Available on ORBi since
17 December 2021
Scientific conference in universities or research centers (Scientific conferences in universities or research centers)
A Recurrent Neural Network-based Surrogate Model for History-Dependent Multi-scale Simulations of Composite Materials
Wu, Ling  ; Noels, Ludovic 
2021
 

Files


Full Text
2021_BRAIA_RNN.pdf
Author postprint (4.84 MB)

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Multiscale; Artificial Intelligence; Recurrent Neural Network
Abstract :
[en] In order to make Computational homogenization affordable, pre-off-line finite element simulations are conducted on the mesoscale problem in order to build a synthetic database that can, in turn, be used to train surrogate models, which can be used as a constitutive law on a classical finite element simulation, speeding up the multi-scale process by several orders. Artificial neural networks (NNWs) offer the possibility to serve as a surrogate model, but a difficulty arises for elasto-plasticity because of its history-dependency. This difficulty can be solved by considering a Recurrent Neural Network (RNN), which uses sequential information [1]. Nevertheless, in order to be accurate under multi-dimensional non-proportional loading conditions, a sufficiently wide database is required in order to perform the training. To this end, a sequential training synthetic database is obtained from finite element simulations on an elasto-plastic RVE subjected to random loading paths. The RNN predictions are thus found to be in agreement with the FE2 simulations, while reducing the computational cost by 4 orders. Nevertheless, such a paradigm is essentially used as a mapping between the macro-stress and macro-strain tensors of the micro-scale boundary value response and the micro-structure information could not be recovered in a so-called localization step. We thus also develop Recurrent Neural Networks (RNNs)-based surrogate of the local micro-structure state variables for complex loading scenarios [2]. In order to address the curse of dimensionality arising because of the large amount of internal state variables in the micro-structure, we enrich the RNN with PCA dimensionality reduction and dimensionality break down, i.e. the use of several RNNs instead of a single one. The sequential training strategy is optimized to allow for GPU usage.
Research center :
A\u0026M - Aérospatiale et Mécanique - ULiège
Disciplines :
Materials science & engineering
Mechanical engineering
Author, co-author :
Wu, Ling ;  Université de Liège - ULiège > Département d'aérospatiale et mécanique > Computational & Multiscale Mechanics of Materials (CM3)
Noels, Ludovic  ;  Université de Liège - ULiège > Département d'aérospatiale et mécanique > Computational & Multiscale Mechanics of Materials (CM3)
Language :
English
Title :
A Recurrent Neural Network-based Surrogate Model for History-Dependent Multi-scale Simulations of Composite Materials
Publication date :
08 December 2021
Event name :
BRAIA Lecture Series on Technology Frontier
Event organizer :
BRAIA
Event place :
Northwestern Polytechnical University, China
Event date :
8 December 2021
Audience :
International
European Projects :
H2020 - 862015 - MOAMMM - Multi-scale Optimisation for Additive Manufacturing of fatigue resistant shock-absorbing MetaMaterials
Funders :
This project has received funding from the European Union´s Horizon 2020 research and innovation programme under grant agreement No. 862015 for the project “Multiscale Optimisation for Additive Manufacturing of fatigue resistant shock-absorbing MetaMaterials (MOAMMM)” of the H2020-EU.1.2.1. - FET Open Programme.
CE - Commission Européenne

Statistics


Number of views
64 (4 by ULiège)
Number of downloads
25 (1 by ULiège)

Bibliography


Similar publications



Contact ORBi