Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem

Abstract A non-intrusive reduced-basis (RB) method is proposed for parametrized unsteady flows. A set of reduced basis functions are extracted from a collection of high-fidelity solutions via a proper orthogonal decomposition (POD), and the coefficients of the reduced basis functions are recovered by a feedforward neural network (NN). As a regression model of the RB method for unsteady flows, the neural network approximates the map between the time/parameter value and the projection coefficients of the high-fidelity solution onto the reduced space. The generation of the reduced basis and the training of the NN are accomplished in the offline stage, thus the RB solution of a new time/parameter value can be recovered via direct outputs of the NN in the online stage. Due to its non-intrusive nature, the proposed RB method, referred as the POD-NN, fully decouples the online stage and the high-fidelity scheme, and is thus able to provide fast and reliable solutions of complex unsteady flows. To test this assertion, the POD-NN method is applied to the reduced order modeling (ROM) of the quasi-one dimensional Continuously Variable Resonance Combustor (CVRC) flow. Numerical results demonstrate the efficiency and robustness of the POD-NN method.

[1]  Karthik Duraisamy,et al.  Reduced-Order Modeling of Model Rocket Combustors , 2017 .

[2]  G. Sod.  A survey of several finite difference methods for systems of nonlinear hyperbolic conservation laws , 1978 .

[3]  J. Hesthaven,et al.  Certified Reduced Basis Methods for Parametrized Partial Differential Equations , 2015 .

[4]  William E. Anderson,et al.  Analysis of Self-Excited Combustion Instabilities Using Two- and Three-Dimensional Simulations , 2013 .

[5]  Martin A. Riedmiller,et al.  Advanced supervised learning in multi-layer perceptrons — From backpropagation to adaptive learning algorithms , 1994 .

[6]  C. P. Skeels,et al.  TVD SCHEMES FOR OPEN CHANNEL FLOW , 1998 .

[7]  Romain Garby.  Simulations of flame stabilization and stability in high-pressure propulsion systems , 2013 .

[8]  Charles L. Merkle,et al.  Determination of heat release response function from 2D hybrid RANS-LES data for the CVRC combustor , 2015 .

[9]  Yvon Maday,et al.  Reduced basis method for the rapid and reliable solution of partial differential equations , 2006 .

[10]  William E. Anderson,et al.  Experimental Study of High-Frequency Combustion Instability in a Continuously Variable Resonance Combustor (CVRC) , 2009 .

[11]  Kuolin Hsu,et al.  Artificial Neural Network Modeling of the Rainfall‐Runoff Process , 1995 .

[12]  Danny C. Sorensen,et al.  Nonlinear Model Reduction via Discrete Empirical Interpolation , 2010, SIAM J. Sci. Comput..

[13]  Gianluigi Rozza,et al.  Reduced Order Methods for Modeling and Computational Reduction , 2013 .

[14]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[15]  William E. Anderson,et al.  Spontaneous Longitudinal Combustion Instability in a Continuously-Variable Resonance Combustor , 2012 .

[16]  Chi-Wang Shu,et al.  Efficient Implementation of Weighted ENO Schemes , 1995 .

[17]  C. Eckart,et al.  The approximation of one matrix by another of lower rank , 1936 .

[18]  P. Sagaut,et al.  Calibrated reduced-order POD-Galerkin system for fluid flow modelling , 2005 .

[19]  Benjamin Stamm,et al.  EFFICIENT GREEDY ALGORITHMS FOR HIGH-DIMENSIONAL PARAMETER SPACES WITH APPLICATIONS TO EMPIRICAL INTERPOLATION AND REDUCED BASIS METHODS ∗ , 2014 .

[20]  Charles L. Merkle,et al.  Coupling between hydrodynamics, acoustics, and heat release in a self-excited unstable combustor , 2015 .

[21]  Karen Willcox,et al.  A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems , 2015, SIAM Rev..

[22]  Gianluigi Rozza,et al.  Supremizer stabilization of POD–Galerkin approximation of parametrized steady incompressible Navier–Stokes equations , 2015 .

[23]  P. Holmes,et al.  The Proper Orthogonal Decomposition in the Analysis of Turbulent Flows , 1993 .

