Recovering missing CFD data for high-order discretizations using deep neural networks and dynamics learning
From MaRDI portal
Publication:2222332
Abstract: Data I/O poses a significant bottleneck in large-scale CFD simulations; thus, practitioners would like to significantly reduce the number of times the solution is saved to disk, yet retain the ability to recover any field quantity (at any time instance) a posteriori. The objective of this work is therefore to accurately recover missing CFD data a posteriori at any time instance, given that the solution has been written to disk at only a relatively small number of time instances. We consider in particular high-order discretizations (e.g., discontinuous Galerkin), as such techniques are becoming increasingly popular for the simulation of highly separated flows. To satisfy this objective, this work proposes a methodology consisting of two stages: 1) dimensionality reduction and 2) dynamics learning. For dimensionality reduction, we propose a novel hierarchical approach. First, the method reduces the number of degrees of freedom within each element of the high-order discretization by applying autoencoders from deep learning. Second, the methodology applies principal component analysis to compress the global vector of encodings. This leads to a low-dimensional state, which associates with a nonlinear embedding of the original CFD data. For dynamics learning, we propose to apply regression techniques (e.g., kernel methods) to learn the discrete-time velocity characterizing the time evolution of this low-dimensional state. A numerical example on a large-scale CFD example characterized by nearly 13 million degrees of freedom illustrates the suitability of the proposed method in an industrial setting.
Recommendations
- Construction of reduced-order models for fluid flows using deep feedforward neural networks
- Combining CFD-EFD-FFD data via gappy proper orthogonal decomposition
- Sparse identification of nonlinear dynamics with low-dimensionalized flow representations
- Deep learning of thermodynamics-aware reduced-order models from data
- Key time steps selection for CFD data based on deep metric learning
Cites work
- A data-driven approximation of the koopman operator: extending dynamic mode decomposition
- A high-order multidomain spectral difference method for the Navier-Stokes equations on unstructured hexahedral grids
- A staggered-grid multidomain spectral method for the compressible Navier-Stokes equations
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Dynamic mode decomposition of numerical and experimental data
- Experimental and numerical studies of the flow over a circular cylinder at Reynolds number 3900
- Facilitating the adoption of unstructured high-order methods amongst a wider community of fluid dynamicists
- Galerkin v. least-squares Petrov-Galerkin projection in nonlinear model reduction
- Gappy data and reconstruction procedures for flow past a cylinder
- Gappy data: to krig or not to krig?
- Heterogeneous computing on mixed unstructured grids with pyfr
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- scientific article; zbMATH DE number 5179816 (Why is no real title available?)
- Linearly recurrent autoencoder networks for learning dynamics
- Low-storage, explicit Runge-Kutta schemes for the compressible Navier-Stokes equations
- Optimization methods for large-scale machine learning
- Random forests
- Reducing the Dimensionality of Data with Neural Networks
- Regularization and Variable Selection Via the Elastic Net
- Spectral difference method for unstructured grids. I. Basic formulation
- Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion
- Surrogate modeling of multiscale models using kernel methods
- The Runge-Kutta local projection P¹-discontinuous-Galerkin finite element method for scalar conservation laws
- Turbulence, Coherent Structures, Dynamical Systems and Symmetry
- Unsteady flow sensing and estimation via the gappy proper orthogonal decomposition
Cited in
(16)- Multi-level convolutional autoencoder networks for parametric prediction of spatio-temporal dynamics
- Data-driven predictions of the Lorenz system
- Non-intrusive reduced-order modeling using uncertainty-aware deep neural networks and proper orthogonal decomposition: application to flood modeling
- Machine learning for fluid flow reconstruction from limited measurements
- Scalable \textit{in situ} compression of transient simulation data using time-dependent bases
- Key time steps selection for CFD data based on deep metric learning
- A tailored convolutional neural network for nonlinear manifold learning of computational physics data using unstructured spatial discretizations
- Sparse identification of nonlinear dynamics with low-dimensionalized flow representations
- Shallow neural networks for fluid flow reconstruction with limited sensors
- Nonintrusive Reduced Order Modelling of Convective Boussinesq Flows
- Launching Drifter Observations in the Presence of Uncertainty
- A causation-based computationally efficient strategy for deploying Lagrangian drifters to improve real-time state estimation
- On the optimality of target-data-dependent kernel greedy interpolation in Sobolev reproducing kernel Hilbert spaces
- A surrogate reduced order model of the unsteady advection dominant problems based on combination of deep autoencoders-LSTM and POD
- Data discovery of lower dimensional equations of turbulent flows
- Symplecticity-preserving prediction of Hamiltonian dynamics by generalized kernel interpolation
This page was built for publication: Recovering missing CFD data for high-order discretizations using deep neural networks and dynamics learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2222332)
