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

Title: Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems

Abstract

A machine learning–based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (eg, random forests, and LASSO) to map a large set of inexpensively computed “error indicators” (ie, features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed by simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering) and subsequently constructs a “local” regression model to predict the time-instantaneous error within each identified region of feature space. We consider 2 uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (eg, time-integrated errors). We then apply the proposed framework to model errors in reduced-order models of nonlinear oil-water subsurface flow simulations, with time-varyingmore » well-control (bottom-hole pressure) parameters. The reduced-order models used in this work entail application of trajectory piecewise linearization in conjunction with proper orthogonal decomposition. Moreover, when the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and well-averaged errors.« less

Authors:
ORCiD logo [1];  [2];  [1]
  1. Stanford Univ., CA (United States). Dept. of Energy Resources Engineering
  2. Sandia National Lab. (SNL-CA), Livermore, CA (United States). Extreme-Scae Data Science and Analytics Dept.
Publication Date:
Research Org.:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1399882
Report Number(s):
SAND-2016-12535J
Journal ID: ISSN 0029-5981; 649856; TRN: US1702976
Grant/Contract Number:  
AC04-94AL85000
Resource Type:
Accepted Manuscript
Journal Name:
International Journal for Numerical Methods in Engineering
Additional Journal Information:
Journal Volume: 112; Journal Issue: 12; Journal ID: ISSN 0029-5981
Publisher:
Wiley
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; error modeling; machine learning; nonlinar dynamical system; POD-TPWL; surrogate model

Citation Formats

Trehan, Sumeet, Carlberg, Kevin T., and Durlofsky, Louis J. Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems. United States: N. p., 2017. Web. doi:10.1002/nme.5583.
Trehan, Sumeet, Carlberg, Kevin T., & Durlofsky, Louis J. Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems. United States. doi:10.1002/nme.5583.
Trehan, Sumeet, Carlberg, Kevin T., and Durlofsky, Louis J. Fri . "Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems". United States. doi:10.1002/nme.5583. https://www.osti.gov/servlets/purl/1399882.
@article{osti_1399882,
title = {Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems},
author = {Trehan, Sumeet and Carlberg, Kevin T. and Durlofsky, Louis J.},
abstractNote = {A machine learning–based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (eg, random forests, and LASSO) to map a large set of inexpensively computed “error indicators” (ie, features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed by simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering) and subsequently constructs a “local” regression model to predict the time-instantaneous error within each identified region of feature space. We consider 2 uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (eg, time-integrated errors). We then apply the proposed framework to model errors in reduced-order models of nonlinear oil-water subsurface flow simulations, with time-varying well-control (bottom-hole pressure) parameters. The reduced-order models used in this work entail application of trajectory piecewise linearization in conjunction with proper orthogonal decomposition. Moreover, when the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and well-averaged errors.},
doi = {10.1002/nme.5583},
journal = {International Journal for Numerical Methods in Engineering},
number = 12,
volume = 112,
place = {United States},
year = {2017},
month = {7}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 9 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Reduced Basis Approximation for Nonlinear Parametrized Evolution Equations based on Empirical Operator Interpolation
journal, January 2012

  • Drohmann, Martin; Haasdonk, Bernard; Ohlberger, Mario
  • SIAM Journal on Scientific Computing, Vol. 34, Issue 2
  • DOI: 10.1137/10081157X

Sequential kriging optimization using multiple-fidelity evaluations
journal, May 2006

  • Huang, D.; Allen, T. T.; Notz, W. I.
  • Structural and Multidisciplinary Optimization, Vol. 32, Issue 5
  • DOI: 10.1007/s00158-005-0587-0

Adjoint-Based, Three-Dimensional Error Prediction and Grid Adaptation
journal, September 2004


General-Purpose Nonlinear Model-Order Reduction Using Piecewise-Polynomial Representations
journal, February 2008

  • Ning Dong, ; Roychowdhury, J.
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Vol. 27, Issue 2
  • DOI: 10.1109/TCAD.2007.907272

Response Surface Models Combining Linear and Euler Aerodynamics for Supersonic Transport Design
journal, January 1999

  • Knill, Duane L.; Giunta, Anthony A.; Baker, Chuck A.
  • Journal of Aircraft, Vol. 36, Issue 1
  • DOI: 10.2514/2.2415

Adjoint Error Estimation and Grid Adaptation for Functional Outputs: Application to Quasi-One-Dimensional Flow
journal, October 2000

  • Venditti, David A.; Darmofal, David L.
  • Journal of Computational Physics, Vol. 164, Issue 1
  • DOI: 10.1006/jcph.2000.6600

