Accelerated evolution of Burgers' turbulence with coarse projective integration and deep learning
- 1 March 2024
- journal article
- research article
- Published by AIP Publishing in Physics of Fluids
- Vol. 36 (3)
Abstract
The evolution of a turbulent flow to a statistically steady state can be cast as a multiscale problem involving energy redistribution processes that take place on the long, large eddy turnover timescale and chaotic processes that take place on the much shorter timescale of the turbulence fluctuations. But the absence of a way to perform super-resolution reconstructions of the instantaneous velocity field from its lower-dimensional moments has prevented the use of standard multiscale computational approaches for accelerating turbulence simulations. We introduce an encoder-decoder recurrent neural network model, an architecture typically used in natural language processing, for translating between the instantaneous velocity field and energy spectrum in a one-dimensional turbulent flow. We use the model in a multiscale simulation scheme to evolve the stochastic Burgers' equation and recover the final, statistically stationary turbulent Burgers' velocity field up to 443 times faster in wall-clock time than using direct numerical simulation alone.Keywords
All Related Versions
This publication has 34 references indexed in Scilit:
- Direct Numerical Simulation (DNS)2021
- Linking Machine Learning with Multiscale Numerics: Data-Driven Discovery of Homogenized EquationsJOM, 2020
- Deep multiscale model learningJournal of Computational Physics, 2020
- A finite-element coarse-grid projection method for incompressible flow simulationsAdvances in Computational Mathematics, 2017
- A coarse-grid projection method for accelerating incompressible flow computationsJournal of Computational Physics, 2013
- Multiscale modeling of incompressible turbulent flowsJournal of Computational Physics, 2012
- Equation-free, coarse-grained computational optimization using timesteppersChemical Engineering Science, 2006
- Analysis of multiscale methods for stochastic differential equationsCommunications on Pure and Applied Mathematics, 2005
- Multiscale Methods in Science and Engineering2005
- A computational strategy for multiscale systems with applications to Lorenz 96 modelJournal of Computational Physics, 2004