An adaptive knowledge-based data-driven approach for turbulence modeling using ensemble learning technique under complex flow configuration: 3D PWR sub-channel with DNS data

Zhu, Yangmo; Dinh, Nam T.; Saini, Nadish; Bolotnov, Igor A.

doi:10.1016/j.nucengdes.2022.111814

An adaptive knowledge-based data-driven approach for turbulence modeling using ensemble learning technique under complex flow configuration: 3D PWR sub-channel with DNS data

Journal Article · Tue May 17 00:00:00 EDT 2022 · Nuclear Engineering and Design

DOI:https://doi.org/10.1016/j.nucengdes.2022.111814· OSTI ID:1977511

Zhu, Yangmo ^[1]; Dinh, Nam T. ^[2]; Saini, Nadish ^[2]; Bolotnov, Igor A. ^[2]

North Carolina State University, Raleigh, NC (United States); OSTI
North Carolina State University, Raleigh, NC (United States)

This work describes a new approach to increase the accuracy of Reynolds-averaged Navier–Stokes (RANS) in modeling turbulence flow leveraging the machine learning technique. Traditionally, different turbulence models for Reynolds stress are developed for different flow patterns based on human knowledge. Each turbulence model has a certain application domain and prediction uncertainty. In recent years, with the rapid improvements of machine learning techniques, researchers start to develop an approach to compensate for the prediction discrepancy of traditional turbulence models with statistical models and data. However, the approach has deficiencies in several aspects. For example, the amount of human knowledge introduced to the statistical model couldn’t be controlled, which makes the statistical model learn from a very naïve stage and limits its application. In this work, a new approach is developed to address those deficiencies. Here, the new approach uses the “ensemble learning” technique to control the amount of human knowledge introduced into the statistical model. Therefore, the new approach could be adaptive to the multiple application domains. In conclusion, according to the results of case study, the new approach shows higher accuracy than both traditional turbulence models and the previous machine learning approach.

View Accepted Manuscript (DOE)

Research Organization:: North Carolina State University, Raleigh, NC (United States)

Sponsoring Organization:: USDOE; USDOE Office of Nuclear Energy (NE), Nuclear Energy University Program (NEUP)

Grant/Contract Number:: NE0008530

OSTI ID:: 1977511

Alternate ID(s):: OSTI ID: 1869323

Journal Information:: Nuclear Engineering and Design, Journal Name: Nuclear Engineering and Design Vol. 393; ISSN 0029-5493

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (21)

