Improved multiscale computational strategies for delamination

TL;DR

This paper introduces a three-scale computational strategy using a LaTIn-based solver and parallelization techniques to reliably simulate delamination in composite materials across multiple scales.

Contribution

It develops a novel multiscale approach with adaptive search directions and scalable parallelization for improved delamination simulation in composites.

Findings

Effective updating of search directions based on interface delamination status

Parallel macro problem implementation ensures scalability

Introduction of a super-macro scale enhances long-range phenomenon modeling

Abstract

This paper presents a three-scale computational strategy for the study of composite modeled at the mesoscale so that delamination can be reliably simulated. The solver is based on a LaTIn approach so that nonlinearities can be tackled at the local scale. We show how search directions which are parameters of the method need to be updated according to the delamination status of interfaces. We also present the parallelization of the macro problem which transmits long-range phenomena and warranty the scalabilty of the method; this parallelization is based on the balancing domain decomposition method, it leads to the introduction of a third (super-macro) scale.