The utilization of data-driven models, for example, the long short-term memory neural network (LSTM) model, has emerged as a potential structural response prediction approach for tackling the issues with the transmission and storage of long-term monitoring data. However, the prediction error of the LSTM models may increase significantly as the total prediction length increases, limiting the applicability of LSTM in response prediction. To tackle this challenge, this study proposes a hybrid prediction method, namely LSTM + compressive sensing (LSTM + CS), which combines LSTM and CS to improve prediction accuracy. CS embeds physical information into the prediction process, thereby the prediction divergence can be well mitigated. Simulated responses of a four-degree-of-freedom system and real-world responses of the Canton Tower are utilized for verification purposes, demonstrating that LSTM + CS can achieve a high prediction performance with limited data and minimal costs. The proposed method has the potential to be an effective response prediction tool to reduce the data storage requirement.
AbdulkaremMSamsudinKRokhaniFZ, et al. Wireless sensor network for structural health monitoring: a contemporary review of technologies, challenges, and future direction. Struct Health Monit2020; 19(3): 693–735.
2.
MishraMLourençoPBRamanaGV.Structural health monitoring of civil engineering structures by using the internet of things: a review. J Build Eng2022; 48: 103954.
3.
LiangJKatoBWangY.Constructing simplified models for dynamic analysis of monopile-supported offshore wind turbines. Ocean Eng2023; 271: 113785.
4.
LinJLChuangMC.Simplified nonlinear modeling for estimating the seismic response of buildings. Eng Struct2023; 279: 115590.
5.
LombaertGDegrandeGKogutJ, et al. The experimental validation of a numerical model for the prediction of railway induced vibrations. J Sound Vibr2006; 297(3–5): 512–535.
6.
GuptaSHusseinMFMDegrandeG, et al. A comparison of two numerical models for the prediction of vibrations from underground railway traffic. Soil Dyn Earthquake Eng2007; 27(7): 608–624.
7.
LiuWLaiZBacsaK, et al. Neural extended Kalman filters for learning and predicting dynamics of structural systems. Struct Health Monit2024; 23(2): 1037–1052.
8.
LiYBaoTGaoZ, et al. A new dam structural response estimation paradigm powered by deep learning and transfer learning techniques. Struct Health Monit2022; 21(3): 770–787.
9.
LiHWNiYQWangYW, et al. Continuous-time state-space neural network and its application in modeling of forced-vibration systems. In: Structural health monitoring2023.
10.
LiHWNiYQWangYW, et al. Modeling of forced-vibration systems using continuous-time state-space neural network. Eng Struct2024; 302: 117329.
11.
LiXZhangW.Physics-informed deep learning model in wind turbine response prediction. Renewable Energy2022; 185: 932–944.
12.
LystadTMFenerciAØisethO.Full long-term extreme buffeting response calculations using sequential Gaussian process surrogate modeling. Eng Struct2023; 292: 116495.
13.
Salcedo-SanzSRojo-ÁlvarezJLMartínez-RamónM, et al. Support vector machines in engineering: an overview. Wiley Int Rev: Data Min Knowl Discovery2014; 4(3): 234–267.
14.
YoungPCRattoM.Statistical emulation of large linear dynamic models. Technometrics2011; 53(1): 29–43.
15.
LeiXFengRDongY, et al. Bayesian-optimized interpretable surrogate model for seismic demand prediction of urban highway bridges. Eng Struct2024; 301: 117307.
16.
ParidaSSBoseSButcherM, et al. SVD enabled data augmentation for machine learning based surrogate modeling of non-linear structures. Eng Struct2023; 280: 115600.
17.
LuKZhangKZhangH, et al. A review of model order reduction methods for large-scale structure systems. Shock Vibr2021; 2021: 1–9.
18.
AllaAKutzJN.Nonlinear model order reduction via dynamic mode decomposition. SIAM J Sci Comput2017; 39(5): B778–B796.
19.
LadevezePChamoinL.On the verification of model reduction methods based on the proper generalized decomposition. Comput Methods Appl Mech Eng2011; 200(23–24): 2032–2047.
