Learning physics-based reduced-order models for a single-injector combustion process

This paper presents a physics-based data-driven method to learn predictive reduced-order models (ROMs) from high-fidelity simulations, and illustrates it in the challenging context of a single-injector combustion process. The method combines the perspectives of model reduction and machine learning. Model reduction brings in the physics of the problem, constraining the ROM predictions to lie on a subspace defined by the governing equations. This is achieved by defining the ROM in proper orthogonal decomposition (POD) coordinates, which embed the rich physics information contained in solution snapshots of a high-fidelity computational fluid dynamics (CFD) model. The machine learning perspective brings the flexibility to use transformed physical variables to define the POD basis. This is in contrast to traditional model reduction approaches that are constrained to use the physical variables of the high-fidelity code. Combining the two perspectives, the approach identifies a set of transformed physical variables that expose quadratic structure in the combustion governing equations and learns a quadratic ROM from transformed snapshot data. This learning does not require access to the high-fidelity model implementation. Numerical experiments show that the ROM accurately predicts temperature, pressure, velocity, species concentrations, and the limit-cycle amplitude, with speedups of more than five orders of magnitude over high-fidelity models. ROM-predicted pressure traces accurately match the phase of the pressure signal and yield good approximations of the limit-cycle amplitude.

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

[2]  Charbel Farhat,et al.  Adaptation of Aeroelastic Reduced-Order Models and Application to an F-16 Configuration , 2007 .

[3]  Matthew F. Barone,et al.  Stable and Efficient Galerkin Reduced Order Models for Non-Linear Fluid Flow. , 2011 .

[4]  Karen E. Willcox,et al.  Balanced Truncation Model Reduction for Lifted Nonlinear Systems , 2019, ArXiv.

[5]  Karen Willcox,et al.  Model reduction for reacting flow applications , 2014 .

[6]  Stefan Volkwein,et al.  POD a-posteriori error estimates for linear-quadratic optimal control problems , 2009, Comput. Optim. Appl..

[7]  Charles L. Merkle,et al.  Analysis of Self-Excited Combustion Instability using a Combination of Two- and Three-Dimensional Simulations , 2012 .

[8]  Karthik Duraisamy,et al.  Exploration of reduced order models for rocket combustion applications , 2018 .

[9]  Matthew E. Harvazinski,et al.  Modeling self-excited combustion instabilities using a combination of two- and three-dimensional simulations , 2012 .

[10]  Petros Boufounos,et al.  Sparse Sensing and DMD-Based Identification of Flow Regimes and Bifurcations in Complex Flows , 2015, SIAM J. Appl. Dyn. Syst..

[11]  Karen Willcox,et al.  Transform & Learn: A data-driven approach to nonlinear model reduction , 2019, AIAA Aviation 2019 Forum.

[12]  Xavier Gloerfelt,et al.  Reliable reduced-order models for time-dependent linearized Euler equations , 2012, J. Comput. Phys..

[13]  Changjin Yoon,et al.  Combustion Dynamics in a Single-Element Lean Direct Injection Gas Turbine Combustor , 2019, Combustion Science and Technology.

[14]  Chenjie Gu,et al.  QLMOR: A Projection-Based Nonlinear Model Order Reduction Approach Using Quadratic-Linear Representation of Nonlinear Systems , 2011, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[15]  R. Murray,et al.  Model reduction for compressible flows using POD and Galerkin projection , 2004 .

[16]  L. Sirovich Turbulence and the dynamics of coherent structures. II. Symmetries and transformations , 1987 .

[17]  Anthony T. Patera,et al.  A posteriori error estimation for reduced-basis approximation of parametrized elliptic coercive partial differential equations : “convex inverse” bound conditioners , 2002 .

[18]  B. R. Noack,et al.  Low Order Galerkin Models for the Actuated Flow Around 2-D Airfoils , 2007 .

[19]  Karen Willcox,et al.  Projection-based model reduction for reacting ows , 2010 .

