Output-based adaptive aerodynamic simulations using convolutional neural networks

DOI10.1016/j.compfluid.2021.104947OpenAlexW3141990187MaRDI QIDQ2245362

Guodong Chen, Krzysztof J. Fidkowski

Publication date: 15 November 2021

Published in: Computers and Fluids (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.compfluid.2021.104947

zbMATH Keywords

mesh adaptationmachine learningconvolutional neural networksoutput error estimation

Mathematics Subject Classification ID

Fluid mechanics (76-XX)

Related Items (3)

Mesh optimization using an improved self-organizing mechanism ⋮ Quasi-optimal \textit{hp}-finite element refinements towards singularities via deep neural network prediction ⋮ Learning Robust Marking Policies for Adaptive Mesh Refinement

Uses Software

• TensorFlow
• Adam
• U-Net
• DeconvNet
• ConvPDE-UQ

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• PDE-constrained optimization with error estimation and control
• \(p\)-multigrid solution of high-order discontinuous Galerkin discretizations of the compressible Navier-Stokes equations
• Predicting goal error evolution from near-initial-information: a learning algorithm
• Adjoint-based \(h\)-\(p\) adaptive discontinuous Galerkin methods for the 2D compressible Euler equations
• Fully anisotropic goal-oriented mesh adaptation for 3D steady Euler equations
• Aerodynamic design via control theory
• Approximate Riemann solvers, parameter vectors, and difference schemes
• Jacobian-free Newton-Krylov methods: a survey of approaches and applications.
• Grid adaptation for functional outputs: application to two-dimensional inviscid flows
• Bayesian calibration of the constants of the \(k-\varepsilon\) turbulence model for a CFD model of street canyon flow
• Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
• Adaptive discontinuous Galerkin finite element methods for the compressible Euler equations.
• Machine-learning error models for approximate solutions to parameterized systems of nonlinear equations
• An optimization-based framework for anisotropic \(hp\)-adaptation of high-order discretizations
• Discretization error control for constrained aerodynamic shape optimization
• ConvPDE-UQ: convolutional neural networks with quantified uncertainty for heterogeneous elliptic partial differential equations on varied domains
• Prediction of aerodynamic flow fields using convolutional neural networks
• A paradigm for data-driven predictive modeling using field inversion and machine learning
• A triangular cut-cell adaptive method for high-order discretizations of the compressible Navier-Stokes equations
• An optimization-based framework for anisotropic simplex mesh adaptation
• The ROMES Method for Statistical Modeling of Reduced-Order-Model Error
• Analysis of Dual Consistency for Discontinuous Galerkin Discretizations of Source Terms
• An Entropy Adjoint Approach to Mesh Refinement
• Adjoint methods for PDEs: a posteriori error analysis and postprocessing by duality
• An optimal control approach to a posteriori error estimation in finite element methods
• Comparison of adjoint-based and feature-based grid adaptation for functional outputs
• Adjoint Recovery of Superconvergent Functionals from PDE Approximations
• Adjoint Consistency Analysis of Discontinuous Galerkin Discretizations
• Learning representations by back-propagating errors
• Turbulence Modeling in the Age of Data

This page was built for publication: Output-based adaptive aerodynamic simulations using convolutional neural networks