Abstract
Coupled hydromechanical finite-element modeling of granular soils, taking into account internal erosion, is computationally prohibitive. Alternative data-driven approaches require large data sets for training and often provide poor generalization ability. To overcome these issues, this study proposed a physics-informed multifidelity residual neural network (PI-MR-NN) modeling strategy. The model was first trained using low-fidelity data to focus on capturing the main underpinning physical laws. Subsequent training on sparser high-fidelity data was then used to calibrate and refine the model. Physical constraints, e.g., boundary conditions, were incorporated through modifications to the loss functions. Feedforward and long short-term memory neural networks were considered as the baseline algorithms for training models. The PI-MR-NN was first used to reproduce synthetic results generated by the soil constitutive model SIMSAND and a published internal erosion model. The developed data-driven model was then applied to simulate the breach of a practical dike-on-foundation case and to predict its temporal responses. All results indicated that the hydromechanical response of porous media can be accurately captured using the proposed PI-MR-NN model. The novel training strategy mitigates the dependency of model performance on the training data set and architecture of the neural network, and the use of physical constraints improves training efficiency and enhances the model’s predictive robustness.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
All data that support the findings of this study are available from the corresponding author upon reasonable request.
Acknowledgments
This research was financially supported by the Research Grants Council (RGC) of Hong Kong Special Administrative Region Government (HKSARG) of China (Grant Nos. 15209119 and 15220221). The last author is supported by the Royal Academy of Engineering under the Research Fellowship Scheme.
References
Alboresha, R. 2016. “Evaluation of the impact of a cavity upon an earth dike (analytical and numerical approaches): Application to the Val d’Orléans area (France).” Ph.D. thesis, Université de Lorraine. https://hal.univ-orraine.fr/tel-01752229/document.
Bahri, Y., J. Kadmon, J. Pennington, S. S. Schoenholz, J. Sohl-Dickstein, and S. Ganguli. 2020. “Statistical mechanics of deep learning.” Ann. Rev. Condens. Matter Phys. 11 (1): 501–528. https://doi.org/10.1146/annurev-conmatphys-031119-050745.
Baiges, J., R. Codina, I. Castañar, and E. Castillo. 2019. “A finite element reduced-order model based on adaptive mesh refinement and artificial neural networks.” Int. J. Numer. Methods Eng. 121 (4): 588–601. https://doi.org/10.1002/nme.6235.
Baydin, A. G., B. A. Pearlmutter, A. A. Radul, and J. M. Siskind. 2018. “Automatic differentiation in machine learning: A survey.” J. Mach. Learn. Res. 18 (1): 5595–5637.
Chen, R. P., P. Zhang, X. Kang, Z. Q. Zhong, Y. Liu, and H. N. Wu. 2019. “Prediction of maximum surface settlement caused by EPB shield tunneling with ANN methods.” Soils Found. 59 (2): 284–295. https://doi.org/10.1016/j.sandf.2018.11.005.
Courant, R., K. Friedrichs, and H. Lewy. 1928. “Über die partiellen Differenzengleichungen der mathematischen Physik.” Math. Ann. 100 (1): 32–74. https://doi.org/10.1007/BF01448839.
DiazDelaO, F. A., and S. Adhikari. 2012. “Bayesian assimilation of multi-fidelity finite element models.” Comput. Struct. 92–93: 206–215.
Fell, R., C. F. Wan, J. Cyganiewicz, and M. Foster. 2003. “Time for development of internal erosion and piping in embankment dams.” J. Geotech. Geoenviron. Eng. 129 (4): 307–314. https://doi.org/10.1061/(ASCE)1090-0241(2003)129:4(307).
Fernández-Godino, M. G., C. Park, N. H. Kim, and R. T. Haftka. 2017. “Review of multi-fidelity models.” Preprint, submitted September 23, 2016. http://arxiv.org/abs/1609.07196v07193.
Fuchs, A., Y. Heider, K. Wang, W. Sun, and M. Kaliske. 2021. “DNN2: A hyper-parameter reinforcement learning game for self-design of neural network based elasto-plastic constitutive descriptions.” Comput. Struct. 249 (Jun): 106505. https://doi.org/10.1016/j.compstruc.2021.106505.
Gal, Y., and Z. Ghahramani. 2015. “Dropout as a Bayesian approximation: Representing model uncertainty in deep learning.” Preprint, submitted June 6, 2015. http://arxiv.org/abs/arXiv:1506.02142.
Graves, A. 2011. “Practical variational inference for neural networks.” In Proc., 24th Int. Conf. on Neural Information Processing Systems. Granada, Spain: Curran Associates.
