Machine-Learning-Based Multiscale Methods for 3D Modelling of Granular Materials by Incorporating History-Dependent State Variables
Open Access
- 10 January 2026
- journal article
- Published by Hosokawa Powder Technology Foundation in KONA Powder and Particle Journal
- Vol. 43, 281-294
Abstract
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.Keywords
This publication has 31 references indexed in Scilit:
- A predictive deep learning framework for path-dependent mechanical behavior of granular materialsActa Geotechnica, 2022
- Deformation accommodating periodic computational domain for a uniform velocity gradientComputer Methods in Applied Mechanics and Engineering, 2021
- Smart constitutive laws: Inelastic homogenization through machine learningComputer Methods in Applied Mechanics and Engineering, 2021
- A machine learning based plasticity model using proper orthogonal decompositionComputer Methods in Applied Mechanics and Engineering, 2020
- Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural networkComputer Methods in Applied Mechanics and Engineering, 2019
- A coupled FEM/DEM approach for hierarchical multiscale modelling of granular mediaInternational Journal for Numerical Methods in Engineering, 2014
- Homogenization and two‐scale simulations of granular materials for different microstructural constraintsInternational Journal for Numerical Methods in Engineering, 2010
- Multi-scale computational homogenization: Trends and challengesJournal of Computational and Applied Mathematics, 2010
- The integration of numerical modeling and physical measurements through inverse analysis in geotechnical engineeringKSCE Journal of Civil Engineering, 2008
- Numerical implementation of a neural network based material model in finite element analysisInternational Journal for Numerical Methods in Engineering, 2004