Machine learning of evolving physics-based material models for multiscale solid mechanics

In this work we present a hybrid physics-based and data-driven learning approach to construct surrogate models for concurrent multiscale simulations of complex material behavior. We start from robust but inflexible physics-based constitutive models and increase their expressivity by allowing a subset of their material parameters to change in...

Physically recurrent neural networks for path-dependent heterogeneous materials: Embedding constitutive models in a data-driven surrogate

Driven by the need to accelerate numerical simulations, the use of machine learning techniques is rapidly growing in the field of computational solid mechanics. Their application is especially advantageous in concurrent multiscale finite element analysis (FE²) due to the exceedingly high computational costs often associated with it...

Neural networks meet physics-based material models: Accelerating concurrent multiscale simulations of path-dependent composite materials

In a concurrent (FE2) multiscale modeling is an increasingly popular approach for modeling complex materials. As such, it is especially suited for modeling composites, as their complex microstructure can be explicitly modeled and nested to each integration point of the macroscale. However, this generality is often associated with exceedingly...

Modelling of Environmental Ageing of Polymers and Polymer Composites—Modular and Multiscale Methods

Service lifetimes of polymers and polymer composites are impacted by environmental ageing. The validation of new composites and their environmental durability involves costly testing programs, thus calling for more affordable and safe alternatives, and modelling is seen as such an alternative. The state-of-the-art models are systematized in...

On-the-fly construction of surrogate constitutive models for concurrent multiscale mechanical analysis through probabilistic machine learning

Concurrent multiscale finite element analysis (FE²) is a powerful approach for high-fidelity modeling of materials for which a suitable macroscopic constitutive model is not available. However, the extreme computational effort associated with computing a nested micromodel at every macroscopic integration point makes FE²...

Numerical and Experimental Investigation of Hygrothermal Aging in Laminated Composites

Although being a crucial step in structural design of laminated composites, prediction of their long-term mechanical performance remains a challenging task for which no comprehensive and reliable solution is currently available. Nevertheless, structures such as wind turbine blades, of which laminated composites constitute the main load bearing...

A combined experimental/numerical investigation on hygrothermal aging of fiber-reinforced composites

This work investigates hygrothermal aging degradation of unidirectional glass/epoxy composite specimens through a combination of experiments and numerical modeling. Aging is performed through immersion in demineralized water. Interlaminar shear testes are performed after multiple conditioning times and after single immersion/redrying cycles....

A multiscale and multiphysics numerical framework for modelling of hygrothermal ageing in laminated composites

In this work, a numerical framework for modelling of hygrothermal ageing in laminated composites is proposed. The model consists of a macroscopic diffusion analysis based on Fick's second law coupled with a multiscale FE² stress analysis in order to take microscopic degradation mechanisms into account. Macroscopic material points...