Heider, Y., H. S. Suh, and W. Sun. 2021. “An offline multi-scale unsaturated poromechanics model enabled by self-designed/self-improved neural networks.” Int. J. Numer. Anal. Methods Geomech. 45 (9): 1212–1237. https://doi.org/10.1002/nag.3196.
Heider, Y., K. Wang, and W. Sun. 2020. “SO(3)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials.” Comput. Methods Appl. Mech. Eng. 363: 112875. https://doi.org/10.1016/j.cma.2020.112875.
Hochreiter, S., and J. Schmidhuber. 1997. “Long short-term memory.” Neural Comput. 9 (8): 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735.
Hu, M., C. Schrank, and K. Regenauer-Lieb. 2020. “Cross-diffusion waves in hydro-poro-mechanics.” J. Mech. Phys. Solids 135 (Feb): 103632. https://doi.org/10.1016/j.jmps.2019.05.015.
Jin, Y.-F., Z.-Y. Yin, S.-L. Shen, and P.-Y. Hicher. 2016a. “Investigation into MOGA for identifying parameters of a critical-state-based sand model and parameters correlation by factor analysis.” Acta Geotech. 11 (5): 1131–1145. https://doi.org/10.1007/s11440-015-0425-5.
Jin, Y.-F., Z.-Y. Yin, S.-L. Shen, and P.-Y. Hicher. 2016b. “Selection of sand models and identification of parameters using an enhanced genetic algorithm.” Int. J. Numer. Anal. Methods Geomech. 40 (8): 1219–1240. https://doi.org/10.1002/nag.2487.
Karniadakis, G. E., I. G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang, and L. Yang. 2021. “Physics-informed machine learning.” Nat. Rev. Phys. 3 (6): 422–440. https://doi.org/10.1038/s42254-021-00314-5.
Karpatne, A., G. Atluri, J. H. Faghmous, M. Steinbach, A. Banerjee, A. Ganguly, S. Shekhar, N. Samatova, and V. Kumar. 2017. “Theory-guided data science: A new paradigm for scientific discovery from data.” IEEE Trans. Knowl. Data Eng. 29 (10): 2318–2331.
Kneifl, J., D. Grunert, and J. Fehr. 2021. “A nonintrusive nonlinear model reduction method for structural dynamical problems based on machine learning.” Int. J. Numer. Methods Eng. 122 (17): 4774–4786.
Lagaris, I. E., A. Likas, and D. I. Fotiadis. 1998. “Artificial neural networks for solving ordinary and partial differential equations.” IEEE Trans. Neural Networks 9 (5): 987–1000.
Lefik, M., and B. A. Schrefler. 2002. “Artificial neural network for parameter identificationsfor an elasto-plastic model of superconductingcable under cyclic loading.” Comput. Struct. 80: 1699–1713. https://doi.org/10.1016/S0045-7949(02)00162-1.
Lin, S.-Q., D.-Y. Tan, J.-H. Yin, and H. Li. 2021. “A Novel approach to surface strain measurement for cylindrical rock specimens under uniaxial compression using distributed fibre optic sensor technology.” Rock Mech. Rock Eng. 54 (12): 6605–6619.
Liu, D., and Y. Wang. 2019a. “Multi-fidelity physics-constrained neural network and its application in materials modeling.” J. Mech. Des. 141 (12): 121403. https://doi.org/10.1115/1.4044400.
Liu, Z., and C. T. Wu. 2019b. “Exploring the 3D architectures of deep material network in data-driven multiscale mechanics.” J. Mech. Phys. Solids 127 (Jun): 20–46. https://doi.org/10.1016/j.jmps.2019.03.004.
Lominé, F., L. Scholtès, L. Sibille, and P. Poullain. 2013. “Modeling of fluid-solid interaction in granular media with coupled lattice Boltzmann/discrete element methods: Application to piping erosion.” Int. J. Numer. Anal. Methods Geomech. 37 (6): 577–596. https://doi.org/10.1002/nag.1109.
Lu, L., M. Dao, P. Kumar, U. Ramamurty, G. E. Karniadakis, and S. Suresh. 2020. “Extraction of mechanical properties of materials through deep learning from instrumented indentation.” Proc. Natl. Acad. Sci. U.S.A. 117 (13): 7052–7062. https://doi.org/10.1073/pnas.1922210117.
Meng, X., H. Babaee, and G. E. Karniadakis. 2021. “Multi-fidelity Bayesian neural networks: Algorithms and applications.” J. Comput. Phys. 438 (Aug): 110361. https://doi.org/10.1016/j.jcp.2021.110361.