A priori convergence of the Greedy algorithm for the parametrized reduced basis method
journal, January 2012

  • Buffa, Annalisa; Maday, Yvon; Patera, Anthony T.
  • ESAIM: Mathematical Modelling and Numerical Analysis, Vol. 46, Issue 3
  • DOI: 10.1051/m2an/2011056

Second-Order Corrections for Surrogate-Based Optimization with Model Hierarchies
conference, June 2012

  • Eldred, Michael; Giunta, Anthony; Collis, S.
  • 10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference
  • DOI: 10.2514/6.2004-4457

Quantification of Model Uncertainty: Calibration, Model Discrepancy, and Identifiability
journal, September 2012

  • Arendt, Paul D.; Apley, Daniel W.; Chen, Wei
  • Journal of Mechanical Design, Vol. 134, Issue 10
  • DOI: 10.1115/1.4007390

Trajectory piecewise quadratic reduced-order model for subsurface flow, with application to PDE-constrained optimization
journal, December 2016


A New Look at Proper Orthogonal Decomposition
journal, January 2003


Adaptive h -refinement for reduced-order models : ADAPTIVE
journal, November 2014

  • Carlberg, Kevin
  • International Journal for Numerical Methods in Engineering, Vol. 102, Issue 5
  • DOI: 10.1002/nme.4800

Grid Adaptation for Functional Outputs: Application to Two-Dimensional Inviscid Flows
journal, February 2002

  • Venditti, David A.; Darmofal, David L.
  • Journal of Computational Physics, Vol. 176, Issue 1
  • DOI: 10.1006/jcph.2001.6967

Improved Production Forecasts and History Matching Using Approximate Fluid-Flow Simulators
journal, September 2004

  • Omre, Henning; Lødøen, Ole Petter
  • SPE Journal, Vol. 9, Issue 03
  • DOI: 10.2118/74691-PA

Long Short-Term Memory
journal, November 1997


A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems
journal, January 2015

  • Benner, Peter; Gugercin, Serkan; Willcox, Karen
  • SIAM Review, Vol. 57, Issue 4
  • DOI: 10.1137/130932715

Galerkin v. least-squares Petrov–Galerkin projection in nonlinear model reduction
journal, February 2017


A posteriori error bounds for reduced-basis approximations of parametrized parabolic partial differential equations
journal, January 2005

  • Grepl, Martin A.; Patera, Anthony T.
  • ESAIM: Mathematical Modelling and Numerical Analysis, Vol. 39, Issue 1
  • DOI: 10.1051/m2an:2005006

Accurate Solution of Bayesian Inverse Uncertainty Quantification Problems Combining Reduced Basis Methods and Reduction Error Models
journal, January 2016

  • Manzoni, A.; Pagani, S.; Lassila, T.
  • SIAM/ASA Journal on Uncertainty Quantification, Vol. 4, Issue 1
  • DOI: 10.1137/140995817

Hybrid Variable Fidelity Optimization by Using a Kriging-Based Scaling Function
journal, November 2005

  • Gano, Shawn E.; Renaud, John E.; Sanders, Brian
  • AIAA Journal, Vol. 43, Issue 11
  • DOI: 10.2514/1.12466

The GNAT method for nonlinear model reduction: Effective implementation and application to computational fluid dynamics and turbulent flows
journal, June 2013

  • Carlberg, Kevin; Farhat, Charbel; Cortial, Julien
  • Journal of Computational Physics, Vol. 242
  • DOI: 10.1016/j.jcp.2013.02.028

Global–local nonlinear model reduction for flows in heterogeneous porous media
journal, August 2015

  • Alotaibi, Manal; Calo, Victor M.; Efendiev, Yalchin
  • Computer Methods in Applied Mechanics and Engineering, Vol. 292
  • DOI: 10.1016/j.cma.2014.10.034

Localized Discrete Empirical Interpolation Method
journal, January 2014

  • Peherstorfer, Benjamin; Butnaru, Daniel; Willcox, Karen
  • SIAM Journal on Scientific Computing, Vol. 36, Issue 1
  • DOI: 10.1137/130924408

Multi-fidelity optimization via surrogate modelling
journal, September 2007

  • Forrester, Alexander I. J.; Sóbester, András; Keane, Andy J.
  • Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 463, Issue 2088
  • DOI: 10.1098/rspa.2007.1900

Constraint reduction procedures for reduced-order subsurface flow models based on POD-TPWL: CONSTRAINT REDUCTION FOR POD-TPWL
journal, March 2015

