Machine-Learning-Based Multiscale Methods for 3D Modelling of Granular Materials by Incorporating History-Dependent State Variables

Wang, Mengqi; Wang, Min; Guan, Shaoheng; Feng, Yuntian

doi:10.14356/kona.2026009

Machine-Learning-Based Multiscale Methods for 3D Modelling of Granular Materials by Incorporating History-Dependent State Variables

Journal Article · Tue Apr 29 00:00:00 EDT 2025 · Kona (Hirakata)

DOI:https://doi.org/10.14356/kona.2026009· OSTI ID:2566453

Wang, Mengqi ^[1]; ^[2]; ^[3]; Feng, Yuntian ^[1]

Swansea University (United Kingdom)
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Swansea University (United Kingdom); Graz University of Technology (Austria)

Over the past decades, the prevalence of machine learning (ML) methods has made the development of ML-based constitutive models for granular materials undoubtedly a popular subject. Numerous studies have been made to feature the loading path or history-dependent stress-strain response of granular media using neural networks. In this work, a novel finite element method (FEM)–ML multiscale approach was developed by incorporating internal variables to improve the simulation accuracy of 3D history-dependent granular materials for the first time. To this end, a surrogate constitutive model based on the single-step-based multi-layer perceptron (MLP) neural network was used to replace representative volume element (RVE) simulations conducted by the discrete element method (DEM) in the multiscale FEM–DEM approach. Although the prediction principle of the MLP aligns with the FEM algorithm, artificially added internal variables are required to differentiate the loading history. To address this issue, history variables associated with the Frobenius norm are proposed to be fed into the MLP coupled with the strain tensor to extract the history-dependent behaviour of granular assemblies. The developed FEM–ML approach was demonstrated in 3D conventional triaxial compression (CTC) simulations. Compared to the multiscale FEM–DEM approach, the proposed FEM–ML method exhibits a significantly improved computational efficiency.

View Accepted Manuscript (DOE)

Research Organization:: Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)

Sponsoring Organization:: USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: 89233218CNA000001

OSTI ID:: 2566453

Report Number(s):: LA-UR--25-24157

Journal Information:: Kona (Hirakata), Journal Name: Kona (Hirakata); ISSN 0288-4534

Publisher:: Hosokawa Powder Technology FoundationCopyright Statement

Country of Publication:: United States

Language:: English

References (30)

