Design of a Broadband Solar Thermal Absorber Using a Deep Neural Network and Experimental Demonstration of Its Performance

Junyong Seo, Pil Hoon Jung, Mingeon Kim, Sounghyeok Yang, Ikjin Lee, Jungchul Lee, Heon Lee, Bong Jae Lee

Research output: Contribution to journalArticle

Abstract

In using nanostructures to design solar thermal absorbers, computational methods, such as rigorous coupled-wave analysis and the finite-difference time-domain method, are often employed to simulate light-structure interactions in the solar spectrum. However, those methods require heavy computational resources and CPU time. In this study, using a state-of-the-art modeling technique, i.e., deep learning, we demonstrate significant reduction of computational costs during the optimization processes. To minimize the number of samples obtained by actual simulation, only regulated amounts are prepared and used as a data set to train the deep neural network (DNN) model. Convergence of the constructed DNN model is carefully examined. Moreover, several analyses utilizing an evolutionary algorithm, which require a remarkable number of performance calculations, are performed using the trained DNN model. We show that deep learning effectively reduces the actual simulation counts compared to the case of a design process without a neural network model. Finally, the proposed solar thermal absorber is fabricated and its absorption performance is characterized.

Original languageEnglish
Article number15028
JournalScientific reports
Volume9
Issue number1
DOIs
Publication statusPublished - 2019 Dec 1

Fingerprint

Neural Networks (Computer)
Hot Temperature
Learning
Nanostructures
Light
Costs and Cost Analysis

ASJC Scopus subject areas

  • General

Cite this

Design of a Broadband Solar Thermal Absorber Using a Deep Neural Network and Experimental Demonstration of Its Performance. / Seo, Junyong; Jung, Pil Hoon; Kim, Mingeon; Yang, Sounghyeok; Lee, Ikjin; Lee, Jungchul; Lee, Heon; Lee, Bong Jae.

In: Scientific reports, Vol. 9, No. 1, 15028, 01.12.2019.

Research output: Contribution to journalArticle

Seo, Junyong ; Jung, Pil Hoon ; Kim, Mingeon ; Yang, Sounghyeok ; Lee, Ikjin ; Lee, Jungchul ; Lee, Heon ; Lee, Bong Jae. / Design of a Broadband Solar Thermal Absorber Using a Deep Neural Network and Experimental Demonstration of Its Performance. In: Scientific reports. 2019 ; Vol. 9, No. 1.
@article{1c1f5d3a0e26415bb9689a7e9b01d16f,
title = "Design of a Broadband Solar Thermal Absorber Using a Deep Neural Network and Experimental Demonstration of Its Performance",
abstract = "In using nanostructures to design solar thermal absorbers, computational methods, such as rigorous coupled-wave analysis and the finite-difference time-domain method, are often employed to simulate light-structure interactions in the solar spectrum. However, those methods require heavy computational resources and CPU time. In this study, using a state-of-the-art modeling technique, i.e., deep learning, we demonstrate significant reduction of computational costs during the optimization processes. To minimize the number of samples obtained by actual simulation, only regulated amounts are prepared and used as a data set to train the deep neural network (DNN) model. Convergence of the constructed DNN model is carefully examined. Moreover, several analyses utilizing an evolutionary algorithm, which require a remarkable number of performance calculations, are performed using the trained DNN model. We show that deep learning effectively reduces the actual simulation counts compared to the case of a design process without a neural network model. Finally, the proposed solar thermal absorber is fabricated and its absorption performance is characterized.",
author = "Junyong Seo and Jung, {Pil Hoon} and Mingeon Kim and Sounghyeok Yang and Ikjin Lee and Jungchul Lee and Heon Lee and Lee, {Bong Jae}",
year = "2019",
month = "12",
day = "1",
doi = "10.1038/s41598-019-51407-2",
language = "English",
volume = "9",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Design of a Broadband Solar Thermal Absorber Using a Deep Neural Network and Experimental Demonstration of Its Performance

AU - Seo, Junyong

AU - Jung, Pil Hoon

AU - Kim, Mingeon

AU - Yang, Sounghyeok

AU - Lee, Ikjin

AU - Lee, Jungchul

AU - Lee, Heon

AU - Lee, Bong Jae

PY - 2019/12/1

Y1 - 2019/12/1

N2 - In using nanostructures to design solar thermal absorbers, computational methods, such as rigorous coupled-wave analysis and the finite-difference time-domain method, are often employed to simulate light-structure interactions in the solar spectrum. However, those methods require heavy computational resources and CPU time. In this study, using a state-of-the-art modeling technique, i.e., deep learning, we demonstrate significant reduction of computational costs during the optimization processes. To minimize the number of samples obtained by actual simulation, only regulated amounts are prepared and used as a data set to train the deep neural network (DNN) model. Convergence of the constructed DNN model is carefully examined. Moreover, several analyses utilizing an evolutionary algorithm, which require a remarkable number of performance calculations, are performed using the trained DNN model. We show that deep learning effectively reduces the actual simulation counts compared to the case of a design process without a neural network model. Finally, the proposed solar thermal absorber is fabricated and its absorption performance is characterized.

AB - In using nanostructures to design solar thermal absorbers, computational methods, such as rigorous coupled-wave analysis and the finite-difference time-domain method, are often employed to simulate light-structure interactions in the solar spectrum. However, those methods require heavy computational resources and CPU time. In this study, using a state-of-the-art modeling technique, i.e., deep learning, we demonstrate significant reduction of computational costs during the optimization processes. To minimize the number of samples obtained by actual simulation, only regulated amounts are prepared and used as a data set to train the deep neural network (DNN) model. Convergence of the constructed DNN model is carefully examined. Moreover, several analyses utilizing an evolutionary algorithm, which require a remarkable number of performance calculations, are performed using the trained DNN model. We show that deep learning effectively reduces the actual simulation counts compared to the case of a design process without a neural network model. Finally, the proposed solar thermal absorber is fabricated and its absorption performance is characterized.

UR - http://www.scopus.com/inward/record.url?scp=85073655670&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073655670&partnerID=8YFLogxK

U2 - 10.1038/s41598-019-51407-2

DO - 10.1038/s41598-019-51407-2

M3 - Article

C2 - 31636300

AN - SCOPUS:85073655670

VL - 9

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 15028

ER -