Nonlinear proper orthogonal decomposition for convection-dominated flows

  title={Nonlinear proper orthogonal decomposition for convection-dominated flows},
  author={Shady E. Ahmed and Omer San and Adil Rasheed and Traian Iliescu},
Autoencoder techniques find increasingly common use in reduced order modeling as a means to create a latent space. This reduced order representation offers a modular data-driven modeling approach for nonlinear dynamical systems when integrated with a time series predictive model. In this letter, we put forth a nonlinear proper orthogonal decomposition (POD) framework, which is an end-to-end Galerkin-free model combining autoencoders with long short-term memory networks for dynamics. By… 

Figures and Tables from this paper


A data driven reduced order model of fluid flow by Auto-Encoder and self-attention deep learning methods
This paper presents a new data-driven non-intrusive reduced-order model(NIROM) that outperforms the traditional Proper orthogonal decomposition (POD) based reducedorder model. This is achieved by
An autoencoder-based reduced-order model for eigenvalue problems with application to neutron diffusion
Using an autoencoder for dimensionality reduction, this paper presents a novel projection-based reduced-order model for eigenvalue problems, compared with the standard POD-Galerkin approach and applied to two test cases taken from the field of nuclear reactor physics.
Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low dimensionalization
We focus on a convolutional neural network (CNN), which has recently been utilized for fluid flow analyses, from the perspective on the influence of various operations inside it by considering some
Kvamsdal. Model fusion with physics-guided machine learning: Projection-based reduced-order modeling
  • Physics of Fluids,
  • 2021
Learning a reduced basis of dynamical systems using an autoencoder.
An autoencoder with latent space penalization is used to discover approximate finite-dimensional manifolds of two canonical partial differential equations and shows that the resulting optimal latent space of the K-S equation is consistent with the dimension of the inertial manifold.
Low-Rank Registration Based Manifolds for Convection-Dominated PDEs
We develop an auto-encoder-type nonlinear dimensionality reduction algorithm to enable the construction of reduced order models of systems governed by convection-dominated nonlinear partial
Model fusion with physics-guided machine learning: Projection-based reduced-order modeling
The unprecedented amount of data generated from experiments, field observations, and large-scale numerical simulations at a wide range of spatiotemporal scales has enabled the rapid advancement of
Multifidelity computing for coupling full and reduced order models.
The proposed interface learning approaches are tested as a new way to address ROM-FOM coupling problems solving nonlinear advection-diffusion flow situations with a bifidelity setup that captures the essence of a broad class of transport processes.
Non-intrusive Nonlinear Model Reduction via Machine Learning Approximations to Low-dimensional Operators
This work proposes a method that enables traditionally intrusive reduced-order models to be accurately approximated in a non-intrusive manner using modern machine-learning regression techniques, and demonstrates the effectiveness of the proposed technique on two types of PDEs.
On closures for reduced order models—A spectrum of first-principle to machine-learned avenues
machine-learned avenues Shady E. Ahmed, Suraj Pawar, Omer San, a) Adil Rasheed, Traian Iliescu, and Bernd R. Noack 5 School of Mechanical & Aerospace Engineering, Oklahoma State University,