Data-driven modelling of the Reynolds stress tensor using random forests with invariance

DOI10.1016/j.compfluid.2020.104497OpenAlexW3012045275MaRDI QIDQ2180004

Mikael L. A. Kaandorp, Richard P. Dwight

Publication date: 13 May 2020

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

Full work available at URL: https://arxiv.org/abs/1810.08794

zbMATH Keywords

random foreststurbulence modellingmachine-learningnon-linear eddy-viscosity closuresReynolds anisotropy tensor

Mathematics Subject Classification ID

Fluid mechanics (76-XX)

Related Items (11)

On the Generalizability of Machine-Learning-Assisted Anisotropy Mappings for Predictive Turbulence Modelling ⋮ Uncertainty quantification for data-driven turbulence modelling with Mondrian forests ⋮ Customized data-driven RANS closures for bi-fidelity LES-RANS optimization ⋮ Multi-objective CFD-driven development of coupled turbulence closure models ⋮ \(S\)-frame discrepancy correction models for data-informed Reynolds stress closure ⋮ Stable \textit{a posteriori} LES of 2D turbulence using convolutional neural networks: backscattering analysis and generalization to higher \(Re\) via transfer learning ⋮ Frame Invariance and Scalability of Neural Operators for Partial Differential Equations ⋮ A highly accurate strategy for data-driven turbulence modeling ⋮ Data-driven RANS closures for three-dimensional flows around bluff bodies ⋮ Conditioning and accurate solutions of Reynolds average Navier–Stokes equations with data-driven turbulence closures ⋮ Data-driven RANS closures for wind turbine wakes under neutral conditions

Uses Software

• BRENT
• PMTK
• mftoolbox
• BartPy

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• BART: Bayesian additive regression trees
• Predictive RANS simulations via Bayesian model-scenario averaging
• Flow over periodic hills -- numerical and experimental study in a wide range of Reynolds numbers
• Machine learning strategies for systems with invariance properties
• The numerical computation of turbulent flows
• Jacobian-free Newton-Krylov methods: a survey of approaches and applications.
• A methodology to evaluate statistical errors in DNS data of plane channel flows
• A novel evolutionary algorithm applied to algebraic modifications of the RANS stress-strain relationship
• A random matrix approach for quantifying model-form uncertainties in turbulence modeling
• A paradigm for data-driven predictive modeling using field inversion and machine learning
• Presentation of anisotropy properties of turbulence, invariants versus eigenvalue approaches
• Robustness of random forests for regression
• Modelling the pressure–strain correlation of turbulence: an invariant dynamical systems approach
• Reynolds number dependence of mean flow structure in square duct turbulence
• Computational Modeling of Turbulent Flows
• A more general effective-viscosity hypothesis
• The return to isotropy of homogeneous turbulence
• Direct numerical simulation of turbulent flow over a backward-facing step
• Turbulent Flows
• Reynolds-averaged Navier–Stokes equations with explicit data-driven Reynolds stress closure can be ill-conditioned
• Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
• Functions of Matrices

This page was built for publication: Data-driven modelling of the Reynolds stress tensor using random forests with invariance