Multi-fidelity Generative Deep Learning Turbulent Flows

Abstract

In computational fluid dynamics, there is an inevitable trade off between accuracy and computational cost. In this work, a novel multi-fidelity deep generative model is introduced for the surrogate modeling of high-fidelity turbulent flow fields given the solution of a computationally inexpensive but inaccurate low-fidelity solver. The resulting surrogate is able to generate physically accurate turbulent realizations at a computational cost magnitudes lower than that of a high-fidelity simulation. The deep generative model developed is a conditional invertible neural network, built with normalizing flows, with recurrent LSTM connections that allow for stable training of transient systems with high predictive accuracy. The model is trained with a variational loss that combines both data-driven and physics-constrained learning. This deep generative model is applied to non-trivial high Reynolds number flows governed by the Navier-Stokes equations including turbulent flow over a backwards facing step at different Reynolds numbers and turbulent wake behind an array of bluff bodies. For both of these examples, the model is able to generate unique yet physically accurate turbulent fluid flows conditioned on an inexpensive low-fidelity solution.

arXiv comment45 pages, 30 figures

Cited by in corpus (9)

• Uncertainty Quantification in Scientific Machine Learning: Methods, Metrics, and Comparisons
• Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels
• Transformers for Modeling Physical Systems
• Multi-fidelity surrogate modeling using long short-term memory networks
• Solving inverse problems using conditional invertible neural networks
• Use of Multifidelity Training Data and Transfer Learning for Efficient Construction of Subsurface Flow Surrogate Models
• Physics-enhanced deep surrogates for partial differential equations
• A multiscale and multicriteria Generative Adversarial Network to synthesize 1-dimensional turbulent fields
• Graph Laplacian-based Bayesian Multi-fidelity Modeling