Abstract
For complex engineering systems, such as trains, planes, and offshore oil platforms, load spectra are cornerstone of their safety designs and fault diagnoses. We demonstrate in this study that well-orchestrated machine learning modeling, in combination with limited experimental data, can effectively reproduce the high-fidelity, history-dependent load spectra in critical sites of complex engineering systems, such as high-speed trains. To meet the need for in-service monitoring, we propose a segmentation and randomization strategy for long-duration historical data processing to improve the accuracy of our data-driven model for long-term load-time history prediction. Results showed the existence of an optimal length of subsequence, which is associated with the characteristic dissipation time of the dynamic system. Moreover, the data-driven model exhibits an excellent generalization capability to accurately predict the load spectra for different levels of passenger-dedicated lines. In brief, we pave the way, from data preprocessing, hyperparameter selection, to learning strategy, on how to capture the nonlinear responses of such a dynamic system, which may then provide a unifying framework that could enable the synergy of computation and in-field experiments to save orders of magnitude of expenses for the load spectrum monitoring of complex engineering structures in service and prevent catastrophic fatigue and fracture in those solids.
Similar content being viewed by others
References
Butler K T, Davies D W, Cartwright H, et al. Machine learning for molecular and materials science. Nature, 2018, 559: 547–555
Rovinelli A, Sangid M D, Proudhon H, et al. Predicting the 3D fatigue crack growth rate of small cracks using multimodal data via Bayesian networks: In-situ experiments and crystal plasticity simulations. J Mech Phys Solids, 2018, 115: 208–229
Tyulyukovskiy E, Huber N. Neural networks for tip correction of spherical indentation curves from bulk metals and thin metal films. J Mech Phys Solids, 2007, 55: 391–418
Shi Z, Tsymbalov E, Dao M, et al. Deep elastic strain engineering of bandgap through machine learning. Proc Natl Acad Sci USA, 2019, 116: 4117–4122
Lu L, Dao M, Kumar P, et al. Extraction of mechanical properties of materials through deep learning from instrumented indentation. Proc Natl Acad Sci USA, 2020, 117: 7052–7062
Wen J, Zou Q, Wei Y. Physics-driven machine learning model on temperature and time-dependent deformation in lithium metal and its finite element implementation. J Mech Phys Solids, 2021, 153: 104481
Xiong J, Zhang T Y, Shi S Q. Machine learning of mechanical properties of steels. Sci China Tech Sci, 2020, 63: 1247–1255
Cai S Z, Mao Z P, Wang Z C, et al. Physics-informed neural networks (PINNs) for fluid mechanics: A review. Acta Mech Sinica PRC, 2021, 37: 1–12
Raccuglia P, Elbert K C, Adler P D F, et al. Machine-learning-assisted materials discovery using failed experiments. Nature, 2016, 533: 73–76
Jha D, Choudhary K, Tavazza F, et al. Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning. Nat Commun, 2019, 10: 5316
Schmidt J, Marques M R G, Botti S, et al. Recent advances and applications of machine learning in solid-state materials science. npj Comput Mater, 2019, 5: 83
Sun W, Zheng Y, Yang K, et al. Machine learning-assisted molecular design and efficiency prediction for high-performance organic photovoltaic materials. Sci Adv, 2019, 5: eaay4275
Senior A W, Evans R, Jumper J, et al. Improved protein structure prediction using potentials from deep learning. Nature, 2020, 577: 706–710
Agius R, Brieghel C, Andersen M A, et al. Machine learning can identify newly diagnosed patients with CLL at high risk of infection. Nat Commun, 2020, 11: 363
Jiao W, Atwal G, Polak P, et al. A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns. Nat Commun, 2020, 11: 728
