Deep material network with cohesive layers: Multi-stage training and interfacial failure analysis

A fundamental issue in multiscale materials modeling and design is the consideration of traction–separation behavior at the interface. By enriching the deep material network (DMN) with cohesive layers, the paper presents a novel data-driven material model which enables accurate and efficient predict...

Full description

Saved in:
Bibliographic Details
Published in:Computer methods in applied mechanics and engineering 2020-05, Vol.363, p.1-25, Article 112913
Main Author: Liu, Zeliang
Format: Article
Language:English
Subjects:
Cohesion
Composites
Computer simulation
Computing time
Debonding analysis
Direct numerical simulation
Failure analysis
Fiber composites
Machine learning
Mathematical models
Model reduction
Multiscale analysis
Path-dependency
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A fundamental issue in multiscale materials modeling and design is the consideration of traction–separation behavior at the interface. By enriching the deep material network (DMN) with cohesive layers, the paper presents a novel data-driven material model which enables accurate and efficient prediction of multiscale responses for heterogeneous materials with interfacial effect. In the newly invoked cohesive building block, the fitting parameters have physical meanings related to the length scale and orientation of the cohesive layer. It is shown that the enriched material network can be effectively optimized via a multi-stage training strategy, with training data generated only from linear elastic direct numerical simulation (DNS). The extrapolation capability of the method to unknown material and loading spaces is demonstrated through the debonding analysis of a unidirectional fiber-reinforced composite, where the interface behavior is governed by an irreversible softening mixed-mode cohesive law. Its predictive accuracy is validated against the nonlinear path-dependent DNS results, and the reduction in computational time is particularly significant. •New architecture of material network enriched by adaptable cohesive networks.•Physics-based building block with a length parameter to describe the size effect.•Multi-stage training strategy to reduce the number of fitting parameters at each stage.•Extrapolation of enriched DMN from linear elasticity to nonlinear failure analysis.
ISSN:0045-7825
DOI:10.1016/j.cma.2020.112913