English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Stress prediction in porous materials and structures is challenging due tothe high computational cost associated with direct numerical simulations.Convolutional Neural Network (CNN) based architectures have recently beenproposed as surrogates to approximate and extrapolate the solution of suchmultiscale simulations. These methodologies are usually limited to 2D problemsdue to the high computational cost of 3D voxel based CNNs. We propose a novelgeometric learning approach based on a Graph Neural Network (GNN) thatefficiently deals with three-dimensional problems by performing convolutionsover 2D surfaces only. Following our previous developments using pixel-basedCNN, we train the GNN to automatically add local fine-scale stress correctionsto an inexpensively computed coarse stress prediction in the porous structureof interest. Our method is Bayesian and generates densities of stress fields,from which credible intervals may be extracted. As a second scientificcontribution, we propose to improve the extrapolation ability of our network bydeploying a strategy of online physics-based corrections. Specifically, wecondition the posterior predictions of our probabilistic predictions to satisfypartial equilibrium at the microscale, at the inference stage. This is doneusing an Ensemble Kalman algorithm, to ensure tractability of the Bayesianconditioning operation. We show that this innovative methodology allows us toalleviate the effect of undesirable biases observed in the outputs of theuncorrected GNN, and improves the accuracy of the predictions in general.

international journal of solids and structures

A graph-based probabilistic geometric deep learning framework with  online enforcement of physical constraints to predict the criticality of  defects in porous materials