Accelerating Eulerian Fluid Simulation With Convolutional Networks
arXiv:1607.03597
Abstract
Efficient simulation of the Navier-Stokes equations for fluid flow is a long standing problem in applied mathematics, for which state-of-the-art methods require large compute resources. In this work, we propose a data-driven approach that leverages the approximation power of deep-learning with the precision of standard solvers to obtain fast and highly realistic simulations. Our method solves the incompressible Euler equations using the standard operator splitting method, in which a large sparse linear system with many free parameters must be solved. We use a Convolutional Network with a highly tailored architecture, trained using a novel unsupervised learning framework to solve the linear system. We present real-time 2D and 3D simulations that outperform recently proposed data-driven methods; the obtained results are realistic and show good generalization properties.
Update website url
References in corpus (1)
Cited by in corpus (72)
- DGM: A deep learning algorithm for solving partial differential equations
- Machine learning accelerated computational fluid dynamics
- Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
- Explainable Machine Learning for Scientific Insights and Discoveries
- Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows
- Learning data driven discretizations for partial differential equations
- Deep Generative Design: Integration of Topology Optimization and Generative Models
- Weak Adversarial Networks for High-dimensional Partial Differential Equations
- Hamiltonian Neural Networks
- Deep learning for determining a near-optimal topological design without any iteration
- A Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometries
- Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledge
- PINNeik: Eikonal solution using physics-informed neural networks
- Artificial Intelligence for Satellite Communication: A Review
- Deep learning observables in computational fluid dynamics
- Wavefield reconstruction inversion via physics-informed neural networks
- Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solvers
- Pricing options and computing implied volatilities using neural networks
- Deep learning algorithm for data-driven simulation of noisy dynamical system
- Classification of Machine Learning Frameworks for Data-Driven Thermal Fluid Models
- Deep learning of thermodynamics-aware reduced-order models from data
- Iterative Surrogate Model Optimization (ISMO): An active learning algorithm for PDE constrained optimization with deep neural networks
- FEA-Net: A Physics-guided Data-driven Model for Efficient Mechanical Response Prediction
- Incorporating Symmetry into Deep Dynamics Models for Improved Generalization
- Digital Twin: Values, Challenges and Enablers
- A Multi-Pass GAN for Fluid Flow Super-Resolution
- Poisson CNN: Convolutional neural networks for the solution of the Poisson equation on a Cartesian mesh
- Modeling human intuitions about liquid flow with particle-based simulation
- Transport-Based Neural Style Transfer for Smoke Simulations
- Teaching the Incompressible Navier-Stokes Equations to Fast Neural Surrogate Models in 3D
- Mosaic Flows: A Transferable Deep Learning Framework for Solving PDEs on Unseen Domains
- Symbolic Pregression: Discovering Physical Laws from Distorted Video
- Physics-Informed Supervised Residual Learning for Electromagnetic Modeling
- Using Deep Learning to Extend the Range of Air-Pollution Monitoring and Forecasting
- Learning to Control PDEs with Differentiable Physics
- Lagrangian Neural Style Transfer for Fluids
- Finite volume method network for acceleration of unsteady computational fluid dynamics: non-reacting and reacting flows
- Learning A Physical Long-term Predictor
- Fluid Simulation on Neural Flow Maps
- Comparing recurrent and convolutional neural networks for predicting wave propagation
- Physics Informed Data Driven model for Flood Prediction: Application of Deep Learning in prediction of urban flood development
- Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
- Machine Learning-Based Optimal Mesh Generation in Computational Fluid Dynamics
- Learning Robotic Manipulation of Granular Media
- 3D Fluid Flow Estimation with Integrated Particle Reconstruction
- A Combined Data-driven and Physics-driven Method for Steady Heat Conduction Prediction using Deep Convolutional Neural Networks
- Non-intrusive and semi-intrusive uncertainty quantification of a multiscale in-stent restenosis model
- Learning Free-Surface Flow with Physics-Informed Neural Networks
- MPMNet: A Data-Driven MPM Framework for Dynamic Fluid-Solid Interaction
- PDE-Driven Spatiotemporal Disentanglement
- Machine Learning Changes the Rules for Flux Limiters
- Frequency-Aware Reconstruction of Fluid Simulations with Generative Networks
- A fluid simulation system based on the MPS method
- Direct Prediction of Steady-State Flow Fields in Meshed Domain with Graph Networks
- Liquid Splash Modeling with Neural Networks
- Using neural networks to solve the 2D Poisson equation for electric field computation in plasma fluid simulations
- Learning Incompressible Fluid Dynamics from Scratch -- Towards Fast, Differentiable Fluid Models that Generalize
- Polynomial Ridge Flowfield Estimation
- Meta-Learning Dynamics Forecasting Using Task Inference
- Towards Sustainable Architecture: 3D Convolutional Neural Networks for Computational Fluid Dynamics Simulation and Reverse DesignWorkflow
- Dynamic Upsampling of Smoke through Dictionary-based Learning
- Multi-Grid Graph Neural Networks with Self-Attention for Computational Mechanics
- Simulating Continuum Mechanics with Multi-Scale Graph Neural Networks
- STENCIL-NET: Data-driven solution-adaptive discretization of partial differential equations
- Efficient Cloth Simulation using Miniature Cloth and Upscaling Deep Neural Networks
- Implicit Neural Solver for Time-dependent Linear PDEs with Convergence Guarantee
- Resistance-Time Co-Modulated PointNet for Temporal Super-Resolution Simulation of Blood Vessel Flows
- Semi-Implicit Neural Solver for Time-dependent Partial Differential Equations
- DEBOSH: Deep Bayesian Shape Optimization
- Data-Driven Continuum Dynamics via Transport-Teleport Duality
- Irregular Convolutional Auto-Encoder on Point Clouds
- Subspace Graph Physics: Real-Time Rigid Body-Driven Granular Flow Simulation