Deep Learning the Physics of Transport Phenomena

2017/09/07 by Amir Barati Farimani, Farimani, Amir Barati, Joseph Gomes +3 · 6 citations
Computer Science · Engineering · Physics and Astronomy · #Computational Physics (physics.comp-ph) #FOS: Computer and information sciences #FOS: Physical sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Model Reduction and Neural Networks #Nuclear Engineering Thermal-Hydraulics

paper · pdf · doi:10.48550/arxiv.1709.02432

openalex publication_date 2017/09/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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

Cited by

• DiffTopo: Solver in the Loop for Inverse Topography via Condition Diffusion Generation
• A Note on the Inception Score
• Deep Euler method: solving ODEs by approximating the local truncation error of the Euler method
• Surrogate modeling of Cellular-Potts Agent-Based Models as a segmentation task using the U-Net neural network architecture
• DiffusionNet: Accelerating the solution of Time-Dependent partial differential equations using deep learning
• Generative diffusion model surrogates for mechanistic agent-based biological models

Related

• Deep Learning Convective Flow Using Conditional Generative Adversarial Networks
• Physics Informed Deep Learning for Flow and Transport in Porous Media
• Fast Modeling and Understanding Fluid Dynamics Systems with\n Encoder-Decoder Networks
• Learning Transport Processes with Machine Intelligence
• Physics Informed Deep Learning for Transport in Porous Media. Buckley Leverett Problem