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

@article{Arora2022PhySRNetPI,
  title={PhySRNet: Physics informed super-resolution network for application in computational solid mechanics},
  author={Rajat Arora},
  journal={ArXiv},
  year={2022},
  volume={abs/2206.15457}
}
Traditional approaches based on finite element analyses have been successfully used to predict the macro-scale behavior of heterogeneous materials (composites, multicomponent alloys, and polycrystals) widely used in industrial applications. However, this necessitates the mesh size to be smaller than the characteristic length scale of the mi-crostructural heterogeneities in the material leading to computationally expensive and time-consuming calculations. The recent advances in deep learning… 

Figures from this paper

References

SHOWING 1-10 OF 50 REFERENCES

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

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

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

TLDR
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.

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

TLDR
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.

Physics-informed deep learning for incompressible laminar flows

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

TLDR
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.

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

TLDR
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.

Prediction of the evolution of the stress field of polycrystals undergoing elastic-plastic deformation with a hybrid neural network model

TLDR
This work uses a neural network with convolutional layers encoding correlations in time and space to predict the evolution of the stress field given only the initial microstructure and external loading, and shows that the stress fields and their rates are in high fidelity with the crystal plasticity data and have no visible artifacts.

Physics-Informed Neural Networks for Nonhomogeneous Material Identification in Elasticity Imaging

TLDR
By employing two neural networks in the model, the capability of material identification ofPINNs is extended to include nonhomogeneous material parameter fields, which enables more flexibility of PINNs in representing complex material properties.