Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism

DOI10.1016/j.jcp.2020.109402zbMath1436.65093OpenAlexW2981433027MaRDI QIDQ777576

Nicolò Ripamonti, Qian Wang, Jan S. Hesthaven

Publication date: 7 July 2020

Published in: Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL: https://infoscience.epfl.ch/record/269133/files/MZ-LSTM%20Ver%201.pdf

zbMATH Keywords

model reductionimplicit-explicit Runge-Kuttaconditioned long-short term memorymemory closurePOD-Galerkin

Mathematics Subject Classification ID

Multistep, Runge-Kutta and extrapolation methods for ordinary differential equations (65L06) Finite element, Rayleigh-Ritz, Galerkin and collocation methods for ordinary differential equations (65L60) Numerical methods for low-rank matrix approximation; matrix compression (65F55)

Related Items (29)

A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables ⋮ Reduced basis methods for time-dependent problems ⋮ Fourier Collocation and Reduced Basis Methods for Fast Modeling of Compressible Flows ⋮ A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs ⋮ Physics-informed machine learning for reduced-order modeling of nonlinear problems ⋮ Model order reduction for compressible flows solved using the discontinuous Galerkin methods ⋮ Neural-network learning of SPOD latent dynamics ⋮ DeepParticle: learning invariant measure by a deep neural network minimizing Wasserstein distance on data generated from an interacting particle method ⋮ Neural-network-augmented projection-based model order reduction for mitigating the Kolmogorov barrier to reducibility ⋮ Simulation of the interaction of light with <scp>3‐D</scp> metallic nanostructures using a proper orthogonal decomposition‐Galerkin reduced‐order discontinuous Galerkin time‐domain method ⋮ A framework for machine learning of model error in dynamical systems ⋮ Variational multiscale super‐resolution: A data‐driven approach for reconstruction and predictive modeling of unresolved physics ⋮ CD-ROM: complemented deep -- reduced order model ⋮ DNN modeling of partial differential equations with incomplete data ⋮ Artificial neural network based correction for reduced order models in computational fluid mechanics ⋮ Extension of dynamic mode decomposition for dynamic systems with incomplete information based on t-model of optimal prediction ⋮ Physics Guided Machine Learning for Variational Multiscale Reduced Order Modeling ⋮ Neural Galerkin schemes with active learning for high-dimensional evolution equations ⋮ Data-driven reduced-order modelling for blood flow simulations with geometry-informed snapshots ⋮ Reduced order modeling for parametrized generalized Newtonian fluid flows ⋮ A nudged hybrid analysis and modeling approach for realtime wake-vortex transport and decay prediction ⋮ A multi-stage deep learning based algorithm for multiscale model reduction ⋮ Operator inference of non-Markovian terms for learning reduced models from partially observed state trajectories ⋮ Data-driven reduced order modeling of poroelasticity of heterogeneous media based on a discontinuous Galerkin approximation ⋮ POD-DL-ROM: enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition ⋮ Modeling synchronization in globally coupled oscillatory systems using model order reduction ⋮ A non-intrusive model order reduction approach for parameterized time-domain Maxwell's equations ⋮ Robust modeling of unknown dynamical systems via ensemble averaged learning ⋮ Latent-space time evolution of non-intrusive reduced-order models using Gaussian process emulation

Uses Software

• redbKIT
• TensorFlow
• Keras
• Adam

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions
• Artificial compressibility method revisited: asymptotic numerical method for incompressible Navier-Stokes equations
• Problem reduction, renormalization, and memory
• Numerical models for differential problems. Translated by Silvia Quarteroni.
• On the stability of the reduced basis method for Stokes equations in parametrized domains
• Stability properties of POD-Galerkin approximations for the compressible Navier-Stokes equations
• Optimal prediction with memory
• Data-based stochastic model reduction for the Kuramoto-Sivashinsky equation
• A dynamic subgrid scale model for large eddy simulations based on the Mori-Zwanzig formalism
• An `empirical interpolation' method: Application to efficient reduced-basis discretization of partial differential equations
• Existence proof for orthogonal dynamics and the Mori-Zwanzig formalism
• Model reduction for compressible flows using POD and Galerkin projection
• Mori-Zwanzig reduced models for uncertainty quantification
• Model reduction of parametrized systems. Selected contributions based on the presentations at the MoRePaS conference, SISSA, Trieste, Italy, October 13--16, 2015
• A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems
• Efficient non-linear model reduction via a least-squares Petrov-Galerkin projection and compressive tensor approximations
• Online Adaptive Model Reduction for Nonlinear Systems via Low-Rank Updates
• Supremizer stabilization of POD-Galerkin approximation of parametrized steady incompressible Navier-Stokes equations
• Adaptiveh-refinement for reduced-order models
• Nonlinear Model Reduction via Discrete Empirical Interpolation
• Stochastic Tools in Mathematics and Science
• Optimal prediction and the Mori–Zwanzig representation of irreversible processes
• On the estimation of the Mori-Zwanzig memory integral
• A priori estimation of memory effects in reduced-order models of nonlinear systems using the Mori–Zwanzig formalism
• Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks
• Data-Driven Discovery of Closure Models
• Model Reduction of MIMO Systems via Tangential Interpolation
• Missing Point Estimation in Models Described by Proper Orthogonal Decomposition
• Model Reduction with Memory and the Machine Learning of Dynamical Systems
• Transport, Collective Motion, and Brownian Motion
• Memory Effects in Irreversible Thermodynamics
• Dynamical Model Reduction Method for Solving Parameter-Dependent Dynamical Systems
• Renormalized Mori–Zwanzig-reduced models for systems without scale separation

This page was built for publication: Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism