Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows

Abstract

With this study we investigate the accuracy of deep learning models for the inference of Reynolds-Averaged Navier-Stokes solutions. We focus on a modernized U-net architecture, and evaluate a large number of trained neural networks with respect to their accuracy for the calculation of pressure and velocity distributions. In particular, we illustrate how training data size and the number of weights influence the accuracy of the solutions. With our best models we arrive at a mean relative pressure and velocity error of less than 3% across a range of previously unseen airfoil shapes. In addition all source code is publicly available in order to ensure reproducibility and to provide a starting point for researchers interested in deep learning methods for physics problems. While this work focuses on RANS solutions, the neural network architecture and learning setup are very generic, and applicable to a wide range of PDE boundary value problems on Cartesian grids.

Code and data available at: https://github.com/thunil/Deep-Flow-Prediction

References in corpus (18)

• Adam: A Method for Stochastic Optimization
• Conditional Generative Adversarial Nets
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Reconstruction of three-dimensional porous media using generative adversarial neural networks
• Interaction Networks for Learning about Objects, Relations and Physics
• Deep Neural Networks for Data-Driven Turbulence Models
• Accelerating Eulerian Fluid Simulation With Convolutional Networks
• Deep Learning the Physics of Transport Phenomena
• Hidden Fluid Mechanics: A Navier-Stokes Informed Deep Learning Framework for Assimilating Flow Visualization Data
• Long-term Forecasting using Higher Order Tensor RNNs
• Visual Interaction Networks
• Prediction of Reynolds Stresses in High-Mach-Number Turbulent Boundary Layers using Physics-Informed Machine Learning
• Learning A Physical Long-term Predictor
• Geodesic Convolutional Shape Optimization
• Application of Convolutional Neural Network to Predict Airfoil Lift Coefficient
• Generating Liquid Simulations with Deformation-aware Neural Networks
• Liquid Splash Modeling with Neural Networks

Cited by in corpus (74)

