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

Abstract

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 whichmore » they were trained. As a result, feature selection techniques, model evaluation, and extrapolation detection are discussed in the context of turbulence modeling applications.« less

Authors:
 [1];  [1]
+ Show Author Affiliations
  1. Sandia National Lab. (SNL-CA), Livermore, CA (United States)
Publication Date:
Research Org.:
Sandia National Lab. (SNL-CA), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1235329
Report Number(s):
SAND-2015-2173J
Journal ID: ISSN 1070-6631; PHFLE6; 579403
Grant/Contract Number:
AC04-94AL85000
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Physics of Fluids (1994)
Additional Journal Information:
Journal Name: Physics of Fluids (1994); Journal Volume: 27; Journal Issue: 8; Journal ID: ISSN 1070-6631
Publisher:
American Institute of Physics
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Reynolds stress modeling; Navier Stokes equations; viscosity; databases; Eddy viscosity closure

Citation Formats

Ling, Julia, and Templeton, Jeremy Alan. Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty. United States: N. p., 2015. Web. doi:10.1063/1.4927765.
Copy to clipboard
Ling, Julia, & Templeton, Jeremy Alan. Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty. United States. doi:10.1063/1.4927765.
Copy to clipboard
Ling, Julia, and Templeton, Jeremy Alan. Tue . "Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty". United States. doi:10.1063/1.4927765. https://www.osti.gov/servlets/purl/1235329.
Copy to clipboard
@article{osti_1235329,
title = {Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty},
author = {Ling, Julia and Templeton, Jeremy Alan},
abstractNote = {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.},
doi = {10.1063/1.4927765},
journal = {Physics of Fluids (1994)},
number = 8,
volume = 27,
place = {United States},
year = {Tue Aug 04 00:00:00 EDT 2015},
month = {Tue Aug 04 00:00:00 EDT 2015}
}
Copy to clipboard
Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
DOI:  10.1063/1.4927765
Copyright Statement
Other availability
Search WorldCat Search WorldCat to find libraries that may hold this journal
Citation Metrics:
Cited by: 20works
Citation information provided by
Web of Science
Save / Share:
Export Metadata
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.

  • ; ; ; ... - Boundary-Layer Meteorology
    Three computational fluid dynamics (CFD) methods with different levels of flow-physics modelling are comprehensively evaluated against high-spatial-resolution wind-tunnel velocity data from step-down street canyons (i.e., a short building downwind of a tall building). The first method is a semi-empirical fast-response approach using the Quick Urban Industrial Complex (QUIC-URB) model. The second method solves the Reynolds-averaged Navier–Stokes (RANS) equations, and the third one utilizes a fully-coupled fluid-structure interaction large-eddy simulation (LES) model with a grid-turbulence inflow generator. Unlike typical point-by-point evaluation comparisons, here the entire two-dimensional wind-tunnel dataset is used to evaluate the dynamics of dominant flow topological features in themore » street canyon. Each CFD method is scrutinized for several geometric configurations by varying the downwind-to-upwind building-height ratio (H d/H u) and street canyon-width to building-width aspect ratio (S / W) for inflow winds perpendicular to the upwind building front face. Disparities between the numerical results and experimental data are quantified in terms of their ability to capture flow topological features for different geometric configurations. Ultimately, all three methods qualitatively predict the primary flow topological features, including a saddle point and a primary vortex. But, the secondary flow topological features, namely an in-canyon separation point and secondary vortices, are only well represented by the LES method despite its failure for taller downwind building cases. Misrepresentation of flow-regime transitions, exaggeration of the coherence of recirculation zones and wake fields, and overestimation of downwards vertical velocity into the canyon are the main defects in QUIC-URB, RANS and LES results, respectively. All three methods underestimate the updrafts and, surprisingly, QUIC-URB outperforms RANS for the streamwise velocity component, while RANS is superior to QUIC-URB for the vertical velocity component in the street canyon.« less
    Cited by 2

