skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty

Journal Article · · Physics of Fluids
DOI:https://doi.org/10.1063/1.4927765· OSTI ID:1235329
 [1];  [1]
  1. Sandia National Lab. (SNL-CA), Livermore, CA (United States)
+ Show Author Affiliations

Reynolds Averaged Navier Stokes (RANS) models are widely used in industry to predict fluid flows, despite their acknowledged deficiencies. Not only do RANS models often produce inaccurate flow predictions, but there are very limited diagnostics available to assess RANS accuracy for a given flow configuration. If experimental or higher fidelity simulation results are not available for RANS validation, there is no reliable method to evaluate RANS accuracy. This paper explores the potential of utilizing machine learning algorithms to identify regions of high RANS uncertainty. Three different machine learning algorithms were evaluated: support vector machines, Adaboost decision trees, and random forests. The algorithms were trained on a database of canonical flow configurations for which validated direct numerical simulation or large eddy simulation results were available, and were used to classify RANS results on a point-by-point basis as having either high or low uncertainty, based on the breakdown of specific RANS modeling assumptions. Classifiers were developed for three different basic RANS eddy viscosity model assumptions: the isotropy of the eddy viscosity, the linearity of the Boussinesq hypothesis, and the non-negativity of the eddy viscosity. It is shown that these classifiers are able to generalize to flows substantially different from those on which they were trained. As a result, feature selection techniques, model evaluation, and extrapolation detection are discussed in the context of turbulence modeling applications.

Research Organization:
Sandia National Lab. (SNL-CA), Livermore, CA (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA)
Grant/Contract Number:
AC04-94AL85000
OSTI ID:
1235329
Report Number(s):
SAND-2015-2173J; PHFLE6; 579403
Journal Information:
Physics of Fluids, Vol. 27, Issue 8; ISSN 1070-6631
Publisher:
American Institute of PhysicsCopyright Statement
Country of Publication:
United States
Language:
English
Citation Metrics:
Cited by: 200 works
Citation information provided by
Web of Science

References (43)

Bayesian uncertainty quantification applied to RANS turbulence models journal December 2011
Bayesian uncertainty analysis with applications to turbulence modeling journal September 2011
Bayesian estimates of parameter variability in the k–ε turbulence model journal February 2014
Tuning a RANS k-e model for jet-in-crossflow simulations report September 2013
Turbulent Flows book July 2012
On predicting the turbulence‐induced secondary flows using nonlinear k ‐ε models journal July 1996
Development and application of a cubic eddy-viscosity model of turbulence journal April 1996
Progress in the development of a Reynolds-stress turbulence closure journal April 1975
An explicit algebraic Reynolds stress model for incompressible and compressible turbulent flows journal January 2000
On explicit algebraic stress models for complex turbulent flows journal September 1993
Turbulent channel and Couette flows using an anisotropic k-epsilon model journal March 1987
RANS modeling of turbulent mixing for a jet in supersonic cross flow: model evaluation and uncertainty quantification conference January 2012
On the non-local geometry of turbulence journal April 2008
Quantification of Structural Uncertainties in the k -w Turbulence Model conference June 2012
Uncertainty Quantification of Structural Uncertainties in RANS Simulations of Complex Flows conference June 2012
New Approaches in Turbulence and Transition Modeling Using Data-driven Techniques conference January 2015
Application of Supervised Learning to Quantify Uncertainties in Turbulence and Combustion Modeling
  • Tracey, Brendan; Duraisamy, Karthik; Alonso, Juan
  • 51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition https://doi.org/10.2514/6.2013-259
conference January 2013
A Machine Learning Strategy to Assist Turbulence Model Development conference January 2015
A numerical study of algebraic flux models for heat and mass transport simulation in complex flows journal October 2010
A numerical study of scalar dispersion downstream of a wall-mounted cube using direct simulations and algebraic flux models journal October 2010
Numerical analysis and modeling of plume meandering in passive scalar dispersion downstream of a wall-mounted cube journal October 2013
Near Wall Modeling for Trailing Edge Slot Film Cooling journal September 2014
Turbulent transport in an inclined jet in crossflow journal October 2013
Direct numerical simulation of turbulent channel flow up to Reτ=590 journal April 1999
Reynolds number dependence of mean flow structure in square duct turbulence journal February 2010
Flow topologies and turbulence scales in a jet-in-cross-flow journal April 2015
Bayesian calibration of a k-ε turbulence model for predictive jet-in-crossflow simulations conference June 2014
Estimation of k-ε parameters using surrogate models and jet-in-crossflow data report November 2014
SIERRA/Fuego: A Multi-Mechanics Fire Environment Simulation Tool conference November 2012
Validation of an FSI Modeling Framework for Internal Captive Carriage Applications conference May 2013
Random Forest:  A Classification and Regression Tool for Compound Classification and QSAR Modeling journal November 2003
Distributed Pasting of Small Votes book January 2002
Creating ensembles of classifiers conference January 2001
The return to isotropy of homogeneous turbulence journal June 2001
Anisotropy analysis of turbulent swirl flow journal January 2014
A more general effective-viscosity hypothesis journal November 1975
Combination of multiple classifiers using local accuracy estimates
  • Woods, K.; Bowyer, K.; Kegelmeyer, W. P.
  • Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.1996.517102
conference January 1996
A Comparison of Ensemble Creation Techniques book January 2004
Diagnosing extrapolation: tree-based density estimation conference January 2004
Analysis of Turbulent Scalar Flux Models for a Discrete Hole Film Cooling Flow journal October 2015
miRNALoc: predicting miRNA subcellular localizations based on principal component scores of physico-chemical properties and pseudo compositions of di-nucleotides journal September 2020
Turbulent Flows journal October 2001
Combination of multiple classifiers using local accuracy estimates journal April 1997