20.
HuangYBeckJLWuS, et al. Bayesian compressive sensing for approximately sparse signals and application to structural health monitoring signals for data loss recovery. Probab Eng Mech2016; 46: 62–79.
21.
HuangYBeckJLWuS, et al. Robust Bayesian compressive sensing for signals in structural health monitoring. Comput Aid Civ Infrastruct Eng2014; 29(3): 160–179.
22.
SherstinskyA.Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network. Phys D2020; 404: 132306.
23.
GersFASchmidhuberJCumminsF.Learning to forget: continual prediction with LSTM. Neural Comput2000; 12(10): 2451–2471.
24.
YuYSiXHuC, et al. A review of recurrent neural networks: LSTM cells and network architectures. Neural Comput2019; 31(7): 1235–1270.
25.
XueYJiangJHongL.A LSTM based prediction model for nonlinear dynamical systems with chaotic itinerancy. Int J Dyn Control2020; 8: 1117–1128.
26.
ZhangRMengLMaoZ, et al. Spatiotemporal deep learning for bridge response forecasting. J Struct Eng2021; 147(6): 04021070.
27.
LiBSpenceSM.Metamodeling through deep learning of high-dimensional dynamic nonlinear systems driven by general stochastic excitation. J Struct Eng2022; 148(11): 04022186.
28.
RaniMDhokSBDeshmukhRB.A systematic review of compressive sensing: concepts, implementations and applications. IEEE Access2018; 6: 4875–4894.
29.
BaoYTangZLiH.Compressive-sensing data reconstruction for structural health monitoring: a machine-learning approach. Struct Health Monit2020; 19(1): 293–304.
30.
KangJRenWXXieYL, et al. An enhanced method to reduce reconstruction error of compressed sensing for structure vibration signals. Mech Syst Signal Process2023; 183: 109585.
31.
WangYHaoH.Damage identification scheme based on compressive sensing. J Comput Civil Eng2015; 29(2): 04014037.
32.
ZhouJBenceKWangY.Operational modal analysis with compressed measurements based on prior information. Measurement2023; 211: 112644.
33.
ZhouJYuSLiH, et al. Automated operational modal analysis for civil engineering structures with compressed measurements. Measurement2023; 223: 113772.
34.
HanZZhaoJLeungH, et al. A review of deep learning models for time series prediction. IEEE Sens J2019; 21(6): 7833–7848.
35.
ChandraRGoyalSGuptaR.Evaluation of deep learning models for multi-step ahead time series prediction. IEEE Access2021; 9: 83105–83123.
36.
LindemannBMüllerTVietzH, et al. A survey on long short-term memory networks for time series prediction. Procedia CIRP2021; 99: 650–655.
37.
XuZChenJShenJ, et al. Recursive long short-term memory network for predicting nonlinear structural seismic response. Eng Struct2022; 250: 113406.
38.
JamesGHCarneTGLaufferJP.The natural excitation technique (NExT) for modal parameter extraction from operating wind turbines. Albuquerque, NM, USA: Sandia National Labs, 1993.
39.
TroppJAGilbertAC.Signal recovery from random measurements via orthogonal matching pursuit. IEEE Trans Inform Theory2007; 53(12): 4655–4666.
40.
KingmaDPBaJ.Adam: a method for stochastic optimization. ArXiv Preprint 2014; arXiv:1412.6980.
41.
NiYQXiaYLiaoWY, et al. Technology innovation in developing the structural health monitoring system for Guangzhou New TV Tower. Struct Control Health Monit2009; 16(1): 73–98.
42.
WanHPNiYQ.Bayesian modeling approach for forecast of structural stress response using structural health monitoring data. J Struct Eng2018; 144(9): 04018130.
43.
WangYWZhangCNiYQ, et al. Bayesian probabilistic assessment of occupant comfort of high-rise structures based on structural health monitoring data. Mech Syst Signal Process2022; 163: 108147.
44.
ZhouJLiHWWangYW, et al. Comparative study of earthquake effects on the Canton Tower based on full-scale measurements. J Build Eng2024; 96: 110430.