A Priori Assessment of Prediction Confidence for Data-Driven Turbulence Modeling Wu, Jin-Long; Wang, Jian-Xun; Xiao, Heng Flow, Turbulence and Combustion, Vol. 99, Issue 1 https://doi.org/10.1007/s10494-017-9807-0	journal	March 2017
Occam's Razor Blumer, Anselm; Ehrenfeucht, Andrzej; Haussler, David Information Processing Letters, Vol. 24, Issue 6 https://doi.org/10.1016/0020-0190(87)90114-1	journal	April 1987
Turbulence modelling for large volumetrically heated liquid pools Dinh, T. N.; Nourgaliev, R. R. Nuclear Engineering and Design, Vol. 169, Issue 1-3 https://doi.org/10.1016/S0029-5493(96)01281-2	journal	June 1997
Data-driven modeling for boiling heat transfer: Using deep neural networks and high-fidelity simulation results Liu, Yang; Dinh, Nam; Sato, Yohei Applied Thermal Engineering, Vol. 144 https://doi.org/10.1016/j.applthermaleng.2018.08.041	journal	November 2018
Evaluation of bubble-induced turbulence using direct numerical simulation Feng, Jinyong; Bolotnov, Igor A. International Journal of Multiphase Flow, Vol. 93 https://doi.org/10.1016/j.ijmultiphaseflow.2017.04.003	journal	July 2017
Slug-to-churn vertical two-phase flow regime transition study using an interface tracking approach Zimmer, Matthew D.; Bolotnov, Igor A. International Journal of Multiphase Flow, Vol. 115 https://doi.org/10.1016/j.ijmultiphaseflow.2019.04.003	journal	June 2019
A paradigm for data-driven predictive modeling using field inversion and machine learning Parish, Eric J.; Duraisamy, Karthik Journal of Computational Physics, Vol. 305 https://doi.org/10.1016/j.jcp.2015.11.012	journal	January 2016
Machine learning strategies for systems with invariance properties Ling, Julia; Jones, Reese; Templeton, Jeremy Journal of Computational Physics, Vol. 318 https://doi.org/10.1016/j.jcp.2016.05.003	journal	August 2016
A turbulence model study for simulating flow inside tight lattice rod bundles Baglietto, E.; Ninokata, H. Nuclear Engineering and Design, Vol. 235, Issue 7 https://doi.org/10.1016/j.nucengdes.2004.10.007	journal	March 2005
Direct numerical simulation of reactor two-phase flows enabled by high-performance computing Fang, Jun; Cambareri, Joseph J.; Brown, Cameron S. Nuclear Engineering and Design, Vol. 330 https://doi.org/10.1016/j.nucengdes.2018.02.024	journal	April 2018
A data-driven framework for error estimation and mesh-model optimization in system-level thermal-hydraulic simulation Bao, Han; Dinh, Nam T.; Lane, Jeffrey W. Nuclear Engineering and Design, Vol. 349 https://doi.org/10.1016/j.nucengdes.2019.04.023	journal	August 2019
Interface capturing simulations of droplet interaction with spacer grids under DFFB conditions Saini, Nadish; Bolotnov, Igor A. Nuclear Engineering and Design, Vol. 364 https://doi.org/10.1016/j.nucengdes.2020.110685	journal	August 2020
Interface capturing simulations of bubble population effects in PWR subchannels Cambareri, Joseph J.; Fang, Jun; Bolotnov, Igor A. Nuclear Engineering and Design, Vol. 365 https://doi.org/10.1016/j.nucengdes.2020.110709	journal	August 2020
Progress in the development of a Reynolds-stress turbulence closure Launder, B. E.; Reece, G. J.; Rodi, W. Journal of Fluid Mechanics, Vol. 68, Issue 3 https://doi.org/10.1017/S0022112075001814	journal	April 1975
Using statistical learning to close two-fluid multiphase flow equations for a simple bubbly system Ma, Ming; Lu, Jiacai; Tryggvason, Gretar Physics of Fluids, Vol. 27, Issue 9 https://doi.org/10.1063/1.4930004	journal	September 2015
Rod Bundle Heat Transfer Thermal-Hydraulic Program Bajorek, Stephen M.; Cheung, Fan-Bill Nuclear Technology, Vol. 205, Issue 1-2 https://doi.org/10.1080/00295450.2018.1510697	journal	July 2018
Physics-informed machine learning approach for reconstructing Reynolds stress modeling discrepancies based on DNS data Wang, Jian-Xun; Wu, Jin-Long; Xiao, Heng Physical Review Fluids, Vol. 2, Issue 3 https://doi.org/10.1103/PhysRevFluids.2.034603	journal	March 2017
Direct numerical simulations of multifluid flows in a vertical channel undergoing topology changes Lu, Jiacai; Tryggvason, Gretar Physical Review Fluids, Vol. 3, Issue 8 https://doi.org/10.1103/PhysRevFluids.3.084401	journal	August 2018
Heat-Transfer Augmentation in Rod Bundles Near Grid Spacers Yao, S. C.; Hochreiter, L. E.; Leech, W. J. Journal of Heat Transfer, Vol. 104, Issue 1 https://doi.org/10.1115/1.3245071	journal	February 1982
Experimental Studies of Spacer Grid Thermal Hydraulics in the Dispersed Flow Film Boiling Regime Riley, M. P.; Mohanta, L.; Cheung, F. B. Nuclear Technology, Vol. 190, Issue 3 https://doi.org/10.13182/NT14-80	journal	June 2015
Quantification of Structural Uncertainties in the k -w Turbulence Model Dow, Eric; Wang, Qiqi 52nd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference https://doi.org/10.2514/6.2011-1762	conference	June 2012

Similar Records

Machine Learning for Turbulence Modeling

Technical Report · Tue Jan 31 23:00:00 EST 2017 · OSTI ID:1761814

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

Journal Article · Mon Aug 03 20:00:00 EDT 2015 · Physics of Fluids · OSTI ID:1235329

Physics-Informed Machine Learning for Predictive Turbulence Modeling: Towards a Complete Framework

Technical Report · Sat Sep 17 00:00:00 EDT 2016 · OSTI ID:1562229

Related Subjects

42 ENGINEERING
RANS
ensemble learning
machine learning
turbulence

An adaptive knowledge-based data-driven approach for turbulence modeling using ensemble learning technique under complex flow configuration: 3D PWR sub-channel with DNS data

Citation Formats

References (21)

Similar Records

Related Subjects