Meng, X., and G. E. Karniadakis. 2020. “A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems.” J. Comput. Phys. 401 (Jan): 109020. https://doi.org/10.1016/j.jcp.2019.109020.
Minisci, E., and M. Vasile. 2013. “Robust design of a reentry unmanned space vehicle by multifidelity evolution control.” AIAA J. 51 (6): 1284–1295. https://doi.org/10.2514/1.J051573.
Moffat, R., R. J. Fannin, and S. J. Garner. 2011. “Spatial and temporal progression of internal erosion in cohesionless soil.” Can. Geotech. J. 48 (3): 399–412. https://doi.org/10.1139/T10-071.
Nagarajan, H. P. N., H. Mokhtarian, H. Jafarian, S. Dimassi, S. Bakrani-Balani, A. Hamedi, E. Coatanéa, G. Gary Wang, and K. R. Haapala. 2019. “Knowledge-based design of artificial neural network topology for additive manufacturing process modeling: A new approach and case study for fused deposition modeling.” J. Mech. Des. 141 (2): 021705. https://doi.org/10.1115/1.4042084.
Peigney, M. 2020. “A Fourier-based machine learning technique with application in engineering.” Int. J. Numer. Methods Eng. 122 (3): 866–897. https://doi.org/10.1002/nme.6565.
Raissi, M., P. Perdikaris, and G. E. Karniadakis. 2019. “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations.” J. Comput. Phys. 378: 686–707. https://doi.org/10.1016/j.jcp.2018.10.045.
Ramuhalli, P., L. Udpa, and S. S. Udpa. 2005. “Finite-element neural networks for solving differential equations.” IEEE Trans. Neural Networks 16 (6): 1381–1392.
Wang, K., and W. Sun. 2018. “A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning.” Comput. Methods Appl. Mech. Eng. 334 (Jun): 337–380. https://doi.org/10.1016/j.cma.2018.01.036.
Wu, Z.-X., Z.-Y. Yin, Y.-F. Jin, and X.-Y. Geng. 2017. “A straightforward procedure of parameters determination for sand: A bridge from critical state based constitutive modelling to finite element analysis.” Eur. J. Environ. Civ. Eng. 23 (12): 1444–1466. https://doi.org/10.1080/19648189.2017.1353442.
Xu, C., C. Wang, F. Ji, and X. Yuan. 2012. “Finite-element neural network-based solving 3-D differential equations in MFL.” IEEE Trans. Magn. 48 (12): 4747–4756.
Yang, J., Z. Y. Yin, F. Laouafa, and P. Y. Hicher. 2019b. “Internal erosion in dike-on-foundation modeled by a coupled hydro-mechanical approach.” Int. J. Numer. Anal. Methods Geomech. 43 (3): 663–683. https://doi.org/10.1002/nag.2877.
Yang, J., Z. Y. Yin, F. Laouafa, and P.-Y. Hicher. 2019a. “Analysis of suffusion in cohesionless soils with randomly distributed porosity and fines content.” Comput. Geotech. 111 (Jul): 157–171. https://doi.org/10.1016/j.compgeo.2019.03.011.
Yang, J., Z.-Y. Yin, F. Laouafa, and P.-Y. Hicher. 2019c. “Modeling coupled erosion and filtration of fine particles in granular media.” Acta Geotech. 14 (6): 1615–1627. https://doi.org/10.1007/s11440-019-00808-8.
Yin, M., X. Zheng, J. D. Humphrey, and G. Em Karniadakis. 2021. “Non-invasive inference of thrombus material properties with physics-informed neural networks.” Comput. Methods Appl. Mech. Eng. 375 (Mar): 113603. https://doi.org/10.1016/j.cma.2020.113603.
Yin, Z.-Y., P.-Y. Hicher, C. Dano, and Y.-F. Jin. 2017. “Modeling mechanical behavior of very coarse granular materials.” J. Eng. Mech. 143 (1): C4016006. https://doi.org/10.1061/(ASCE)EM.1943-7889.0001059.
Zhang, P., Y. F. Jin, and Z. Y. Yin. 2021a. “Machine learning–based uncertainty modelling of mechanical properties of soft clays relating to time-dependent behavior and its application.” Int. J. Numer. Anal. Methods Geomech. 45 (11): 1588–1602. https://doi.org/10.1002/nag.3215.
Zhang, P., Y. Yang, and Z.-Y. Yin. 2021b. “BiLSTM-based soil–structure interface modeling.” Int. J. Geomech. 21 (7): 04021096. https://doi.org/10.1061/(ASCE)GM.1943-5622.0002058.
