Daytime radiative cooling multilayer films designed by a machine learning method and genetic algorithm

Siyuan Li; Meng An; Zhiheng Zheng; Yuchun Gou; Wenlei Lian; Wei Yu; Ping Zhang

doi:10.1364/AO.486726

Applied Optics
Vol. 62,
Issue 16,
pp. 4359-4369
(2023)
•https://doi.org/10.1364/AO.486726

Daytime radiative cooling multilayer films designed by a machine learning method and genetic algorithm

Siyuan Li, Meng An, Zhiheng Zheng, Yuchun Gou, Wenlei Lian, Wei Yu, and Ping Zhang

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Siyuan Li, Meng An, Zhiheng Zheng, Yuchun Gou, Wenlei Lian, Wei Yu, and Ping Zhang, "Daytime radiative cooling multilayer films designed by a machine learning method and genetic algorithm," Appl. Opt. 62, 4359-4369 (2023)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Abstract

Recently, there has been growing interest and attention towards daytime radiative cooling. This cooling technology is considered a potentially significant alternative to traditional cooling methods because of its neither energy consumption nor harmful gas emission during operation. In this paper, a daytime radiative cooling emitter (DRCE) consisting of polydimethylsiloxane, silicon dioxide, and aluminum nitride from top to bottom on a silver-silicon substrate was designed by a machine learning method (MLM) and genetic algorithm to achieve daytime radiative cooling. The optimal DRCE had 94.43% average total hemispherical emissivity in the atmospheric window wavelength band and 98.25% average total hemispherical reflectivity in the solar radiation wavelength band. When the ambient temperature was 30°C, and the power of solar radiation was about ${900}\;{{\rm W/m}^2}$, the net cooling power of the optimal DRCE could achieve ${140.38}\;{{\rm W/m}^2}$. The steady-state temperature of that could be approximately 9.08°C lower than the ambient temperature. This paper provides a general research strategy for MLM-driven design of DRCE.

Full Article | PDF Article

More Like This

Antireflection and radiative cooling difunctional coating design for silicon solar cells

Yiteng Tu, Xinyu Tan, Xiongbo Yang, Guiguang Qi, Kun Yan, and Zhe Kang
Opt. Express 31(14) 22296-22307 (2023)

Metal-free design of a multilayered metamaterial with chirped Bragg grating for enhanced radiative cooling

D. Osuna Ruiz, C. Lezaun, A. E. Torres-García, and M. Beruete
Opt. Express 31(14) 22698-22709 (2023)

Daytime radiative cooler using porous TiO₂: new approach

Mounir Zahir and Mourad Benlattar
Appl. Opt. 59(30) 9400-9408 (2020)

Previous Article Next Article

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (12)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (5)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (10)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

	Quantity (Unit)	Feature 1 (µm)	Feature 2 (µm)	Feature 3 (µm)	Feature 4 (Integer)	Feature 5 (Integer)	Feature 6 (Integer)	Target $τ$
Group 1	24000	0–20	0–20	0–20	0,1	2–4	2–4	0 – $\sqrt{2}$
Group 2	19729	0–15	0–15	0–15	0,1	2–4	2–4	0– $\sqrt{2}$
Test set	1200	0–20	0–20	0–20	0,1	2–4	2–4	0 – $\sqrt{2}$

Method	RMSE	$R^{2}$
KNN	0.028	0.88
AdaBoost	0.043	0.53
XGBoost	0.013	0.96
LightGBM	0.010	0.98
GBDT	0.015	0.94

Method	RMSE	$R^{2}$
KNN	0.040	0.59
AdaBoost	0.048	0.40
XGBoost	0.015	0.94
LightGBM	0.013	0.96
GBDT	0.018	0.92

Method	Time Consuming
TMM	7783.37 s
MLM	115.81 s

No.	Time	Material	Reference	Cooling Power ( $W / m^{2}$ )	$T_{a m b} - T_{d e v}$
1	2017	$P D M S / {S i O}_{2} / A g$	[46]	$127 {W / m}^{2}$	8.2°C
2	2017	$T P X / {S i O}_{2}$	[47]	$93 {W / m}^{2}$	/
3	2018	$P (V d F - H F P)_{H P}$	[48]	$96 {W / m}^{2}$	6°C
4	2019	PDMS/Al	[49]	$120 {W / m}^{2}$	2°C–6°C
5	2020	${S i O}_{2} / {S i}_{3} N_{4} / {A l}_{2} O_{3} / A g - S i$	[21]	$73.8 {W / m}^{2}$	6.2°C–8.2°C
6	2020	$P D M S / {A l}_{2} O_{3}$	[50]	$90.8 {W / m}^{2}$	5.1°C–7°C
7	2021	$A l O N / {S i O}_{2} / A l$	[11]	$110 {W / m}^{2}$	7°C
8	This paper	$P D M S / {S i O}_{2} / A l N / A g / S i$	/	$140.38 {W / m}^{2}$	9.08°C

Abstract

Data availability

Cited By

Figures (12)

Tables (5)

Equations (10)

Applied Optics