Physics-informed graph neural Galerkin networks: a unified framework for solving PDE-governed forward and inverse problems

DOI10.1016/J.CMA.2021.114502OpenAlexW3183994952WikidataQ114196856 ScholiaQ114196856MaRDI QIDQ2072742

Han Gao, Jian-Xun Wang, Matthew J. Zahr

Publication date: 26 January 2022

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2107.12146

zbMATH Keywords

inverse problempartial differential equationsmechanicsphysics-informed machine learninggraph convolutional neural networks

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Numerical methods for inverse problems for boundary value problems involving PDEs (65N21) Numerical methods for partial differential equations, boundary value problems (65N99)

Related Items (40)

CAN-PINN: a fast physics-informed neural network based on coupled-automatic-numerical differentiation method ⋮ Physics-informed PointNet: a deep learning solver for steady-state incompressible flows and thermal fields on multiple sets of irregular geometries ⋮ PhySR: physics-informed deep super-resolution for spatiotemporal data ⋮ Enforcing continuous symmetries in physics-informed neural network for solving forward and inverse problems of partial differential equations ⋮ Isogeometric analysis-based physics-informed graph neural network for studying traffic jam in neurons ⋮ Model discovery of compartmental models with graph-supported neural networks ⋮ Frame Invariance and Scalability of Neural Operators for Partial Differential Equations ⋮ On the use of graph neural networks and shape‐function‐based gradient computation in the deep energy method ⋮ Fully probabilistic deep models for forward and inverse problems in parametric PDEs ⋮ A framework based on symbolic regression coupled with eXtended physics-informed neural networks for gray-box learning of equations of motion from data ⋮ Novel DeepONet architecture to predict stresses in elastoplastic structures with variable complex geometries and loads ⋮ A metalearning approach for physics-informed neural networks (PINNs): application to parameterized PDEs ⋮ A decoupled physics-informed neural network for recovering a space-dependent force function in the wave equation from integral overdetermination data ⋮ Adversarial deep energy method for solving saddle point problems involving dielectric elastomers ⋮ Transfer learning based physics-informed neural networks for solving inverse problems in engineering structures under different loading scenarios ⋮ A graph convolutional autoencoder approach to model order reduction for parametrized PDEs ⋮ Random vibration of hysteretic systems under Poisson white noise excitations ⋮ Physics-integrated neural differentiable (PiNDiff) model for composites manufacturing ⋮ A symmetry group based supervised learning method for solving partial differential equations ⋮ Differentiable hybrid neural modeling for fluid-structure interaction ⋮ Learning the flux and diffusion function for degenerate convection-diffusion equations using different types of observations ⋮ Robust variational physics-informed neural networks ⋮ A finite element-inspired hypergraph neural network: application to fluid dynamics simulations ⋮ A microstructure-based graph neural network for accelerating multiscale simulations ⋮ Non-diffusive neural network method for hyperbolic conservation laws ⋮ An alternating flux learning method for multidimensional nonlinear conservation laws ⋮ Physics-informed \textsc{MeshGraphNets} (PI-MGNs): neural finite element solvers for non-stationary and nonlinear simulations on arbitrary meshes ⋮ Phase field smoothing-PINN: a neural network solver for partial differential equations with discontinuous coefficients ⋮ Energy-informed graph transformer model for solid mechanical analyses ⋮ Deep learning in computational mechanics: a review ⋮ Interpretable physics-encoded finite element network to handle concentration features and multi-material heterogeneity in hyperelasticity ⋮ f-PICNN: a physics-informed convolutional neural network for partial differential equations with space-time domain ⋮ Physics-informed machine learning for the inverse design of wave scattering clusters ⋮ The novel graph transformer-based surrogate model for learning physical systems ⋮ MHDnet: physics-preserving learning for solving magnetohydrodynamics problems ⋮ Neural differentiable modeling with diffusion-based super-resolution for two-dimensional spatiotemporal turbulence ⋮ Interpretable a-posteriori error indication for graph neural network surrogate models ⋮ Taylor series error correction network for super-resolution of discretized partial differential equation solutions ⋮ IGA-Graph-Net: isogeometric analysis-reuse method based on graph neural networks for topology-consistent models ⋮ Emulation of cardiac mechanics using graph neural networks

Uses Software

• DiffSharp
• Adam
• DeepXDE
• PhyCNN
• PhyGeoNet
• VarNet
• NSFnets
• hp-VPINNs
• DiscretizationNet

Cites Work

• Unnamed Item
• Unnamed Item
• Weak adversarial networks for high-dimensional partial differential equations
• A mixed finite element approximation of the Navier-Stokes equations
• The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
• Machine learning in cardiovascular flows modeling: predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks
• Multi-level convolutional autoencoder networks for parametric prediction of spatio-temporal dynamics
• \textit{hp}-VPINNs: variational physics-informed neural networks with domain decomposition
• Discretizationnet: a machine-learning based solver for Navier-Stokes equations using finite volume discretization
• NSFnets (Navier-Stokes flow nets): physics-informed neural networks for the incompressible Navier-Stokes equations
• Physics-informed neural networks for high-speed flows
• Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
• Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems
• Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
• Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks
• Physics-informed multi-LSTM networks for metamodeling of nonlinear structures
• FEA-Net: a physics-guided data-driven model for efficient mechanical response prediction
• An energy approach to the solution of partial differential equations in computational mechanics via machine learning: concepts, implementation and applications
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Prediction of aerodynamic flow fields using convolutional neural networks
• Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
• DeepXDE: A Deep Learning Library for Solving Differential Equations
• Discrete nodal domain theorems

This page was built for publication: Physics-informed graph neural Galerkin networks: a unified framework for solving PDE-governed forward and inverse problems