A neural network approach for the blind deconvolution of turbulent flows

DOI10.1017/jfm.2017.637zbMath1421.76134arXiv1706.00912OpenAlexW3098093095MaRDI QIDQ5231546

Romit Maulik, Omer San

Publication date: 27 August 2019

Published in: Journal of Fluid Mechanics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1706.00912

zbMATH Keywords

computational methodsturbulence modelling

Mathematics Subject Classification ID

Related Items (32)

DeepMoD: deep learning for model discovery in noisy data ⋮ A robust super-resolution reconstruction model of turbulent flow data based on deep learning ⋮ Subgrid modelling for two-dimensional turbulence using neural networks ⋮ Data-Driven Filtered Reduced Order Modeling of Fluid Flows ⋮ A machine learning framework for LES closure terms ⋮ Invariant data-driven subgrid stress modeling in the strain-rate eigenframe for large eddy simulation ⋮ Hybrid analysis and modeling, eclecticism, and multifidelity computing toward digital twin revolution ⋮ A perspective on machine learning methods in turbulence modeling ⋮ Physics-informed regularization and structure preservation for learning stable reduced models from data with operator inference ⋮ Variational multiscale super‐resolution: A data‐driven approach for reconstruction and predictive modeling of unresolved physics ⋮ A highly accurate strategy for data-driven turbulence modeling ⋮ Invariant data-driven subgrid stress modeling on anisotropic grids for large eddy simulation ⋮ Data-driven wall modeling for turbulent separated flows ⋮ On the improvement of the extrapolation capability of an iterative machine-learning based RANS framework ⋮ The anisotropic graph neural network model with multiscale and nonlinear characteristic for turbulence simulation ⋮ An artificial neural network framework for reduced order modeling of transient flows ⋮ Sub-grid scale model classification and blending through deep learning ⋮ Using machine learning to detect the turbulent region in flow past a circular cylinder ⋮ Interpreting neural network models of residual scalar flux ⋮ Unsupervised deep learning for super-resolution reconstruction of turbulence ⋮ A turbulent eddy-viscosity surrogate modeling framework for Reynolds-averaged Navier-Stokes simulations ⋮ Prediction of turbulent heat transfer using convolutional neural networks ⋮ Reynolds-averaged Navier–Stokes equations with explicit data-driven Reynolds stress closure can be ill-conditioned ⋮ Nonlinear mode decomposition with convolutional neural networks for fluid dynamics ⋮ Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression ⋮ The use of the Reynolds force vector in a physics informed machine learning approach for predictive turbulence modeling ⋮ Subgrid-scale model for large-eddy simulation of isotropic turbulent flows using an artificial neural network ⋮ Toward neural-network-based large eddy simulation: application to turbulent channel flow ⋮ Kinetic-energy-flux-constrained model using an artificial neural network for large-eddy simulation of compressible wall-bounded turbulence ⋮ A physics-inspired alternative to spatial filtering for large-eddy simulations of turbulent flows ⋮ A unified understanding of scale-resolving simulations and near-wall modelling of turbulent flows using optimal finite-element projections ⋮ Spatiotemporally dynamic implicit large eddy simulation using machine learning classifiers

Uses Software

• ICALAB

Cites Work

• Discovering governing equations from data by sparse identification of nonlinear dynamical systems
• Approximate deconvolution models of turbulence. Analysis, phenomenology and numerical analysis.
• Machine learning strategies for systems with invariance properties
• A simple and stable scale-similarity model for large eddy simulation: Energy balance and existence of weak solutions
• A novel evolutionary algorithm applied to algebraic modifications of the RANS stress-strain relationship
• Hidden physics models: machine learning of nonlinear partial differential equations
• Multilayer feedforward networks are universal approximators
• Neural network modeling for near wall turbulent flow.
• Convergence tests for the piecewise parabolic method and Navier-Stokes solutions for homogeneous compressible turbulence
• High-order methods for decaying two-dimensional homogeneous isotropic turbulence
• A paradigm for data-driven predictive modeling using field inversion and machine learning
• Explicit and implicit LES closures for Burgers turbulence
• Regression and the Moore-Penrose pseudoinverse
• Compressive sensing based machine learning strategy for characterizing the flow around a cylinder with limited pressure measurements
• An approximate deconvolution procedure for large-eddy simulation
• The approximate deconvolution model for large-eddy simulations of compressible flows and its application to shock-turbulent-boundary-layer interaction
• Matrices
• Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
• Deep learning in fluid dynamics

This page was built for publication: A neural network approach for the blind deconvolution of turbulent flows