Zhang, P., Z. Y. Yin, and Y. F. Jin. 2021c. “Bayesian neural network-based uncertainty modelling: Application to soil compressibility and undrained shear strength prediction.” Can. Geotech. J. https://doi.org/10.1139/cgj-2020-0751.
Zhang, P., Z. Y. Yin, and Y. F. Jin. 2021d. “State-of-the-art review of machine learning applications in constitutive modeling of soils.” Arch. Comput. Method Eng. 28 (5): 3661–3686. https://doi.org/10.1007/s11831-020-09524-z.
Zhang, P., Z. Y. Yin, Y. F. Jin, and G. L. Ye. 2020a. “An AI-based model for describing cyclic characteristics of granular materials.” Int. J. Numer. Anal. Methods Geomech. 44 (9): 1315–1335. https://doi.org/10.1002/nag.3063.
Zhang, P., Z. Y. Yin, Y. Zheng, and F. P. Gao. 2020b. “A LSTM surrogate modelling approach for caisson foundations.” Ocean Eng. 204 (May): 107263. https://doi.org/10.1016/j.oceaneng.2020.107263.
Zhang, P., and Z.-Y. Yin. 2021. “A novel deep learning-based modelling strategy from image of particles to mechanical properties for granular materials with CNN and BiLSTM.” Comput. Methods Appl. Mech. Eng. 382 (Aug): 113858. https://doi.org/10.1016/j.cma.2021.113858.
Zhang, Q., Y. Chen, Z. Yang, and E. Darve. 2021e. “Multi-constitutive neural network for large deformation poromechanics problem.” Preprint, submitted October 11, 2020. http://arxiv.org/abs/2010.15549.
Zhang, Q., X. Yan, and J. Shao. 2021f. “Fluid flow through anisotropic and deformable double porosity media with ultra-low matrix permeability: A continuum framework.” J. Pet. Sci. Eng. 200: 108349. https://doi.org/10.1016/j.petrol.2021.108349.
Information & Authors
Information
Published In
Journal of Engineering Mechanics
Volume 148 • Issue 4 • April 2022
Copyright
© 2022 American Society of Civil Engineers.
History
Received: Sep 1, 2021
Accepted: Dec 21, 2021
Published online: Feb 10, 2022
Published in print: Apr 1, 2022
Discussion open until: Jul 10, 2022
ASCE Technical Topics:
- Artificial intelligence and machine learning
- Clays
- Computer programming
- Computing in civil engineering
- Education
- Engineering fundamentals
- Erosion
- Finite element method
- Geology
- Geomechanics
- Geotechnical engineering
- Granular soils
- Hydraulic engineering
- Hydrologic models
- Hydromechanics
- Methodology (by type)
- Models (by type)
- Neural networks
- Numerical methods
- Physical models
- Piping erosion
- Practice and Profession
- Soil mechanics
- Soils (by type)
- Training
- Water and water resources
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
- Stephen K. Suryasentana, Brian B. Sheil, Bruno Stuyts, Multifidelity Data Fusion for the Estimation of Static Stiffness of Suction Caisson Foundations in Layered Soil, Journal of Geotechnical and Geoenvironmental Engineering, 10.1061/JGGEFK.GTENG-11819, 150, 8, (2024).
- Pin Zhang, Zhen-Yu Yin, Brian Sheil, Multifidelity Constitutive Modeling of Stress-Induced Anisotropic Behavior of Clay, Journal of Geotechnical and Geoenvironmental Engineering, 10.1061/JGGEFK.GTENG-11222, 150, 3, (2024).
- Ning Zhang, Annan Zhou, Yin-Fu Jin, Zhen-Yu Yin, Shui-Long Shen, An enhanced deep learning method for accurate and robust modelling of soil stress–strain response, Acta Geotechnica, 10.1007/s11440-023-01813-8, (2023).
- Sheng Zhang, Peng Lan, Hai-Chao Li, Chen-Xi Tong, Daichao Sheng, Physics-informed neural networks for consolidation of soils, Engineering Computations, 10.1108/EC-08-2021-0492, 39, 7, (2845-2865), (2022).
- Geng-Fu He, Pin Zhang, Zhen-Yu Yin, Yin-Fu Jin, Yi Yang, Multi-fidelity data-driven modelling of rate-dependent behaviour of soft clays, Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 10.1080/17499518.2022.2149815, (1-13), (2022).
- Pin Zhang, Zhen‐Yu Yin, Yin‐Fu Jin, Brian Sheil, Physics‐constrained hierarchical data‐driven modelling framework for complex path‐dependent behaviour of soils, International Journal for Numerical and Analytical Methods in Geomechanics, 10.1002/nag.3370, 46, 10, (1831-1850), (2022).