Homogenization and two-scale simulations of granular materials for different microstructural constraints Miehe, C.; Dettmar, J.; Zäh, D. International Journal for Numerical Methods in Engineering, Vol. 83, Issue 8-9 https://doi.org/10.1002/nme.2875	journal	August 2010
A coupled FEM/DEM approach for hierarchical multiscale modelling of granular media: HIERARCHICAL MULTISCALE MODELLING OF GRANULAR MEDIA Guo, Ning; Zhao, Jidong International Journal for Numerical Methods in Engineering, Vol. 99, Issue 11 https://doi.org/10.1002/nme.4702	journal	June 2014
Numerical implementation of a neural network based material model in finite element analysis: NEURAL NETWORK BASED MATERIAL MODEL Hashash, Y. M. A.; Jung, S.; Ghaboussi, J. International Journal for Numerical Methods in Engineering, Vol. 59, Issue 7 https://doi.org/10.1002/nme.905	journal	January 2004
A predictive deep learning framework for path-dependent mechanical behavior of granular materials Ma, Gang; Guan, Shaoheng; Wang, Qiao Acta Geotechnica, Vol. 17, Issue 8, p. 3463-3478 https://doi.org/10.1007/s11440-021-01419-y	journal	January 2022
The integration of numerical modeling and physical measurements through inverse analysis in geotechnical engineering Hashash, Y. M. A.; Song, H. KSCE Journal of Civil Engineering, Vol. 12, Issue 3 https://doi.org/10.1007/s12205-008-0165-2	journal	May 2008
Statistical theories for the elastic moduli of two-dimensional assemblies of granular materials Kruyt, N. P.; Rothenburg, L. International Journal of Engineering Science, Vol. 36, Issue 10 https://doi.org/10.1016/S0020-7225(98)00003-2	journal	August 1998
Constitutive modeling of geomaterials from non-uniform material tests Sidarta, D. E.; Ghaboussi, J. Computers and Geotechnics, Vol. 22, Issue 1 https://doi.org/10.1016/S0266-352X(97)00035-9	journal	January 1998
Multi-scale computational homogenization: Trends and challenges Geers, M. G. D.; Kouznetsova, V. G.; Brekelmans, W. A. M. Journal of Computational and Applied Mathematics, Vol. 234, Issue 7 https://doi.org/10.1016/j.cam.2009.08.077	journal	August 2010
Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural network Ghavamian, F.; Simone, A. Computer Methods in Applied Mechanics and Engineering, Vol. 357 https://doi.org/10.1016/j.cma.2019.112594	journal	December 2019
A machine learning based plasticity model using proper orthogonal decomposition Huang, Dengpeng; Fuhg, Jan Niklas; Weißenfels, Christian Computer Methods in Applied Mechanics and Engineering, Vol. 365 https://doi.org/10.1016/j.cma.2020.113008	journal	June 2020
Smart constitutive laws: Inelastic homogenization through machine learning Logarzo, Hernan J.; Capuano, German; Rimoli, Julian J. Computer Methods in Applied Mechanics and Engineering, Vol. 373 https://doi.org/10.1016/j.cma.2020.113482	journal	January 2021
Deformation accommodating periodic computational domain for a uniform velocity gradient Wang, Min; Zhang, Duan Z. Computer Methods in Applied Mechanics and Engineering, Vol. 374 https://doi.org/10.1016/j.cma.2020.113607	journal	February 2021
3D multiscale modeling of strain localization in granular media Guo, Ning; Zhao, Jidong Computers and Geotechnics, Vol. 80 https://doi.org/10.1016/j.compgeo.2016.01.020	journal	December 2016
Modelling the non-coaxiality of soils from the view of cross-anisotropy Tian, Yu; Yao, Yang-Ping Computers and Geotechnics, Vol. 86 https://doi.org/10.1016/j.compgeo.2017.01.013	journal	June 2017
Data-driven strain–stress modelling of granular materials via temporal convolution neural network Wang, Mengqi; Qu, Tongming; Guan, Shaoheng Computers and Geotechnics, Vol. 152 https://doi.org/10.1016/j.compgeo.2022.105049	journal	December 2022
Strain localisation in granular media Desrues, Jacques; Andò, Edward Comptes Rendus Physique, Vol. 16, Issue 1 https://doi.org/10.1016/j.crhy.2015.01.001	journal	January 2015
Towards data-driven constitutive modelling for granular materials via micromechanics-informed deep learning Qu, Tongming; Di, Shaocheng; Feng, Y. T. International Journal of Plasticity, Vol. 144 https://doi.org/10.1016/j.ijplas.2021.103046	journal	September 2021
