A data-driven approach to full-field nonlinear stress distribution and failure pattern prediction in composites using deep learning

DOI10.1016/j.cma.2022.115126OpenAlexW4281673431MaRDI QIDQ2145129

Maryam Shakiba, Anuj Karpatne, Reza Sepasdar

Publication date: 17 June 2022

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

Full work available at URL: https://arxiv.org/abs/2104.04485

zbMATH Keywords

composite materialsmachine learningmicro-mechanicsdeep learningfailure patternfull-field stress prediction

Mathematics Subject Classification ID

Composite and mixture properties (74E30) Stress (74A10)

Related Items (1)

Prediction of local elasto-plastic stress and strain fields in a two-phase composite microstructure using a deep convolutional neural network

Uses Software

• ParaView
• TensorFlow
• U-Net
• SciANN
• UNet++
• F3DAM

Cites Work

• A 3D interface-enriched generalized finite element method for weakly discontinuous problems with complex internal geometries
• Strain- and stress-based continuum damage models. I. Formulation
• On continuum damage-elastoplasticity at finite strains. A computational framework
• SciANN: a Keras/Tensorflow wrapper for scientific computations and physics-informed deep learning using artificial neural networks
• Geometric deep learning for computational mechanics. I: Anisotropic hyperelasticity
• Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow
• An intelligent nonlinear meta element for elastoplastic continua: deep learning using a new time-distributed residual U-net architecture
• \(\mathrm{SO}(3)\)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials
• A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning
• An interface-enriched generalized FEM for problems with discontinuous gradient fields
• Finite-deformation irreversible cohesive elements for three-dimensional crack-propagation analysis

This page was built for publication: A data-driven approach to full-field nonlinear stress distribution and failure pattern prediction in composites using deep learning