, SunGuoqing, ZhangKaiyue.Localisati-on of Geiger acoustic emission source based on all-phase analysis and several times cross-correlation[J].Rock and Soil Mechanics,2018, 39(4):1362‒1368.
ZhangYu, LiuJiacheng, FengShu, et al.Research on acoustic emission source localization method for discontinuous structure with holes[J].Chinese Journal of Scientific Instrument,2023,44(11):282‒289.
HuangLinqi, WuXin, LiXibing,et al.Influence of sensor array on MS/AE source location accuracy in rock mass[J].Transactions of Nonferrous Metals Society of China,2023,33(1):254‒274.
JiangRuochen, XuNuwen, DaiFeng,et al.Research on microseismic location based on fast marching upwind linear interpolation method[J].Rock and Soil Mechanics,2019,40(9):3697‒3708.
DongLongjun, LiXibing, MaJu,et al.Three-dimensional analytical comprehensive solutions for acoustic emission/microseismic sources of unknown velocity system[J].Chinese Journal of Rock Mechanics and Engineering,2017,36(1):186‒197.
JinguoLü, JiangYaodong, ZhaoYixin,et al.Study of microseismic positioning based on steady simulated annealing-simplex hybrid algorithm[J].Rock and Soil Mechanics,2013,34(8):2195‒2203.
LiYuanhui, ChenZhiyang, XuShida.Influence of wave velocity range on accuracy of PSO location algorithm[J].Journal of Northeastern University(Natural Science),2021,42(4):561‒566.
YaoKefu, SuHuaizhi, YangLifu,et al.Pattern recognition combination model for locating damage in concrete faced rockfill dams using acoustic emission and its experimental verification[J].Journal of Hydroelectric Engineering,2021,40(7):131‒140.
XuXiaoyang, SuHuaizhi, YanXiaoqun,et al.A combined locating method of acoustic emission sources for rockfill dam face slab damage based on wave velocity correction[J].Advances in Science and Technology of Water Resources,2023,43(4):86‒91.
XiaoXiaochun, DingZhen, DingXin,et al.Joint positioning method of P-wave arrival time picking based analysis-genetic algorithm and its application[J].Rock and Soil Mechanics,2024,45(7):2195‒2207.
ZhangXiaoping, ZhuHangkai, LiuQuansheng,et al.Research on microseismic event locating in layered rock masses based on Snell's law and Cuckoo search algorithm[J].Chinese Journal of Rock Mechanics and Engineering,2021,40(7):1383‒1391.
EbrahimkhanlouA, DubucB, SalamoneS.A generalizable deep learning framework for localizing and characterizing acoustic emission sources in riveted metallic panels[J].Mechanical Systems and Signal Processing,2019,130:248‒272. doi:10.1016/j.ymssp.2019.04.050
[40]
YangLi, XuFeiyun.A novel acoustic emission sources localization and identification method in metallic plates based on stacked denoising autoencoders[J].IEEE Access,2020,8:141123‒141142. doi:10.1109/access.2020.3012521
[41]
EbrahimkhanlouA, SalamoneS.Single-sensor acoustic em-ission source localization in plate-like structures using deep learning[J].Aerospace,2018,5(2):50. doi:10.3390/aerospace5020050
[42]
HesserD F, KocurG K, MarkertB.Active source localization in wave guides based on machine learning[J].Ultrasonics,2020,106:106144. doi:10.1016/j.ultras.2020.106144
[43]
ZhaoQi, GlaserS D.Relocating acoustic emission in rocks with unknown velocity structure with machine learning[J].Rock Mechanics and Rock Engineering,2020,53(5):2053‒2061. doi:10.1007/s00603-019-02028-8
[44]
HuangXuhui, HanMing, DengYiming.A hybrid GAN-inception deep learning approach for enhanced coordinate-based acoustic emission source localization[J].Applied Sciences,2024,14(19):8811. doi:10.3390/app14198811
[45]
ZhaoZhimin, ChenNianzhong.Spatial-temporal graph convolutional networks (STGCN) based method for localizing acoustic emission sources in composite panels[J].Composite Structures,2023,323:117496. doi:10.1016/j.compstruct.2023.117496
[46]
HaileM A, ZhuE, HsuC,et al.Deep machine learning for detection of acoustic wave reflections[J].Structural Health Monitoring,2020,19(5):1340‒1350. doi:10.1177/1475921719881642
[47]
ChenNianzhong, ZhaoZhimin, LinLin.A hybrid deep learning method for AE source localization for heterostructure of wind turbine blades[J].Marine Structures,2024,94:103562. doi:10.1016/j.marstruc.2023.103562
[48]
ZhangLijun, LiKewei, WangHang,et al.MFC‒PINN:A method to improve the accuracy and robustness of acoustic emission source planar localization[J].Measurement,2024,235:114995. doi:10.1016/j.measurement.2024.114995
[49]
WuXin, ZhaoHongxia, LuoXiaoyu,et al.Study on acoustic emission event source location of rock damage based on BP neural network[J].Journal of Safety Science and Technology,2021,17(8):36‒42.
