• Corpus ID: 245218872

Machine Learning-Accelerated Computational Solid Mechanics: Application to Linear Elasticity

  title={Machine Learning-Accelerated Computational Solid Mechanics: Application to Linear Elasticity},
  author={Rajat Arora},
  • Rajat Arora
  • Published 16 December 2021
  • Computer Science
  • ArXiv
This work presents a novel physics-informed deep learning based super-resolution framework to reconstruct high-resolution deformation fields from low-resolution counterparts, obtained from coarse mesh simulations or experiments. We leverage the governing equations and boundary conditions of the physical system to train the model without using any high-resolution labeled data. The proposed approach is applied to obtain the super-resolved deformation fields from the low-resolution stress and… 

Figures from this paper

PhySRNet: Physics informed super-resolution network for application in computational solid mechanics

The proposed framework provides possibilities for guiding future subgrid-scale models for modeling complex phenomena occurring at small spatial and temporal scales and opens the door to applying machine learning and traditional numerical approaches in tandem to reduce computational complexity accelerate scientific discovery and engineering design.

Physics-informed neural networks for modeling rate- and temperature-dependent plasticity

A physics-informed neural network based framework to model the strain-rate and temperature dependence of the deformation fields in elastic-viscoplastic solids and a fundamental challenge involving selection of appropriate model outputs so that the mechanical problem can be faithfully solved using neural networks is highlighted.

A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method

By properly designing the network architecture in PINN, the deep learning model has the potential to solve the unknowns in a heterogeneous domain without any available initial data from other sources.



MESHFREEFLOWNET: A Physics-Constrained Deep Continuous Space-Time Super-Resolution Framework

This work proposes MESHFREEFLOWNET, a novel deep learning-based super-resolution framework to generate continuous (grid-free) spatio-temporal solutions from the lowresolution inputs, and provides an opensource implementation of the method that supports arbitrary combinations of PDE constraints.

Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows

The present model reconstructs high-resolved turbulent flows from very coarse input data in space, and also reproduces the temporal evolution for appropriately chosen time interval, suggesting that the present method can perform a range of flow reconstructions in support of computational and experimental efforts.

PhyGeoNet: Physics-Informed Geometry-Adaptive Convolutional Neural Networks for Solving Parametric PDEs on Irregular Domain

A novel physics-constrained CNN learning architecture is proposed, aiming to learn solutions of parametric PDEs on irregular domains without any labeled data, and elliptic coordinate mapping is introduced to enable coordinate transforms between the irregular physical domain and regular reference domain.

Using Physics-Informed Super-Resolution Generative Adversarial Networks for Subgrid Modeling in Turbulent Reactive Flows

This work presents a novel subgrid modeling approach based on a generative adversarial network (GAN), which is trained with unsupervised deep learning (DL) using adversarial and physics-informed losses to improve the generalization capability, especially extrapolation, of the network.

Super-resolution analysis with machine learning for low-resolution flow data

The hybrid Downsampled Skip-Connection Multi-Scale (DSC/MS) model is presented, which can reconstruct the flow field accurately from coarse input flow field data and the possibility of a machine-learned model for superresolution in experimental and computational fluid dynamics is discussed.

Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels

This work presents a novel physics-informed DL-based SR solution using convolutional neural networks (CNN), which is able to produce HR flow fields from low-resolution (LR) inputs in high-dimensional parameter space by leveraging the conservation laws and boundary conditions of fluid flows.

tempoGAN: A Temporally Coherent, Volumetric GAN for Super-resolution Fluid Flow

This work represents a first approach to synthesize four-dimensional physics fields with neural networks based on a conditional generative adversarial network that is designed for the inference of three-dimensional volumetric data, and generates consistent and detailed results.

Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations

This two part treatise introduces physics informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given law of physics described by general nonlinear partial differential equations and demonstrates how these networks can be used to infer solutions topartial differential equations, and obtain physics-informed surrogate models that are fully differentiable with respect to all input coordinates and free parameters.