Multi-layer perceptron-based data-driven multiscale modelling of granular materials with a novel Frobenius norm-based internal variable

作     者：Mengqi Wang Y.T.Feng Shaoheng Guan Tongming Qu Mengqi Wang;Y.T.Feng;Shaoheng Guan;Tongming Qu

作者机构：Zienkiewicz Centre for Computational EngineeringFaculty of Science and EngineeringSwansea UniversitySwanseaWalesSA18EPUK Institute of Theoretical and Computational PhysicsGraz University of TechnologyGraz8010Austria Department of Civil and Environmental EngineeringHong Kong University of Science and TechnologyClearwater BayKowloonHong KongChina 

出 版 物：《Journal of Rock Mechanics and Geotechnical Engineering》 (岩石力学与岩土工程学报（英文版）)

年 卷 期：2024年第16卷第6期

页      面：2198-2218页

核心收录：

学科分类：081401[工学-岩土工程] 08[工学] 0818[工学-地质资源与地质工程] 0814[工学-土木工程] 

基　　金：supported by the National Natural Science Foundation of China(NSFC)(Grant No.12072217) 

主　　题：Granular materials History-dependence Multi-layer perceptron(MLP) Discrete element method FEM-DEM Machine learning 

摘      要：One objective of developing machine learning(ML)-based material models is to integrate them with well-established numerical methods to solve boundary value problems(BVPs).In the family of ML models,recurrent neural networks(RNNs)have been extensively applied to capture history-dependent constitutive responses of granular materials,but these multiple-step-based neural networks are neither sufficiently efficient nor aligned with the standard finite element method(FEM).Single-step-based neural networks like the multi-layer perceptron(MLP)are an alternative to bypass the above issues but have to introduce some internal variables to encode complex loading *** this work,one novel Frobenius norm-based internal variable,together with the Fourier layer and residual architectureenhanced MLP model,is crafted to replicate the history-dependent constitutive features of representative volume element(RVE)for granular *** obtained ML models are then seamlessly embedded into the FEM to solve the BVP of a biaxial compression case and a rigid strip footing *** obtained solutions are comparable to results from the FEM-DEM multiscale modelling but achieve significantly improved *** results demonstrate the applicability of the proposed internal variable in enabling MLP to capture highly nonlinear constitutive responses of granular materials.