[24]  Erich Novak,et al.  High dimensional polynomial interpolation on sparse grids , 2000, Adv. Comput. Math..

[25]  Gianluigi Rozza,et al.  Certified reduced basis approximation for parametrized partial differential equations and applications , 2011 .

[26]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[27]  E. Schmidt.  Zur Theorie der linearen und nichtlinearen Integralgleichungen , 1907 .

[28]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[29]  Qian Wang,et al.  Compact high order finite volume method on unstructured grids I: Basic formulations and one-dimensional schemes , 2016, J. Comput. Phys..

[30]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[31]  Laurent Selle,et al.  Prediction and Analysis of Combustion Instabilities in a Model Rocket Engine , 2014 .

[32]  Jean-Antoine Désidéri,et al.  Stability Properties of POD–Galerkin Approximations for the Compressible Navier–Stokes Equations , 2000 .

[33]  A. Patera,et al.  A PRIORI CONVERGENCE OF THE GREEDY ALGORITHM FOR THE PARAMETRIZED REDUCED BASIS METHOD , 2012 .

[34]  Charles L. Merkle,et al.  Computational Investigation of Acoustics and Instabilities in a Longitudinal-Mode Rocket Combustor , 2008 .

[35]  Charles L. Merkle,et al.  Quasi-1D modeling of heat release for the study of longitudinal combustion instability , 2018 .

[36]  S. Osher,et al.  Efficient implementation of essentially non-oscillatory shock-capturing schemes,II , 1989 .

[37]  Wolfgang Dahmen,et al.  Convergence Rates for Greedy Algorithms in Reduced Basis Methods , 2010, SIAM J. Math. Anal..

[39]  Venkateswaran Sankaran,et al.  Examination of mode shapes in an unstable model combustor , 2011 .

[40]  M. W Gardner,et al.  Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences , 1998 .

[41]  Benjamin Peherstorfer,et al.  Localized Discrete Empirical Interpolation Method , 2014, SIAM J. Sci. Comput..

[42]  B. Haasdonk,et al.  REDUCED BASIS METHOD FOR FINITE VOLUME APPROXIMATIONS OF PARAMETRIZED LINEAR EVOLUTION EQUATIONS , 2008 .

[43]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[44]  H. P. Lee,et al.  PROPER ORTHOGONAL DECOMPOSITION AND ITS APPLICATIONS—PART I: THEORY , 2002 .

[45]  Peter Benner,et al.  Reduced-Order Modeling and ROM-Based Optimization of Batch Chromatography , 2013, ENUMATH.

[46]  Francesco Nasuti,et al.  Numerical issues in modeling combustion instability by quasi-1D Euler equations , 2017 .

[47]  Charbel Farhat,et al.  Reduced-order fluid/structure modeling of a complete aircraft configuration , 2006 .

[48]  A. Chatterjee.  An introduction to the proper orthogonal decomposition , 2000 .

[49]  Fabien Casenave,et al.  A nonintrusive reduced basis method applied to aeroacoustic simulations , 2014, Adv. Comput. Math..

[50]  Yoshua Bengio,et al.  Practical Recommendations for Gradient-Based Training of Deep Architectures , 2012, Neural Networks: Tricks of the Trade.

[51]  Jan S. Hesthaven,et al.  Non-intrusive reduced order modeling of nonlinear problems using neural networks , 2018, J. Comput. Phys..

[52]  David Amsallem,et al.  Efficient model reduction of parametrized systems by matrix discrete empirical interpolation , 2015, J. Comput. Phys..

[53]  N. Nguyen,et al.  An ‘empirical interpolation’ method: application to efficient reduced-basis discretization of partial differential equations , 2004 .

[54]  William E. Anderson,et al.  Combustion Instability of Gaseous Fuels in a Continuously Variable Resonance Chamber (CVRC) , 2008 .

[55]  C. Merkle,et al.  Investigation of the Stability of POD-Galerkin Techniques for Reduced Order Model Development , 2016 .

[56]  Jan S. Hesthaven,et al.  Reduced order modeling for nonlinear structural analysis using Gaussian process regression , 2018, Computer Methods in Applied Mechanics and Engineering.

[57]  Matthew F. Barone,et al.  Stable Galerkin reduced order models for linearized compressible flow , 2009, J. Comput. Phys..