Cited By (35)

Uncertainty and global sensitivity analysis of bladed disk statics with material anisotropy and root geometry variations journal October 2019
Prediction of Reynolds stresses in high-Mach-number turbulent boundary layers using physics-informed machine learning journal December 2018
A Priori Assessment of Prediction Confidence for Data-Driven Turbulence Modeling journal March 2017
Zonal Eddy Viscosity Models Based on Machine Learning journal February 2019
The Current State of High-Fidelity Simulations for Main Gas Path Turbomachinery Components and Their Industrial Impact journal February 2019
Progress Variable Variance and Filtered Rate Modelling Using Convolutional Neural Networks and Flamelet Methods journal April 2019
Reynolds averaged turbulence modelling using deep neural networks with embedded invariance journal October 2016
A neural network approach for the blind deconvolution of turbulent flows journal October 2017
Subgrid modelling for two-dimensional turbulence using neural networks journal November 2018
Deep learning of vortex-induced vibrations journal December 2018
Sub-grid scale model classification and blending through deep learning journal May 2019
Construction of reduced-order models for fluid flows using deep feedforward neural networks journal June 2019
Prediction of turbulent heat transfer using convolutional neural networks journal November 2019
Hydrodynamic object identification with artificial neural models journal August 2019
Using field inversion to quantify functional errors in turbulence closures journal April 2016
Prediction model of velocity field around circular cylinder over various Reynolds numbers by fusion convolutional neural networks based on pressure on the cylinder journal April 2018
Machine learning methods for turbulence modeling in subsonic flows around airfoils journal January 2019
Fast flow field prediction over airfoils using deep learning approach journal May 2019
Application of two-branch deep neural network to predict bubble migration near elastic boundaries journal October 2019
Direct Numerical Simulation of Turbulent Flow Inside a Differentially Heated Composite Cavity journal May 2019
Reynolds-Averaged Turbulence Modeling Using Deep Learning with Local Flow Features: An Empirical Approach journal February 2020
Modeling subgrid-scale force and divergence of heat flux of compressible isotropic turbulence by artificial neural network journal October 2019
A Machine Learning Approach for Determining the Turbulent Diffusivity in Film Cooling Flows journal December 2017
Kernel-Smoothed Proper Orthogonal Decomposition–Based Emulation for Spatiotemporally Evolving Flow Dynamics Prediction journal December 2019
Future Directions in Computational Fluid Dynamics conference June 2017
Towards Integrated Field Inversion and Machine Learning With Embedded Neural Networks for RANS Modeling conference January 2019
A Machine Learning Approach for Determining the Turbulent Diffusivity in Film Cooling Flows
  • Milani, Pedro M.; Ling, Julia; Saez-Mischlich, Gonzalo
  • ASME Turbo Expo 2017: Turbomachinery Technical Conference and Exposition, Volume 5A: Heat Transfer https://doi.org/10.1115/gt2017-63299
conference August 2017
Deep Learning of Vortex Induced Vibrations text January 2018
A neural network approach for the blind deconvolution of turbulent flows text January 2017
Physics-Informed Machine Learning Approach for Augmenting Turbulence Models: A Comprehensive Framework text January 2018
Prediction of Reynolds Stresses in High-Mach-Number Turbulent Boundary Layers using Physics-Informed Machine Learning text January 2018
Sub-grid scale model classification and blending through deep learning text January 2018
Construction of Reduced Order Models for Fluid Flows Using Deep Feedforward Neural Networks text January 2019
Modal Analysis of Fluid Flows: Applications and Outlook preprint January 2019
Hydrodynamic object identification with artificial neural models text January 2019

Similar Records

Adaptive downstream tensorial eddy viscosity for hybrid large‐eddy simulations‐Reynolds‐averaged‐Navier–Stokes simulations
Journal Article · Tue Feb 02 00:00:00 EST 2021 · International Journal for Numerical Methods in Fluids · OSTI ID:1235329
Machine Learning Assisted Safety Modeling and Analysis of Advanced Reactors
Technical Report · Mon Dec 20 00:00:00 EST 2021 · OSTI ID:1235329
+2 more
Simulation of Wingtip Vortex Flows with Reynolds-Averaged Navier–Stokes and Scale-Resolving Simulation Methods
Journal Article · Fri Jan 11 00:00:00 EST 2019 · AIAA Journal · OSTI ID:1235329

Related Subjects

97 MATHEMATICS AND COMPUTING
Reynolds stress modeling
Navier Stokes equations
viscosity
databases
Eddy viscosity closure