Stabilized neural ordinary differential equations for long-time forecasting of dynamical systems

DOI10.1016/j.jcp.2022.111838OpenAlexW4312054954MaRDI QIDQ2112549

Yanyan Li

Publication date: 11 January 2023

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

Full work available at URL: https://arxiv.org/abs/2203.15706

zbMATH Keywords

partial differential equationsreduced-order modelsneural ordinary differential equations

Mathematics Subject Classification ID

Numerical methods for partial differential equations, initial value and time-dependent initial-boundary value problems (65Mxx) Parabolic equations and parabolic systems (35Kxx) Qualitative properties of solutions to partial differential equations (35Bxx)

Related Items (5)

Dynamics of a data-driven low-dimensional model of turbulent minimal Couette flow ⋮ Physics-agnostic and physics-infused machine learning for thin films flows: modelling, and predictions from small data ⋮ A multifidelity deep operator network approach to closure for multiscale systems ⋮ Incremental neural controlled differential equations for modeling of path-dependent material behavior ⋮ Learning subgrid-scale models with neural ordinary differential equations

Uses Software

• PyTorch
• LANDO

Cites Work

• Unnamed Item
• Discovering governing equations from data by sparse identification of nonlinear dynamical systems
• Reservoir computing approaches to recurrent neural network training
• Global dissipativity and inertial manifolds for diffusive Burgers equations with low-wavenumber instability
• A data-driven approximation of the koopman operator: extending dynamic mode decomposition
• Inertial manifolds for the Kuramoto-Sivashinsky equation and an estimate of their lowest dimension
• Spectral barriers and inertial manifolds for dissipative partial differential equations
• Nonlinear Galerkin methods: The finite elements case
• The nonlinear Galerkin method in computational fluid dynamics
• The energy method, stability, and nonlinear convection
• Estimates on the lowest dimension of inertial manifolds for the Kuramoto-Sivashinsky equation in the general case
• Time-series learning of latent-space dynamics for reduced-order model closure
• Weak SINDy for partial differential equations
• Stationary two-dimensional turbulence statistics using a Markovian forcing scheme
• Dynamic Mode Decomposition with Control
• Cartography of high-dimensional flows: A visual guide to sections and slices
• Applied Koopmanism
• Postprocessing the Galerkin Method: a Novel Approach to Approximate Inertial Manifolds
• Turbulent Flows
• Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks
• Weak SINDy: Galerkin-Based Data-Driven Model Selection
• Inertial manifolds and finite-dimensional reduction for dissipative PDEs
• Fourth-Order Time-Stepping for Stiff PDEs
• Evaluating the dimension of an inertial manifold for the Kuramoto-Sivashinsky equation.

This page was built for publication: Stabilized neural ordinary differential equations for long-time forecasting of dynamical systems