A deep learning perspective into the figure-of-merit of thermoelectric materials

Russlan Jaafreh; Kang Yoo Seong; Jung Gu Kim; Kotiba Hamad

doi:10.1016/j.matlet.2022.132299

A deep learning perspective into the figure-of-merit of thermoelectric materials

Russlan Jaafreh
, Kang Yoo Seong
, Jung Gu Kim
, Kotiba Hamad

Department of Advanced Materials Science and Engineering

Sungkyunkwan University

Research output: Contribution to journal › Article › peer-review

23 Scopus citations

Abstract

In the present work, we employed a deep learning (DL) technique to investigate thermoelectric (TE) materials performance. The figure-of-merit (ZT) determined experimentally for various compositions at a wide range of temperatures with composition features were used to train the DL model. The validation results showed that the built DL model exhibited a reliable accuracy with a cross-validation score R² value of 0.91. In addition, by this model, the experimental behaviors of TE materials (ZT vs temperature) were successfully reproduced and a general prediction of 300,000 composition have been done.

Original language	English
Article number	132299
Journal	Materials Letters
Volume	319
DOIs	https://doi.org/10.1016/j.matlet.2022.132299
State	Published - 15 Jul 2022

Keywords

Deep learning
Figure-of-merit
Seebeck effect
Thermoelectric materials

Access to Document

10.1016/j.matlet.2022.132299

Cite this

@article{de0da561331d4612b566f6dc8478f780,

title = "A deep learning perspective into the figure-of-merit of thermoelectric materials",

abstract = "In the present work, we employed a deep learning (DL) technique to investigate thermoelectric (TE) materials performance. The figure-of-merit (ZT) determined experimentally for various compositions at a wide range of temperatures with composition features were used to train the DL model. The validation results showed that the built DL model exhibited a reliable accuracy with a cross-validation score R2 value of 0.91. In addition, by this model, the experimental behaviors of TE materials (ZT vs temperature) were successfully reproduced and a general prediction of 300,000 composition have been done.",

keywords = "Deep learning, Figure-of-merit, Seebeck effect, Thermoelectric materials",

author = "Russlan Jaafreh and \{Yoo Seong\}, Kang and Kim, \{Jung Gu\} and Kotiba Hamad",

note = "Publisher Copyright: {\textcopyright} 2022 Elsevier B.V.",

year = "2022",

month = jul,

day = "15",

doi = "10.1016/j.matlet.2022.132299",

language = "English",

volume = "319",

journal = "Materials Letters",

issn = "0167-577X",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - A deep learning perspective into the figure-of-merit of thermoelectric materials

AU - Jaafreh, Russlan

AU - Yoo Seong, Kang

AU - Kim, Jung Gu

AU - Hamad, Kotiba

PY - 2022/7/15

Y1 - 2022/7/15

N2 - In the present work, we employed a deep learning (DL) technique to investigate thermoelectric (TE) materials performance. The figure-of-merit (ZT) determined experimentally for various compositions at a wide range of temperatures with composition features were used to train the DL model. The validation results showed that the built DL model exhibited a reliable accuracy with a cross-validation score R2 value of 0.91. In addition, by this model, the experimental behaviors of TE materials (ZT vs temperature) were successfully reproduced and a general prediction of 300,000 composition have been done.

AB - In the present work, we employed a deep learning (DL) technique to investigate thermoelectric (TE) materials performance. The figure-of-merit (ZT) determined experimentally for various compositions at a wide range of temperatures with composition features were used to train the DL model. The validation results showed that the built DL model exhibited a reliable accuracy with a cross-validation score R2 value of 0.91. In addition, by this model, the experimental behaviors of TE materials (ZT vs temperature) were successfully reproduced and a general prediction of 300,000 composition have been done.

KW - Deep learning

KW - Figure-of-merit

KW - Seebeck effect

KW - Thermoelectric materials

UR - https://www.scopus.com/pages/publications/85129549913

U2 - 10.1016/j.matlet.2022.132299

DO - 10.1016/j.matlet.2022.132299

M3 - Article

AN - SCOPUS:85129549913

SN - 0167-577X

VL - 319

JO - Materials Letters

JF - Materials Letters

M1 - 132299

ER -

A deep learning perspective into the figure-of-merit of thermoelectric materials

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this