Recovering missing CFD data for high-order discretizations using deep neural networks and dynamics learning

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.

arXiv commentAccepted in Journal of Computational Physics

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Auto-Encoding Variational Bayes
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• A Data-Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition
• Deep learning for universal linear embeddings of nonlinear dynamics
• State Representation Learning for Control: An Overview
• Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images
• Deep Dynamical Modeling and Control of Unsteady Fluid Flows
• Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data
• A Kernel-Based Approach to Data-Driven Koopman Spectral Analysis
• Learning to Linearize Under Uncertainty

Cited by in corpus (16)

• Data-driven discovery of coordinates and governing equations
• Shallow Neural Networks for Fluid Flow Reconstruction with Limited Sensors
• Assessment of supervised machine learning methods for fluid flows
• Multi-level Convolutional Autoencoder Networks for Parametric Prediction of Spatio-temporal Dynamics
• Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low-dimensionalization
• Sparse identification of nonlinear dynamics with low-dimensionalized flow representations
• Assessment of unsteady flow predictions using hybrid deep learning based reduced order models
• Data-driven discovery of intrinsic dynamics
• Non-Intrusive Reduced-Order Modeling Using Uncertainty-Aware Deep Neural Networks and Proper Orthogonal Decomposition: Application to Flood Modeling
• Nonlinear proper orthogonal decomposition for convection-dominated flows
• Tensor-based flow reconstruction from optimally located sensor measurements
• A Tailored Convolutional Neural Network for Nonlinear Manifold Learning of Computational Physics Data using Unstructured Spatial Discretizations
• On the optimality of target-data-dependent kernel greedy interpolation in Sobolev Reproducing Kernel Hilbert Spaces
• Model order reduction with neural networks: Application to laminar and turbulent flows
• Geometry encoding for numerical simulations
• Nonintrusive reduced order modeling of convective Boussinesq flows