Embedding Consistent Recurrent Neural Networks in Finite Element Simulations for Path-Dependent Damage Prediction | Zenodo
Skip to main
You are using an outdated browser. Please upgrade your browser to improve your experience.
Published July 5, 2026
| Version v1
Working paper
Open
Embedding Consistent Recurrent Neural Networks in Finite Element Simulations for Path-Dependent Damage Prediction
Authors/Creators
Kõrgesaar, Mihkel<br>(Supervisor)1
KURBAN, HASAN<br>(Editor)2
Yatkın, Muhammed Adil1
Show affiliations
1.
Tallinn University of Technology
2.
Hamad bin Khalifa University
Description
Abstract :
Capturing sheet-metal localization physics with a trained surrogate inside a finite element (FEM) solver requires the<br>surrogate to deliver consistent predictions irrespective of the solver’s strain-increment count, which is difficult to<br>control in non-linear explicit codes. We address this by re-implementing a trained recurrent neural network (RNN)<br>damage criterion in Fortran and embed it as a live, increment-by-increment fracture criterion inside an Abaqus/Ex-<br>plicit user material subroutine (VUMAT), advancing the network state alongside the solver. Two architectures are<br>compared: a SimpleRNN and the proposed Consistent RNN (ConsRNN), whose transition function renders predic-<br>tions invariant to the number of strain increments along a fixed deformation path. Both are trained on bilinear strain<br>paths and evaluated under varying temporal discretizations and nonlinear histories. SimpleRNN predictions drift<br>as the increment count increases, which disqualifies it for embedding; ConsRNN maintains a stable response, at a<br>modest accuracy cost relative to SimpleRNN on multilinear paths. Deployed at structural scale, the embedded Con-<br>sRNN surrogate reproduces the global force–displacement response of a clamped steel plate to within 0.82% of peak<br>force relative to an established two-parameter fracture criterion. The results establish increment-count consistency<br>as the decisive property for embedding recurrent surrogates in explicit FEM solvers and demonstrate the first such<br>deployment at structural scale.
Files
ConsRNN_Abaqus_Deployment_Codes.zip
Files<br>(891.0 MB)
Name<br>Size
ConsRNN_Abaqus_Deployment_Codes.zip
md5:a9a907df898e5c77107367789ff856c0
891.0 MB
Preview
Download
Additional details
Software
Programming language
Python
Fortran
Views
Downloads
Show more details
All versions<br>This version
Views
Total views
Downloads
Total downloads
Data volume
Total data volume
0 Bytes<br>0 Bytes
More info on how stats are collected....
Versions
External resources
Indexed in
OpenAIRE
Communities
Details
DOI
DOI Badge
DOI
10.5281/zenodo.21207967
Markdown
[](https://doi.org/10.5281/zenodo.21207967)
reStructuredText
.. image:: https://zenodo.org/badge/DOI/10.5281/zenodo.21207967.svg<br>:target: https://doi.org/10.5281/zenodo.21207967
HTML
Image URL
https://zenodo.org/badge/DOI/10.5281/zenodo.21207967.svg
Target URL
https://doi.org/10.5281/zenodo.21207967
Resource type<br>Working paper
Publisher<br>Zenodo
Rights
License
Creative Commons Attribution 4.0 International
The Creative Commons Attribution license allows re-distribution and re-use of a licensed work on the condition that the creator is appropriately credited.
Read more
Citation
Export
Technical metadata
Created
July 5, 2026
Modified
July 5, 2026
Jump up
This site uses cookies. Find out more on how we use cookies
Accept all cookies<br>Accept only essential cookies