Deep active learning for constitutive modelling of granular materials: From representative volume elements to implicit finite element modelling Qu, Tongming; Guan, Shaoheng; Feng, Y. T. International Journal of Plasticity, Vol. 164 https://doi.org/10.1016/j.ijplas.2023.103576	journal	May 2023
A micro–macro homogenization approach for discrete particle assembly – Cosserat continuum modeling of granular materials Li, Xikui; Liu, Qipeng; Zhang, Junbo International Journal of Solids and Structures, Vol. 47, Issue 2 https://doi.org/10.1016/j.ijsolstr.2009.09.033	journal	January 2010
An anisotropic elastic–viscoplastic model for soft clays Yin, Zhen-Yu; Chang, Ching S.; Karstunen, Minna International Journal of Solids and Structures, Vol. 47, Issue 5 https://doi.org/10.1016/j.ijsolstr.2009.11.004	journal	March 2010
Multiscale modeling and characterization of granular matter: From grain kinematics to continuum mechanics Andrade, J. E.; Avila, C. F.; Hall, S. A. Journal of the Mechanics and Physics of Solids, Vol. 59, Issue 2 https://doi.org/10.1016/j.jmps.2010.10.009	journal	February 2011
Multi-layer perceptron-based data-driven multiscale modelling of granular materials with a novel Frobenius norm-based internal variable Wang, Mengqi; Feng, Y. T.; Guan, Shaoheng Journal of Rock Mechanics and Geotechnical Engineering, Vol. 16, Issue 6 https://doi.org/10.1016/j.jrmge.2024.02.003	journal	June 2024
Multiscale framework for behavior prediction in granular media Andrade, José E.; Tu, Xuxin Mechanics of Materials, Vol. 41, Issue 6 https://doi.org/10.1016/j.mechmat.2008.12.005	journal	June 2009
Interface modeling in incompressible media using level sets in Escript Gross, L.; Bourgouin, L.; Hale, A. J. Physics of the Earth and Planetary Interiors, Vol. 163, Issue 1-4 https://doi.org/10.1016/j.pepi.2007.04.004	journal	August 2007
A strain energy-based elastic parameter calibration method for lattice/bonded particle modelling of solid materials Qu, Tongming; Feng, Y. T.; Zhao, Jidong Powder Technology, Vol. 410 https://doi.org/10.1016/j.powtec.2022.117860	journal	September 2022
Cyclic behavior of saturated soft clay under stress path with bidirectional shear stresses Hu, Xiuqing; Zhang, Yan; Guo, Lin Soil Dynamics and Earthquake Engineering, Vol. 104 https://doi.org/10.1016/j.soildyn.2017.10.016	journal	January 2018
Stress-Strain Behavior of Geomaterials in Loading Reversal Simulated by Time-Delay Neural Networks Basheer, I. A. Journal of Materials in Civil Engineering, Vol. 14, Issue 3 https://doi.org/10.1061/(ASCE)0899-1561(2002)14:3(270)	journal	June 2002
BiLSTM-Based Soil–Structure Interface Modeling Zhang, Pin; Yang, Yi; Yin, Zhen-Yu International Journal of Geomechanics, Vol. 21, Issue 7 https://doi.org/10.1061/(ASCE)GM.1943-5622.0002058	journal	July 2021
Selection of Methodology for Neural Network Modeling of Constitutive Hystereses Behavior of Soils Basheer, Imad A. Computer-Aided Civil and Infrastructure Engineering, Vol. 15, Issue 6 https://doi.org/10.1111/0885-9507.00206	journal	November 2000
Numerical simulations of deviatoric shear deformation of granular media Thornton, C. Géotechnique, Vol. 50, Issue 1 https://doi.org/10.1680/geot.2000.50.1.43	journal	February 2000

Similar Records

An adaptive, data-driven multiscale approach for dense granular flows

Journal Article · Fri Oct 31 20:00:00 EDT 2025 · Computer Methods in Applied Mechanics and Engineering · OSTI ID:3015371

An adaptive granular representative volume element model with an evolutionary periodic boundary for hierarchical multiscale analysis

Journal Article · Mon Jan 04 19:00:00 EST 2021 · International Journal for Numerical Methods in Engineering · OSTI ID:1774465

The Combined Finite-Discrete Element Method applied to the Study of Rock Fracturing Behavior in 3D

Conference · Fri Dec 31 23:00:00 EST 2010 · OSTI ID:1011073

Related Subjects

granular materials
machine learning
multiscale modelling
muti-layer Perceptron
path-dependent constitutive response

Machine-Learning-Based Multiscale Methods for 3D Modelling of Granular Materials by Incorporating History-Dependent State Variables

Citation Formats

References (30)

Similar Records

Related Subjects