Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A fast and accurate domain decomposition nonlinear manifold reduced order model

Journal Article · · Computer Methods in Applied Mechanics and Engineering

Here, this paper integrates nonlinear-manifold reduced order models (NM-ROMs) with domain decomposition (DD). NM ROMs approximate the full order model (FOM) state in a nonlinear-manifold by training a shallow, sparse autoencoder using FOM snapshot data. These NM-ROMs can be advantageous over linear-subspace ROMs (LS-ROMs) for problems with slowly decaying Kolmogorov n-width. However, the number of NM-ROM parameters that need to be trained scales with the size of the FOM. Moreover, for “extreme-scale” problems, the storage of high-dimensional FOM snapshots alone can make ROM training expensive. To alleviate the training cost, this paper applies DD to the FOM, computes NM-ROMs on each subdomain, and couples them to obtain a global NM-ROM. This approach has several advantages: Subdomain NM-ROMs can be trained in parallel, involve fewer parameters to be trained than global NM-ROMs, require smaller subdomain FOM dimensional training data, and can be tailored to subdomain specific features of the FOM. The shallow, sparse architecture of the autoencoder used in each subdomain NM-ROM allows application of hyper-reduction (HR), reducing the complexity caused by nonlinearity and yielding computational speedup of the NM-ROM. This paper provides the first application of NM-ROM (with HR) to a DD problem. In particular, this paper details an algebraic DD reformulation of the FOM, training a NM-ROM with HR for each sub domain, and a sequential quadratic programming (SQP) solver to evaluate the coupled global NM-ROM. Theoretical convergence results for the SQP method and a priori and a posteriori error estimates for the DD NM-ROM with HR are provided. The proposed DD NM-ROM with HR approach is numerically compared to a DD LS-ROM with HR on the 2D steady-state Burgers’ equation, showing an order of magnitude improvement in accuracy of the proposed DD NM-ROM over the DD LS-ROM.

Research Organization:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA); USDOE Laboratory Directed Research and Development (LDRD) Program; US Air Force Office of Scientific Research (AFOSR)
Grant/Contract Number:
AC52-07NA27344; SC0023164
OSTI ID:
2371783
Alternate ID(s):
OSTI ID: 2328659
Report Number(s):
LLNL--JRNL-849457; 1075130
Journal Information:
Computer Methods in Applied Mechanics and Engineering, Journal Name: Computer Methods in Applied Mechanics and Engineering Vol. 425; ISSN 0045-7825
Publisher:
ElsevierCopyright Statement
Country of Publication:
United States
Language:
English

References (46)

Numerical Methods for Parameter Estimation in Nonlinear Differential Algebraic Equations journal August 2007
Data-driven model order reduction of quadratic-bilinear systems: The Loewner framework for QB systems journal July 2018
Nonlinear model order reduction based on local reduced-order bases: NONLINEAR MODEL REDUCTION BASED ON LOCAL REDUCED-ORDER BASES
  • Amsallem, David; Zahr, Matthew J.; Farhat, Charbel
  • International Journal for Numerical Methods in Engineering, Vol. 92, Issue 10 https://doi.org/10.1002/nme.4371
journal June 2012
Port reduction in parametrized component static condensation: approximation and a posteriori error estimation: PORT REDUCTION IN PARAMETRIZED COMPONENT STATIC CONDENSATION journal July 2013
Transported snapshot model order reduction approach for parametric, steady-state fluid flows containing parameter-dependent shocks: Model order reduction for fluid flows containing shocks journal December 2018
Approximation by superpositions of a sigmoidal function journal December 1989
Rigorous and effective a-posteriori error bounds for nonlinear problems—application to RB methods journal March 2020
Non-linear Manifold Reduced-Order Models with Convolutional Autoencoders and Reduced Over-Collocation Method journal February 2023
A deep domain decomposition method based on Fourier features journal May 2023
A reduced basis hybrid method for the coupling of parametrized domains represented by fluidic networks journal May 2012
Component-wise reduced order model lattice-type structure design journal August 2021
Domain-decomposition least-squares Petrov–Galerkin (DD-LSPG) nonlinear model reduction journal October 2021
Reduced order models for Lagrangian hydrodynamics journal January 2022
Stress-constrained topology optimization of lattice-like structures using component-wise reduced order models journal October 2022
A one-shot overlapping Schwarz method for component-based model reduction: application to nonlinear elasticity journal February 2023
Explicit synchronous partitioned scheme for coupled reduced order models based on composite reduced bases journal December 2023
Iterative methods for model reduction by domain decomposition journal June 2009
The GNAT method for nonlinear model reduction: Effective implementation and application to computational fluid dynamics and turbulent flows journal June 2013
Conservative model reduction for finite-volume models journal October 2018
Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders journal November 2019
Space–time reduced order model for large-scale linear dynamical systems with application to Boltzmann transport problems journal January 2021
A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder journal February 2022
Quadratic approximation manifold for mitigating the Kolmogorov barrier in nonlinear projection-based model order reduction journal September 2022
Local Lagrangian reduced-order modeling for the Rayleigh-Taylor instability by solution manifold decomposition journal January 2023
A framework for the solution of the generalized realization problem journal September 2007
Numerical solution of saddle point problems journal April 2005
A Reduced-Basis Element Method journal January 2002
A Static condensation Reduced Basis Element method : approximation and a posteriori error estimation journal November 2012
Reduced basis techniques for nonlinear conservation laws journal April 2015
A discontinuous Galerkin reduced basis element method for elliptic problems journal February 2016
Advection modes by optimal mass transfer journal February 2014
D3M: A Deep Domain Decomposition Method for Partial Differential Equations journal January 2020
QLMOR: A Projection-Based Nonlinear Model Order Reduction Approach Using Quadratic-Linear Representation of Nonlinear Systems journal September 2011
Mesh Independence for Nonlinear Least Squares Problems with Norm Constraints journal February 1993
Two-Sided Projection Methods for Nonlinear Model Order Reduction journal January 2015
Online Adaptive Model Reduction for Nonlinear Systems via Low-Rank Updates journal January 2015
Model Reduction of Bilinear Systems in the Loewner Framework journal January 2016
Interpolation of Functions with Parameter Dependent Jumps by Transformed Snapshots journal January 2017
Space--Time Least-Squares Petrov--Galerkin Projection for Nonlinear Model Reduction journal January 2019
The Shifted Proper Orthogonal Decomposition: A Mode Decomposition for Multiple Transport Phenomena journal January 2018
SNS: A Solution-Based Nonlinear Subspace Method for Time-Dependent Model Order Reduction journal January 2020
Localized Model Reduction for Nonlinear Elliptic Partial Differential Equations: Localized Training, Partition of Unity, and Adaptive Enrichment journal June 2023
The Reduced Basis Element Method: Application to a Thermal Fin Problem journal January 2004
Karhunen–Loève procedure for gappy data journal January 1995
Adaptive Port Reduction in Static Condensation journal January 2012
Efficient Space–Time Reduced Order Model for Linear Dynamical Systems in Python Using Less than 120 Lines of Code journal July 2021

Similar Records

Domain-decomposition nonlinear manifold reduced order model
Software · Fri May 03 20:00:00 EDT 2024 · OSTI ID:code-141711

A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder
Journal Article · Thu Nov 11 23:00:00 EST 2021 · Journal of Computational Physics · OSTI ID:1843130