Segler M H S, Preuss M, Waller M P. Planning chemical syntheses with deep neural networks and symbolic AI. Nature, 2018, 555: 604–610
Biamonte J, Wittek P, Pancotti N, et al. Quantum machine learning. Nature, 2017, 549: 195–202
Rocchetto A, Aaronson S, Severini S, et al. Experimental learning of quantum states. Sci Adv, 2019, 5: eaau1946
Zhang L, Lin D Y, Wang H, et al. Active learning of uniformly accurate interatomic potentials for materials simulation. Phys Rev Mater, 2019, 3: 023804
Zhang Y, Wang H, Chen W, et al. DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models. Comput Phys Commun, 2020, 253: 107206
DeVries P M R, Thompson T B, Meade B J. Enabling large-scale viscoelastic calculations via neural network acceleration. Geophys Res Lett, 2017, 44: 2662–2669
Holtzman B K, Paté A, Paisley J, et al. Machine learning reveals cyclic changes in seismic source spectra in Geysers geothermal field. Sci Adv, 2018, 4: eaao2929
Reichstein M, Camps-Valls G, Stevens B, et al. Deep learning and process understanding for data-driven Earth system science. Nature, 2019, 566: 195–204
Kopp W, Monti R, Tamburrini A, et al. Deep learning for genomics using Janggu. Nat Commun, 2020, 11: 3488
Lei Y, Yang B, Jiang X, et al. Applications of machine learning to machine fault diagnosis: A review and roadmap. Mech Syst Signal Process, 2020, 138: 106587
Wang R X, Gao X, Gao J M, et al. An artificial immune and incremental learning inspired novel framework for performance pattern identification of complex electromechanical systems. Sci China Tech Sci, 2020, 63: 1–13
Martins T, Infante V, Sousa L, et al. Numerical and experimental study of aircraft structural health. Int J Fatigue, 2020, 132: 105348
Zuo H, Bi K, Hao H, et al. Influence of earthquake ground motion modelling on the dynamic responses of offshore wind turbines. Soil Dyn Earthq Eng, 2019, 121: 151–167
Zhou Y, Sun L. Effects of high winds on a long-span sea-crossing bridge based on structural health monitoring. J Wind Eng Indust Aero Dyn, 2018, 174: 260–268
Chen D, Xiao Q, Mou M, et al. Study on establishment of standardized load spectrum on bogie frames of high-speed trains. Acta Mech Sin, 2019, 35: 812–827
Anderson P W. More is different: Broken symmetry and the nature of the hierarchical structure of science. Science, 1972, 177: 393–396
Diamanti K, Soutis C. Structural health monitoring techniques for aircraft composite structures. Prog Aerosp Sci, 2010, 46: 342–352
Fujino Y, Siringoringo D M, Ikeda Y, et al. Research and implementations of structural monitoring for bridges and buildings in Japan. Engineering, 2019, 5: 1093–1119
Raissi M, Yazdani A, Karniadakis G E. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science, 2020, 367: 1026–1030
Zhang R, Chen Z, Chen S, et al. Deep long short-term memory networks for nonlinear structural seismic response prediction. Comput Struct, 2019, 220: 55–68
Zhang W J, Qin J, Mei F, et al. Short-term power load forecasting using integrated methods based on long short-term memory. Sci China Tech Sci, 2020, 63: 614–624
Jiang Y H, Yu Y F, Huang J Q, et al. Li-ion battery temperature estimation based on recurrent neural networks. Sci China Tech Sci, 2021, 64: 1335–1344
Wu P, Sun J, Chang X, et al. Data-driven reduced order model with temporal convolutional neural network. Comput Methods Appl Mech Eng, 2020, 360: 112766
Mozaffar M, Bostanabad R, Chen W, et al. Deep learning predicts path-dependent plasticity. Proc Natl Acad Sci USA, 2019, 116: 26414–26420
Gorji M B, Mozaffar M, Heidenreich J N, et al. On the potential of recurrent neural networks for modeling path dependent plasticity. J Mech Phys Solids, 2020, 143: 103972
Bergen K J, Johnson P A, de Hoop M V, et al. Machine learning for data-driven discovery in solid Earth geoscience. Science, 2019, 363: aau0323
Zhang S G. Study on testing and establishment method for the load spectrum of bogie frame for high-speed trains. Sci China Ser E, 2008, 51: 2142–2151
