From Deep to Physics-Informed Learning of Turbulence: Diagnostics

Abstract

We describe tests validating progress made toward acceleration and automation of hydrodynamic codes in the regime of developed turbulence by three Deep Learning (DL) Neural Network (NN) schemes trained on Direct Numerical Simulations of turbulence. Even the bare DL solutions, which do not take into account any physics of turbulence explicitly, are impressively good overall when it comes to qualitative description of important features of turbulence. However, the early tests have also uncovered some caveats of the DL approaches. We observe that the static DL scheme, implementing Convolutional GAN and trained on spatial snapshots of turbulence, fails to reproduce intermittency of turbulent fluctuations at small scales and details of the turbulence geometry at large scales. We show that the dynamic NN schemes, namely LAT-NET and Compressed Convolutional LSTM, trained on a temporal sequence of turbulence snapshots are capable to correct for the caveats of the static NN. We suggest a path forward towards improving reproducibility of the large-scale geometry of turbulence with NN.

arXiv comment8 pages, 3 figures

References in corpus (5)

• Generative Adversarial Text to Image Synthesis
• A Deep Learning based Approach to Reduced Order Modeling for Turbulent Flow Control using LSTM Neural Networks
• Deep Learning the Physics of Transport Phenomena
• An Efficient Deep Learning Technique for the Navier-Stokes Equations: Application to Unsteady Wake Flow Dynamics
• Lat-Net: Compressing Lattice Boltzmann Flow Simulations using Deep Neural Networks

Cited by in corpus (18)

• Surrogate Modeling for Fluid Flows Based on Physics-Constrained Deep Learning Without Simulation Data
• Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systems
• Zermelo's problem: Optimal point-to-point navigation in 2D turbulent flows using Reinforcement Learning
• Deep unsupervised learning of turbulence for inflow generation at various Reynolds numbers
• Sub-grid scale model classification and blending through deep learning
• Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
• Reservoir computing model of two-dimensional turbulent convection
• Compressed Convolutional LSTM: An Efficient Deep Learning framework to Model High Fidelity 3D Turbulence
• Generative Modeling of Turbulence
• Embedding Hard Physical Constraints in Neural Network Coarse-Graining of 3D Turbulence
• Solving Bayesian Inverse Problems via Variational Autoencoders
• Combining Differentiable PDE Solvers and Graph Neural Networks for Fluid Flow Prediction
• Learning the structure of wind: A data-driven nonlocal turbulence model for the atmospheric boundary layer
• Toward the Fully Physics-Informed Echo State Network -- an ODE Approximator Based on Recurrent Artificial Neurons
• A wall model based on neural networks for LES of turbulent flows over periodic hills
• Uncertainty Quantification of Locally Nonlinear Dynamical Systems using Neural Networks
• Adversarial sampling of unknown and high-dimensional conditional distributions
• Unsupervised Machine Learning to Teach Fluid Dynamicists to Think in 15 Dimensions