Deep Learning the Physics of Transport Phenomena

Abstract

We have developed a new data-driven paradigm for the rapid inference, modeling and simulation of the physics of transport phenomena by deep learning. Using conditional generative adversarial networks (cGAN), we train models for the direct generation of solutions to steady state heat conduction and incompressible fluid flow purely on observation without knowledge of the underlying governing equations. Rather than using iterative numerical methods to approximate the solution of the constitutive equations, cGANs learn to directly generate the solutions to these phenomena, given arbitrary boundary conditions and domain, with high test accuracy (MAE$<$1\%) and state-of-the-art computational performance. The cGAN framework can be used to learn causal models directly from experimental observations where the underlying physical model is complex or unknown.

Cited by in corpus (10)

• Compressed Convolutional LSTM: An Efficient Deep Learning framework to Model High Fidelity 3D Turbulence
• Encoding Invariances in Deep Generative Models
• A Combined Data-driven and Physics-driven Method for Steady Heat Conduction Prediction using Deep Convolutional Neural Networks
• Pedestrian Wind Factor Estimation in Complex Urban Environments
• FD-Net with Auxiliary Time Steps: Fast Prediction of PDEs using Hessian-Free Trust-Region Methods
• DiffusionNet: Accelerating the solution of Time-Dependent partial differential equations using deep learning
• Deep learning of material transport in complex neurite networks
• Deep-learning PDEs with unlabeled data and hardwiring physics laws
• Finite Difference Neural Networks: Fast Prediction of Partial Differential Equations
• Application of Video-to-Video Translation Networks to Computational Fluid Dynamics