[20]  John T. Hwang,et al.  Review and Unification of Methods for Computing Derivatives of Multidisciplinary Computational Models , 2013 .

[21]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

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

[23]  Rick Chartrand,et al.  Numerical differentiation of noisy, nonsmooth, multidimensional data , 2017, 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[24]  I. Mezić,et al.  Spectral analysis of nonlinear flows , 2009, Journal of Fluid Mechanics.

[25]  Karen Willcox,et al.  Nonlinear Model Order Reduction via Lifting Transformations and Proper Orthogonal Decomposition , 2018, AIAA Journal.

[26]  Mark N. Glauser,et al.  Reduced Order Models for a High Speed Jet with Time-Resolved PIV , 2013 .

[27]  Karthik Duraisamy,et al.  Challenges in Reduced Order Modeling of Reacting Flows , 2018, 2018 Joint Propulsion Conference.

[28]  Steven L. Brunton,et al.  Data-driven discovery of partial differential equations , 2016, Science Advances.

[29]  N. Nguyen,et al.  EFFICIENT REDUCED-BASIS TREATMENT OF NONAFFINE AND NONLINEAR PARTIAL DIFFERENTIAL EQUATIONS , 2007 .

[30]  P. Schmid,et al.  Dynamic mode decomposition of numerical and experimental data , 2008, Journal of Fluid Mechanics.

[31]  Harbir Antil,et al.  Galerkin v. least-squares Petrov-Galerkin projection in nonlinear model reduction , 2015, J. Comput. Phys..

[32]  ProblemsPer Christian HansenDepartment The L-curve and its use in the numerical treatment of inverse problems , 2000 .

[33]  I. Knowles,et al.  METHODS FOR NUMERICAL DIFFERENTIATION OF NOISY DATA , 2014 .

[34]  Earl H. Dowell,et al.  Minimal subspace rotation on the Stiefel manifold for stabilization and enhancement of projection-based reduced order models for the compressible Navier-Stokes equations , 2015, J. Comput. Phys..

[35]  Qian Wang,et al.  Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem , 2019, J. Comput. Phys..

[36]  David R. Williams,et al.  Reduced-order unsteady aerodynamic models at low Reynolds numbers , 2013, Journal of Fluid Mechanics.

[37]  Hayden Schaeffer,et al.  Extracting Sparse High-Dimensional Dynamics from Limited Data , 2017, SIAM J. Appl. Math..

[38]  Karthik Duraisamy,et al.  Investigations and Improvement of Robustness of Reduced-Order Models of Reacting Flow , 2019, AIAA Journal.

[39]  C. Chui,et al.  Article in Press Applied and Computational Harmonic Analysis a Randomized Algorithm for the Decomposition of Matrices , 2022 .

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

[41]  S. Brunton,et al.  Discovering governing equations from data by sparse identification of nonlinear dynamical systems , 2015, Proceedings of the National Academy of Sciences.

[42]  K. Willcox,et al.  Data-driven operator inference for nonintrusive projection-based model reduction , 2016 .

[43]  Renee C. Swischuk Physics-based machine learning and data-driven reduced-order modeling , 2019 .

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

[45]  Charbel Farhat,et al.  Toward Real-Time Computational-Fluid-Dynamics-Based Aeroelastic Computations Using a Database of Reduced-Order Information , 2010 .

[46]  C. Westbrook,et al.  Simplified Reaction Mechanisms for the Oxidation of Hydrocarbon Fuels in Flames , 1981 .

[47]  Karen Willcox,et al.  Model Order Reduction for Reacting Flows: Laminar Gaussian Flame Applications , 2017 .

[48]  Benjamin Peherstorfer,et al.  Projection-based model reduction: Formulations for physics-based machine learning , 2019, Computers & Fluids.

[49]  K. Willcox,et al.  Aerodynamic Data Reconstruction and Inverse Design Using Proper Orthogonal Decomposition , 2004 .

[50]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .