Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks
A data-driven model for rapid prediction of the steady-state heat conduction of a hot object with arbitrary geometry is developed. Mathematically, the steady-state heat conduction can be described by the Laplace's equation, where a heat (spatial) diffusion term dominates the governing equation....
Guardado en:
Autores principales: | , , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Elsevier
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/f28e11a437dd48879d0f642583b7cc06 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:f28e11a437dd48879d0f642583b7cc06 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:f28e11a437dd48879d0f642583b7cc062021-11-18T04:48:57ZData-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks2214-157X10.1016/j.csite.2021.101651https://doaj.org/article/f28e11a437dd48879d0f642583b7cc062021-12-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2214157X21008145https://doaj.org/toc/2214-157XA data-driven model for rapid prediction of the steady-state heat conduction of a hot object with arbitrary geometry is developed. Mathematically, the steady-state heat conduction can be described by the Laplace's equation, where a heat (spatial) diffusion term dominates the governing equation. As the intensity of the heat diffusion only depends on the gradient of the temperature field, the temperature distribution of the steady-state heat conduction displays strong features of locality. Therefore, a convolution neural network-based data-driven model is proposed, which is good at capturing local features (sub-invariant) thus can be treated as numerical discretization in some sense. Furthermore, a signed distance function (SDF) is proposed to represent the geometry of the problem, which contains more information than the binary image. The hot objects in the training datasets are composed of simple geometries, the geometry is different in size, shape, orientation, and location. After training, the data-driven model can accurately predict steady-state heat conduction of hot objects with complex geometries which have never been seen by the network; and the prediction speed is more than one order faster than numerical simulation. The outstanding performance of the network model indicates the potential of the approach for applications of engineering optimization and design in future.Jiang-Zhou PengXianglei LiuNadine AubryZhihua ChenWei-Tao WuElsevierarticleHeat transferHeat conductionData-driven modelConvolution neural networksSigned distance functionEngineering (General). Civil engineering (General)TA1-2040ENCase Studies in Thermal Engineering, Vol 28, Iss , Pp 101651- (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
Heat transfer Heat conduction Data-driven model Convolution neural networks Signed distance function Engineering (General). Civil engineering (General) TA1-2040 |
spellingShingle |
Heat transfer Heat conduction Data-driven model Convolution neural networks Signed distance function Engineering (General). Civil engineering (General) TA1-2040 Jiang-Zhou Peng Xianglei Liu Nadine Aubry Zhihua Chen Wei-Tao Wu Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
description |
A data-driven model for rapid prediction of the steady-state heat conduction of a hot object with arbitrary geometry is developed. Mathematically, the steady-state heat conduction can be described by the Laplace's equation, where a heat (spatial) diffusion term dominates the governing equation. As the intensity of the heat diffusion only depends on the gradient of the temperature field, the temperature distribution of the steady-state heat conduction displays strong features of locality. Therefore, a convolution neural network-based data-driven model is proposed, which is good at capturing local features (sub-invariant) thus can be treated as numerical discretization in some sense. Furthermore, a signed distance function (SDF) is proposed to represent the geometry of the problem, which contains more information than the binary image. The hot objects in the training datasets are composed of simple geometries, the geometry is different in size, shape, orientation, and location. After training, the data-driven model can accurately predict steady-state heat conduction of hot objects with complex geometries which have never been seen by the network; and the prediction speed is more than one order faster than numerical simulation. The outstanding performance of the network model indicates the potential of the approach for applications of engineering optimization and design in future. |
format |
article |
author |
Jiang-Zhou Peng Xianglei Liu Nadine Aubry Zhihua Chen Wei-Tao Wu |
author_facet |
Jiang-Zhou Peng Xianglei Liu Nadine Aubry Zhihua Chen Wei-Tao Wu |
author_sort |
Jiang-Zhou Peng |
title |
Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
title_short |
Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
title_full |
Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
title_fullStr |
Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
title_full_unstemmed |
Data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
title_sort |
data-driven modeling of geometry-adaptive steady heat conduction based on convolutional neural networks |
publisher |
Elsevier |
publishDate |
2021 |
url |
https://doaj.org/article/f28e11a437dd48879d0f642583b7cc06 |
work_keys_str_mv |
AT jiangzhoupeng datadrivenmodelingofgeometryadaptivesteadyheatconductionbasedonconvolutionalneuralnetworks AT xiangleiliu datadrivenmodelingofgeometryadaptivesteadyheatconductionbasedonconvolutionalneuralnetworks AT nadineaubry datadrivenmodelingofgeometryadaptivesteadyheatconductionbasedonconvolutionalneuralnetworks AT zhihuachen datadrivenmodelingofgeometryadaptivesteadyheatconductionbasedonconvolutionalneuralnetworks AT weitaowu datadrivenmodelingofgeometryadaptivesteadyheatconductionbasedonconvolutionalneuralnetworks |
_version_ |
1718425075168837632 |