ChenYangkang, SaadO M, SavvaidisA,et al.3D microseismic monitoring using machine learning[J].Journal of Geophysical Research:Solid Earth,2022,127(3):e2021JB023842. doi:10.1029/2021jb023842
[52]
ChenJie, ChenZiyang, PuYuanyuan.Acoustic emission sour-ce localization in rocks based on spectral analysis and convolutional neural network[J].Chinese Journal of Rock Mechanics and Engineering,2022,41(Supp2):3271‒3271.
van den Ende, AmpueroJ P.Automated seismic source characterisation using deep graph neural networks[J].Geophysical Research Letters,2020,47(17):e2020GL088690. doi:10.1029/2020gl088690
[55]
ZhouYubao, LiangMinfei, YueXinling.Deep residual lea-rning for acoustic emission source localization in A steel-concrete composite slab[J].Construction and Building Materials,2024,411:134220. doi:10.1016/j.conbuildmat.2023.134220
[56]
ShenJingshi, ZengXiaodong, JiangMingshun.Acoustic emission location method research based on FBG Network and BP Neural Network[J].Journal of Vibration,Measurement & Diagnosis,2018,38(4):816‒820.
ChenBingrui, FengXiating, FuQiqing,et al.Integration and high precision intelligence microseismic monitoring technology and its application in deep rock engineering[J].Rock and Soil Mechanics,2020,41(7):2422‒2431.
LinFeng, LiShulin, XueYunliang,et al.Microseismic sources location methods based on different initial values[J].Chinese Journal of Rock Mechanics and Engineering,2010,29(5):996‒1002.
ZhangHaijiang, NadeauR M, ToksozM N.Locating nonvolcanic tremors beneath the San Andreas Fault using a station-pair double-difference location method[J].Geoph-ysical Research Letters,2010,37(13):2010GL043577. doi:10.1029/2010gl043577
[63]
KhanA, SohailA, ZahooraU,et al.A survey of the recent architectures of deep convolutional neural networks[J].Artificial Intelligence Review,2020,53(8):5455‒5516. doi:10.1007/s10462-020-09825-6
[64]
KiranyazS, AvciO, AbdeljaberO,et al.1D convolutional neural networks and applications:A survey[J].Mechanical Systems and Signal Processing,2021,151:107398. doi:10.1016/j.ymssp.2020.107398
[65]
VyV, LeeY, BakJ,et al.Damage localization using acoustic emission sensors via convolutional neural network and continuous wavelet transform[J].Mechanical Systems and Signal Processing,2023,204:110831. doi:10.1016/j.ymssp.2023.110831
[66]
GauderD, BiehlerM, GölzJ,et al.In-process acoustic pore detection in milling using deep learning[J].CIRP Journal of Manufacturing Science and Technology,2022,37:125‒133. doi:10.1016/j.cirpj.2022.01.008
[67]
DonatiG, ZonziniF, De MarchiL.Tiny deep learning architectures enabling sensor-near acoustic data processing and defect localization[J].Computers,2023,12(7):129. doi:10.3390/computers12070129