  • He, J.; Durlofsky, L. J.
  • International Journal for Numerical Methods in Engineering, Vol. 103, Issue 1
  • DOI: 10.1002/nme.4874

Regularized kernel PCA for the efficient parameterization of complex geological models
journal, October 2016


Bayesian calibration of computer models
journal, August 2001

  • Kennedy, Marc C.; O'Hagan, Anthony
  • Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 63, Issue 3
  • DOI: 10.1111/1467-9868.00294

Approximation and Model Management in Aerodynamic Optimization with Variable-Fidelity Models
journal, November 2001

  • Alexandrov, Natalia M.; Lewis, Robert Michael; Gumbert, Clyde R.
  • Journal of Aircraft, Vol. 38, Issue 6
  • DOI: 10.2514/2.2877

The ROMES Method for Statistical Modeling of Reduced-Order-Model Error
journal, January 2015

  • Drohmann, Martin; Carlberg, Kevin
  • SIAM/ASA Journal on Uncertainty Quantification, Vol. 3, Issue 1
  • DOI: 10.1137/140969841

Enhanced linearized reduced-order models for subsurface flow simulation
journal, September 2011


Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
journal, October 2016

  • Ling, Julia; Kurzawski, Andrew; Templeton, Jeremy
  • Journal of Fluid Mechanics, Vol. 807
  • DOI: 10.1017/jfm.2016.615

Linearized reduced-order models for subsurface flow simulation
journal, February 2010


A novel evolutionary algorithm applied to algebraic modifications of the RANS stress–strain relationship
journal, November 2016


Provably Convergent Multifidelity Optimization Algorithm Not Requiring High-Fidelity Derivatives
journal, May 2012

  • March, Andrew; Willcox, Karen
  • AIAA Journal, Vol. 50, Issue 5
  • DOI: 10.2514/1.J051125

Efficient non-linear model reduction via a least-squares Petrov-Galerkin projection and compressive tensor approximations
journal, October 2010

  • Carlberg, Kevin; Bou-Mosleh, Charbel; Farhat, Charbel
  • International Journal for Numerical Methods in Engineering, Vol. 86, Issue 2
  • DOI: 10.1002/nme.3050

The post-processing approach in the finite element method—part 1: Calculation of displacements, stresses and other higher derivatives of the displacements
journal, June 1984

  • Babuška, I.; Miller, A.
  • International Journal for Numerical Methods in Engineering, Vol. 20, Issue 6
  • DOI: 10.1002/nme.1620200610

Design and Analysis of "Noisy" Computer Experiments
journal, October 2006

  • Forrester, Alexander I. J.; Keane, Andy J.; Bressloff, Neil W.
  • AIAA Journal, Vol. 44, Issue 10
  • DOI: 10.2514/1.20068

A trajectory piecewise-linear approach to model order reduction and fast simulation of nonlinear circuits and micromachined devices
journal, February 2003

  • Rewienski, M.; White, J.
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Vol. 22, Issue 2
  • DOI: 10.1109/TCAD.2002.806601

Complexity Reduction of Multiphase Flows in Heterogeneous Porous Media
journal, February 2016

  • Ghommem, Mehdi; Gildin, Eduardo; Ghasemi, Mohammadreza
  • SPE Journal, Vol. 21, Issue 01
  • DOI: 10.2118/167295-PA

EM-ANN models for microstrip vias and interconnects in dataset circuits
journal, January 1996

  • Watson, P. M.; Gupta, K. C.
  • IEEE Transactions on Microwave Theory and Techniques, Vol. 44, Issue 12
  • DOI: 10.1109/22.554584

Interpretation of Well-Block Pressures in Numerical Reservoir Simulation With Nonsquare Grid Blocks and Anisotropic Permeability
journal, June 1983

  • Peaceman, D. W.
  • Society of Petroleum Engineers Journal, Vol. 23, Issue 03
  • DOI: 10.2118/10528-PA

Application of POD and DEIM on dimension reduction of non-linear miscible viscous fingering in porous media
journal, August 2011

  • Chaturantabut, Saifon; Sorensen, Danny C.
  • Mathematical and Computer Modelling of Dynamical Systems, Vol. 17, Issue 4
  • DOI: 10.1080/13873954.2011.547660

Reduced-Order Modeling for Compositional Simulation by Use of Trajectory Piecewise Linearization
journal, October 2014

  • He, Jincong; Durlofsky, Louis J.
  • SPE Journal, Vol. 19, Issue 05
  • DOI: 10.2118/163634-PA

Random Forests
journal, January 2001