• Machine Learning in Aerodynamic Shape Optimization
• A Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometries
• Robust active flow control over a range of Reynolds numbers using an artificial neural network trained through deep reinforcement learning
• A Physics-informed Diffusion Model for High-fidelity Flow Field Reconstruction
• Deep Learning in Deterministic Computational Mechanics
• Physics-informed PointNet: A deep learning solver for steady-state incompressible flows and thermal fields on multiple sets of irregular geometries
• Wavefield reconstruction inversion via physics-informed neural networks
• Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low-dimensionalization
• Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solvers
• Graph neural networks for laminar flow prediction around random 2D shapes
• Assessment of unsteady flow predictions using hybrid deep learning based reduced order models
• Solving inverse problems using conditional invertible neural networks
• End-to-end Wind Turbine Wake Modelling with Deep Graph Representation Learning
• Learned Turbulence Modelling with Differentiable Fluid Solvers: Physics-based Loss-functions and Optimisation Horizons
• Multi-fidelity prediction of fluid flow and temperature field based on transfer learning using Fourier Neural Operator
• Teaching the Incompressible Navier-Stokes Equations to Fast Neural Surrogate Models in 3D
• Learning Mesh-Based Simulation with Graph Networks
• Uncertainty-aware Surrogate Models for Airfoil Flow Simulations with Denoising Diffusion Probabilistic Models
• Latent-space time evolution of non-intrusive reduced-order models using Gaussian process emulation
• Numerical investigation of minimum drag profiles in laminar flow using deep learning surrogates
• Manifold learning-based polynomial chaos expansions for high-dimensional surrogate models
• Aero-Nef: Neural Fields for Rapid Aircraft Aerodynamics Simulations
• Finite Volume Graph Network(FVGN): Predicting unsteady incompressible fluid dynamics with finite volume informed neural network
• On the Use of Neural Networks for Full Waveform Inversion
• Physics-driven Learning of the Steady Navier-Stokes Equations using Deep Convolutional Neural Networks
• 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
• A Finite Element-Inspired Hypergraph Neural Network: Application to Fluid Dynamics Simulations
• Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
• Deep Learning for Real-Time Aerodynamic Evaluations of Arbitrary Vehicle Shapes
• Physically Interpretable Feature Learning and Inverse Design of Supercritical Airfoils
• Predicting fluid-structure interaction with graph neural networks
• Solving the Discretised Neutron Diffusion Equations using Neural Networks
• Flowfield prediction of airfoil off-design conditions based on a modified variational autoencoder
• Machine Learning-Based Optimal Mesh Generation in Computational Fluid Dynamics
• An accuracy-enhanced transonic flow prediction method fusing deep learning and reduced-order model
• Robust training approach of neural networks for fluid flow state estimations
• A Combined Data-driven and Physics-driven Method for Steady Heat Conduction Prediction using Deep Convolutional Neural Networks
• Neural Networks for Generating Better Local Optima in Topology Optimization
• DrivAerNet: A Parametric Car Dataset for Data-Driven Aerodynamic Design and Prediction
• Solving Partial Differential Equations with Point Source Based on Physics-Informed Neural Networks
• C(NN)FD -- a deep learning framework for turbomachinery CFD analysis
• Graph Neural Network Based Surrogate Model of Physics Simulations for Geometry Design
• Benchmarking Convolutional Neural Network and Graph Neural Network based Surrogate Models on a Real-World Car External Aerodynamics Dataset
• Generative Adversarial Networks with Physical Evaluators for Spray Simulation of Pintle Injector
• Validation and parameterization of a novel physics-constrained neural dynamics model applied to turbulent fluid flow
• Multi-Fidelity Machine Learning Applied to Steady Fluid Flows
• Predicting Change, Not States: An Alternate Framework for Neural PDE Surrogates
• DrivAer Transformer: A high-precision and fast prediction method for vehicle aerodynamic drag coefficient based on the DrivAerNet++ dataset
• Learning Incompressible Fluid Dynamics from Scratch -- Towards Fast, Differentiable Fluid Models that Generalize
• Using neural networks to solve the 2D Poisson equation for electric field computation in plasma fluid simulations
• Multi-GPU Approach for Training of Graph ML Models on large CFD Meshes
• Polynomial Chaos Expansions on Principal Geodesic Grassmannian Submanifolds for Surrogate Modeling and Uncertainty Quantification
• Polynomial Ridge Flowfield Estimation
• Data-Driven Modeling of Geometry-Adaptive Steady Heat Transfer based on Convolutional Neural Networks: Heat Conduction
• Deep learning modelling of manufacturing and build variations on multi-stage axial compressors aerodynamics
• Fast buffet onset prediction and optimization method based on a pre-trained flowfield prediction model
• Parametric 3D Convolutional Autoencoder for the Prediction of Flow Fields in a Bed Configuration of Hot Particles
• Towards high-accuracy deep learning inference of compressible turbulent flows over aerofoils
• An Extensible Benchmark Suite for Learning to Simulate Physical Systems
• Simulating Continuum Mechanics with Multi-Scale Graph Neural Networks
• TripOptimizer: Generative 3D Shape Optimization and Drag Prediction using Triplane VAE Networks
• A study on using image based machine learning methods to develop the surrogate models of stamp forming simulations
• Geometry encoding for numerical simulations
• Real-time simulation of parameter-dependent fluid flows through deep learning-based reduced order models
• PICT -- A Differentiable, GPU-Accelerated Multi-Block PISO Solver for Simulation-Coupled Learning Tasks in Fluid Dynamics
• Adaptation of AI-accelerated CFD Simulations to the IPU platform
• Data-driven nonlinear aerodynamics models with certifiably optimal boundedness properties
• Learning invariance preserving moment closure model for Boltzmann-BGK equation
• Rapid feasibility assessment of components formed through hot stamping: A deep learning approach
• Subspace Graph Physics: Real-Time Rigid Body-Driven Granular Flow Simulation
• Gaussian Field Representations for Turbulent Flow: Compression, Scale Separation, and Physical Fidelity
• U-Net-Based Surrogate Model For Evaluation of Microfluidic Channels
• Fine Tuning Swimming Locomotion Learned from Mosquito Larvae