Ma S, Sun S, Wang B, et al. Estimating load spectra probability distributions of train bogie frames by the diffusion-based kernel density method. Int J Fatigue, 2020, 132: 105352
Yuan Z, Chen X, Ma L, et al. A segmented load spectrum model for high-speed trains and its inflection stress as an indicator for line quality. Int J Fatigue, 2021, 148: 106221
Klemenc J, Fajdiga M. Prediction of loading spectra under diverse operating conditions by a localised basis function neural network. Int J Fatigue, 2005, 27: 555–568
Yang Z, Song Z, Zhao X, et al. Time-domain extrapolation method for tractor drive shaft loads in stationary operating conditions. Bio Syst Eng, 2021, 210: 143–155
Zou Y, Zhang Y, Mao H. Fault diagnosis on the bearing of traction motor in high-speed trains based on deep learning. Alex Eng J, 2021, 60: 1209–1219
Chen J, Feng X, Jiang L, et al. State of charge estimation of lithiumion battery using denoising autoencoder and gated recurrent unit recurrent neural network. Energy, 2021, 227: 120451
Veeramsetty V, Reddy K R, Santhosh M, et al. Short-term electric power load forecasting using random forest and gated recurrent unit. Electr Eng, 2022, 104: 307–329
Patil S A, Raj L A, Singh B K. Prediction of IoT traffic using the gated recurrent unit neural network- (GRU-NN-) based predictive model. Secur Commun Netw, 2021, 2021: 1425732
Jordan M I, Mitchell T M. Machine learning: Trends, perspectives, and prospects. Science, 2015, 349: 255–260
Clough R W, Penzien J. Dynamics of Structures. 3rd ed. Berkeley: Computers & Structures Inc., 2003
Hochreiter S. Untersuchungen zu dynamischen neuronalen Netzen (in German). Dissertation for Master’s Degree. Munich: Technische Universität München, 1991
Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput, 1997, 9: 1735–1780
Cho K, Van Merrienboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). Doha: Association for Computational Linguistics, 2014. 1724–1734
Murdoch W J, Singh C, Kumbier K, et al. Definitions, methods, and applications in interpretable machine learning. Proc Natl Acad Sci USA, 2019, 116: 22071–22080
Chollet F. Keras Version: 2.3.0, 2019. https://keras.io. (accessed on 2020-06-28)
Kingma D P, Ba J. Adam: A method for stochastic optimization. ar-Xiv: 1412.6980
nCode 9.0. Hottinger Baldwin Messtechnik Company, Darmstadt, Germany, 2013
Amzallag C, Gerey J, Robert J, et al. Standardization of the rainflow counting method for fatigue analysis. Int J Fatigue, 1994, 16: 287–293
Miner M A. Cumulative damage in fatigue. J Appl Mech, 1945, 12: A159–A164
Blalock D, Ortiz J J G, Frankle J, et al. What is the state of neural network pruning? arXiv: 2003.03033
Yu B, Kumbier K. Veridical data science. Proc Natl Acad Sci USA, 2020, 117: 3920–3929
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported by the Basic Science Center of the National Natural Science Foundation of China for “Multiscale Problems in Nonlinear Mechanics” (Grant No. 11988102), the National Key Research and Development Program of China (Grant Nos. 2017YFB0202800 and 2016YFB1200602), the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDB22020200), and the Science Challenge Project (Grant No. TZ2018002).
Supporting Information
The supporting information is available online at https://tech.scichina.com and https://link.springer.com. The supporting materials are published as submitted, without typesetting or editing. The responsibility for scientific accuracy and content remains entirely with the authors.
Rights and permissions
About this article
Cite this article
Chen, X., Yuan, Z., Li, Q. et al. A computational method for the load spectra of large-scale structures with a data-driven learning algorithm. Sci. China Technol. Sci. 66, 141–154 (2023). https://doi.org/10.1007/s11431-021-2068-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11431-021-2068-8