Surrogate convolutional neural network models for steady computational fluid dynamics simulations

DOI10.1553/etna_vol56s235zbMath1496.76110OpenAlexW3164867305MaRDI QIDQ2672202

Matthias Eichinger, Axel Klawonn, Alexander Heinlein

Publication date: 8 June 2022

Published in: ETNA. Electronic Transactions on Numerical Analysis (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1553/etna_vol56s235

zbMATH Keywords

steady Navier-Stokes equationstransfer learningU-netiterative sequential learningreduced order surrogate model

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Navier-Stokes equations for incompressible viscous fluids (76D05) Basic methods in fluid mechanics (76M99)

Related Items

\(\mathrm{U}^p\)-net: a generic deep learning-based time stepper for parameterized spatio-temporal dynamics ⋮ DNN-MG: a hybrid neural network/finite element method with applications to 3D simulations of the Navier-Stokes equations

Uses Software

• OpenFOAM
• redbKIT
• TensorFlow
• Keras
• Adam
• U-Net

Cites Work

• Unnamed Item
• Unnamed Item
• Data driven approximation of parametrized PDEs by reduced basis and neural networks
• Model order reduction: Theory, research aspects and applications. Selected papers based on the presentations at the workshop `Model order reduction, coupled problems and optimization', Leiden, The Netherlands, September 19--23, 2005.
• The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
• Estimating the time-dependent contact rate of SIR and SEIR models in mathematical epidemiology using physics-informed neural networks
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• The finite volume method in computational fluid dynamics. An advanced introduction with OpenFOAM and Matlab
• An artificial neural network as a troubled-cell indicator
• A calculation procedure for heat, mass and momentum transfer in three- dimensional parabolic flows
• A comparison of reduced-order modeling approaches using artificial neural networks for PDEs with bifurcating solutions
• A New Look at Proper Orthogonal Decomposition
• Stationary Flow Predictions Using Convolutional Neural Networks
• Machine Learning in Adaptive Domain Decomposition Methods---Predicting the Geometric Location of Constraints
• Reduced Basis Methods for Partial Differential Equations