Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Deep learning of nonlinear flame fronts development due to Darrieus–Landau instability

Yu, Rixin LU (2023) In APL Machine Learning 1.
Abstract
The Darrieus–Landau instability is studied using a data-driven, deep neural network approach. The task is set up to learn a time-advancement operator mapping any given flame front to a future time. A recurrent application of such an operator rolls out a long sequence of predicted flame fronts, and a learned operator is required to not only make accurate short-term predictions but also reproduce characteristic nonlinear behavior, such as fractal front structures and detached flame pockets. Using two datasets of flame front solutions obtained from a heavy-duty direct numerical simulation and a light-duty modeling equation, we compare the performance of three state-of-art operator-regression network methods: convolutional neural networks,... (More)
The Darrieus–Landau instability is studied using a data-driven, deep neural network approach. The task is set up to learn a time-advancement operator mapping any given flame front to a future time. A recurrent application of such an operator rolls out a long sequence of predicted flame fronts, and a learned operator is required to not only make accurate short-term predictions but also reproduce characteristic nonlinear behavior, such as fractal front structures and detached flame pockets. Using two datasets of flame front solutions obtained from a heavy-duty direct numerical simulation and a light-duty modeling equation, we compare the performance of three state-of-art operator-regression network methods: convolutional neural networks, Fourier neural operator (FNO), and deep operator network. We show that, for learning complicated front evolution, FNO gives the best recurrent predictions in both the short and long term. A consistent extension allowing the operator-regression networks to handle complicated flame front shape is achieved by representing the latter as an implicit curve. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
APL Machine Learning
volume
1
article number
026106
pages
19 pages
publisher
American Institute of Physics (AIP)
ISSN
2770-9019
DOI
10.1063/5.0139857
language
English
LU publication?
yes
id
27aa091b-4780-42cd-946b-619064058dc8
date added to LUP
2024-04-04 15:04:17
date last changed
2024-04-10 10:56:44
@article{27aa091b-4780-42cd-946b-619064058dc8,
  abstract     = {{The Darrieus–Landau instability is studied using a data-driven, deep neural network approach. The task is set up to learn a time-advancement operator mapping any given flame front to a future time. A recurrent application of such an operator rolls out a long sequence of predicted flame fronts, and a learned operator is required to not only make accurate short-term predictions but also reproduce characteristic nonlinear behavior, such as fractal front structures and detached flame pockets. Using two datasets of flame front solutions obtained from a heavy-duty direct numerical simulation and a light-duty modeling equation, we compare the performance of three state-of-art operator-regression network methods: convolutional neural networks, Fourier neural operator (FNO), and deep operator network. We show that, for learning complicated front evolution, FNO gives the best recurrent predictions in both the short and long term. A consistent extension allowing the operator-regression networks to handle complicated flame front shape is achieved by representing the latter as an implicit curve.}},
  author       = {{Yu, Rixin}},
  issn         = {{2770-9019}},
  language     = {{eng}},
  month        = {{04}},
  publisher    = {{American Institute of Physics (AIP)}},
  series       = {{APL Machine Learning}},
  title        = {{Deep learning of nonlinear flame fronts development due to Darrieus–Landau instability}},
  url          = {{http://dx.doi.org/10.1063/5.0139857}},
  doi          = {{10.1063/5.0139857}},
  volume       = {{1}},
  year         = {{2023}},
}