  • ; - Computers and Fluids

  • ; ; - Combustion and Flame
    Numerical simulation results are presented for two axisymmetric, nonluminous turbulent piloted jet diffusion flames: Sandia Flame D (SFD) and Delft Flame III (DFIII). Turbulence is represented by a Reynolds stress transport model, while chemistry is modeled by means of steady laminar flamelets. We use the preassumed PDF approach for turbulence-chemistry interaction. A weighted sum of gray gases model is used for the gas radiative properties. The radiative transfer equation is solved using the discrete ordinates method in the conservative finite-volume formulation. The radiative loss leads to a decrease in mean temperature, but does not significantly influence the flow and mixingmore » fields, in terms either of mean values or of rms values of fluctuations. A systematic analysis of turbulence-radiation interaction (TRI) is carried out. By considering five different TRI formulations, and comparing also with a simple optically thin model, individual TRI contributions are isolated and quantified. For both flames, effects are demonstrated of (1) influence of temperature fluctuations on the mean Planck function, (2) temperature and composition fluctuations on the mean absorption coefficient, and (3) correlation between absorption coefficient and Planck function. The strength of the last effect is stronger in DFIII than in SFD, because of stronger turbulence-chemistry interaction and lower mean temperature in DFIII. The impact of the choice of TRI model on the prediction of the temperature-sensitive minor species NO is determined in a postprocessing step with fixed flow and mixing fields. Best agreement for NO is obtained using the most complete representation of TRI. (author)« less

  • ; ; ; ... - Journal of Computational Physics
    Despite their well-known limitations, Reynolds-Averaged Navier–Stokes (RANS) models are still the workhorse tools for turbulent flow simulations in today's engineering analysis, design and optimization. While the predictive capability of RANS models depends on many factors, for many practical flows the turbulence models are by far the largest source of uncertainty. As RANS models are used in the design and safety evaluation of many mission-critical systems such as airplanes and nuclear power plants, quantifying their model-form uncertainties has significant implications in enabling risk-informed decision-making. In this work we develop a data-driven, physics-informed Bayesian framework for quantifying model-form uncertainties in RANS simulations.more » Uncertainties are introduced directly to the Reynolds stresses and are represented with compact parameterization accounting for empirical prior knowledge and physical constraints (e.g., realizability, smoothness, and symmetry). An iterative ensemble Kalman method is used to assimilate the prior knowledge and observation data in a Bayesian framework, and to propagate them to posterior distributions of velocities and other Quantities of Interest (QoIs). We use two representative cases, the flow over periodic hills and the flow in a square duct, to evaluate the performance of the proposed framework. Both cases are challenging for standard RANS turbulence models. Simulation results suggest that, even with very sparse observations, the obtained posterior mean velocities and other QoIs have significantly better agreement with the benchmark data compared to the baseline results. At most locations the posterior distribution adequately captures the true model error within the developed model form uncertainty bounds. The framework is a major improvement over existing black-box, physics-neutral methods for model-form uncertainty quantification, where prior knowledge and details of the models are not exploited. This approach has potential implications in many fields in which the governing equations are well understood but the model uncertainty comes from unresolved physical processes. - Highlights: • Proposed a physics–informed framework to quantify uncertainty in RANS simulations. • Framework incorporates physical prior knowledge and observation data. • Based on a rigorous Bayesian framework yet fully utilizes physical model. • Applicable for many complex physical systems beyond turbulent flows.« less

  • ; ; ; ... - Water Resources Research
    Climate, groundwater extraction, and surface water flows have complex nonlinear relationships with groundwater level in agricultural regions. To better understand the relative importance of each driver and predict groundwater level change, we develop a new ensemble modeling framework based on spectral analysis, machine learning, and uncertainty analysis, as an alternative to complex and computationally expensive physical models. We apply and evaluate this new approach in the context of two aquifer systems supporting agricultural production in the United States: the High Plains aquifer (HPA) and the Mississippi River Valley alluvial aquifer (MRVA). We select input data sets by using a combinationmore » of mutual information, genetic algorithms, and lag analysis, and then use the selected data sets in a Multilayer Perceptron network architecture to simulate seasonal groundwater level change. As expected, model results suggest that irrigation demand has the highest influence on groundwater level change for a majority of the wells. The subset of groundwater observations not used in model training or cross-validation correlates strongly (R > 0.8) with model results for 88 and 83% of the wells in the HPA and MRVA, respectively. In both aquifer systems, the error in the modeled cumulative groundwater level change during testing (2003-2012) was less than 2 m over a majority of the area. Here, we conclude that our modeling framework can serve as an alternative approach to simulating groundwater level change and water availability, especially in regions where subsurface properties are unknown.« less