Multiscale computational modelling is challenging due to the high computational cost of direct numerical simulation by finite elements. To address this issue, concurrent multiscale methods use the solution of cheaper macroscale surrogates as boundary conditions to microscale sliding windows. The microscale problems remain a numerically challenging operation both in terms of implementation and cost. In this work we propose to replace the local microscale solution by an Encoder-Decoder Convolutional Neural Network that will generate fine-scale stress corrections to coarse predictions around unresolved microscale features, without prior parametrisation of local microscale problems. We deploy a Bayesian approach providing credible intervals to evaluate the uncertainty of the predictions, which is then used to investigate the merits of a selective learning framework. We will demonstrate the capability of the approach to predict equivalent stress fields in porous structures using linearised and finite strain elasticity theories.

The growing need for creating i rfacei wiyh ipuecifc wetng puropuertiei, i ch ai i puerh rdopuhobic behavior, aiki for novel meyhodi for yheir efcieny deiign. In yhii work, a faiy compu yational meyhod for yhe eval ation of puaterned i puerh rdopuhobic i rfacei ii inyrod ced. The h dropuhobiciy of a i rface ii q antifed in energ yermi yhro gh an objective f nction. The increaied compu yational coiy led yo yhe puarallelization of yhe meyhod wiyh yhe Meiiage Paiiing Inyerface (MPI) comm nication puroyocol yhay enablei calc lationi on diiyrib yed memor i iyemi allowing for puarameyric inveitigationi ay accepuyable time framei. The meyhod ii demoniyrayed for a i rface coniiiting of an arra of puillari wiyh inveryed conical (fr iy m) geomeyr . The puarallel ipueed pu achieved allowi for low coiy puarameyric inveitigationi on yhe efecy of yhe fne feay rei (c rvay re and ilopuei) of yhe puillari on yhe i puerh dopuhobiciy of yhe i rface and conieq enyl for yhe oputimization of i puerh rdopuhobic i rfacei.

Stress prediction in porous materials and structures is challenging due to the high computational cost associated with direct numerical simulations. Convolutional Neural Network (CNN) based architectures have recently been proposed as surrogates to approximate and extrapolate the solution of such multiscale simulations. These methodologies are usually limited to 2D problems due to the high computational cost of 3D voxel based CNNs. We propose a novel geometric learning approach based on a Graph Neural Network (GNN) that efficiently deals with three-dimensional problems by performing convolutions over 2D surfaces only. Following our previous developments using pixel-based CNN, we train the GNN to automatically add local fine-scale stress corrections to an inexpensively computed coarse stress prediction in the porous structure of interest. Our method is Bayesian and generates densities of stress fields, from which credible intervals may be extracted. As a second scientific contribution, we propose to improve the extrapolation ability of our network by deploying a strategy of online physics-based corrections. Specifically, we condition the posterior predictions of our probabilistic predictions to satisfy partial equilibrium at the microscale, at the inference stage. This is done using an Ensemble Kalman algorithm, to ensure tractability of the Bayesian conditioning operation. We show that this innovative methodology allows us to alleviate the effect of undesirable biases observed in the outputs of the uncorrected GNN, and improves the accuracy of the predictions in general.

No abstract

No abstract