Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural network

DOI10.1016/j.cma.2019.112594zbMath1442.65142OpenAlexW2969744700WikidataQ127334722 ScholiaQ127334722MaRDI QIDQ2179209

Publication date: 12 May 2020

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL: http://resolver.tudelft.nl/uuid:3832b270-d25c-4c06-8a96-000d62298d9d

zbMATH Keywords

recurrent neural network viscoplasticity machine learning multiscale modeling deep learning strain softening

Mathematics Subject Classification ID

Small-strain, rate-dependent theories of plasticity (including theories of viscoplasticity) (74C10) Finite element, Rayleigh-Ritz, Galerkin and collocation methods for ordinary differential equations (65L60)

Related Items (41)

Learning constitutive relations using symmetric positive definite neural networks ⋮ Multiscale modeling of inelastic materials with thermodynamics-based artificial neural networks (TANN) ⋮ Microstructure-guided deep material network for rapid nonlinear material modeling and uncertainty quantification ⋮ An intelligent nonlinear meta element for elastoplastic continua: deep learning using a new time-distributed residual U-net architecture ⋮ Accelerated offline setup of homogenized microscopic model for multi‐scale analyses using neural network with knowledge transfer ⋮ Thermodynamically consistent machine-learned internal state variable approach for data-driven modeling of path-dependent materials ⋮ Micromechanics-based surrogate models for the response of composites: a critical comparison between a classical mesoscale constitutive model, hyper-reduction and neural networks ⋮ A comparative study on different neural network architectures to model inelasticity ⋮ Recurrent and convolutional neural networks in structural dynamics: a modified attention steered encoder-decoder architecture versus LSTM versus GRU versus TCN topologies to predict the response of shock wave-loaded plates ⋮ Intelligent stiffness computation for plate and beam structures by neural network enhanced finite element analysis ⋮ Surrogate modeling for the homogenization of elastoplastic composites based on RBF interpolation ⋮ A machine learning-based viscoelastic-viscoplastic model for epoxy nanocomposites with moisture content ⋮ \(\mathrm{FE^{ANN}}\): an efficient data-driven multiscale approach based on physics-constrained neural networks and automated data mining ⋮ A monolithic hyper ROM \(\mathrm{FE}^2\) method with clustered training at finite deformations ⋮ Pre-trained transformer model as a surrogate in multiscale computational homogenization framework for elastoplastic composite materials subjected to generic loading paths ⋮ Incremental neural controlled differential equations for modeling of path-dependent material behavior ⋮ Physics-based self-learning spiking neural network enhanced time-integration scheme for computing viscoplastic structural finite element response ⋮ Graph-based representation of history-dependent material response in the data-driven computational mechanics framework ⋮ Transfer learning of recurrent neural network‐based plasticity models ⋮ Cooperative data-driven modeling ⋮ Reducing internal variables and improving efficiency in data-driven modelling of anisotropic damage from RVE simulations ⋮ On-the-fly construction of surrogate constitutive models for concurrent multiscale mechanical analysis through probabilistic machine learning ⋮ A framework for neural network based constitutive modelling of inelastic materials ⋮ Learning adaptive coarse basis functions of FETI-DP ⋮ Physically recurrent neural networks for path-dependent heterogeneous materials: embedding constitutive models in a data-driven surrogate ⋮ A recurrent neural network-accelerated multi-scale model for elasto-plastic heterogeneous materials subjected to random cyclic and non-proportional loading paths ⋮ Micromechanics-based material networks revisited from the interaction viewpoint; robust and efficient implementation for multi-phase composites ⋮ Data-driven reduced homogenization for transient diffusion problems with emergent history effects ⋮ Learning constitutive models from microstructural simulations via a non-intrusive reduced basis method ⋮ An FE-DMN method for the multiscale analysis of short fiber reinforced plastic components ⋮ Surrogate modeling of elasto-plastic problems via long short-term memory neural networks and proper orthogonal decomposition ⋮ Recurrent neural networks (RNNs) learn the constitutive law of viscoelasticity ⋮ TONR: an exploration for a novel way combining neural network with topology optimization ⋮ Smart constitutive laws: inelastic homogenization through machine learning ⋮ High performance reduction technique for multiscale finite element modeling (HPR-FE\(^2\)): towards industrial multiscale FE software ⋮ Interaction-based material network: a general framework for (porous) microstructured materials ⋮ Recurrent neural networks (RNNs) with dimensionality reduction and break down in computational mechanics; application to multi-scale localization step ⋮ Physics-based self-learning recurrent neural network enhanced time integration scheme for computing viscoplastic structural finite element response ⋮ Material Modeling via Thermodynamics-Based Artificial Neural Networks ⋮ A robust unsupervised neural network framework for geometrically nonlinear analysis of inelastic truss structures ⋮ Hydro-mechanical multiscale numerical manifold model of the three-dimensional heterogeneous poro-elasticity

Cites Work

This page was built for publication: Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural network