Accelerated Discovery of Advanced Thermoelectric Materials via Transfer Learning

Mengke Li; Hongmei Yuan; Yufeng LuoXiaoyi MaHuijun LiuWentao JiangYing Fang

首页> 外文期刊>Advanced energy materials >Accelerated Discovery of Advanced Thermoelectric Materials via Transfer Learning

【24h】

Accelerated Discovery of Advanced Thermoelectric Materials via Transfer Learning

机译：Accelerated Discovery of Advanced Thermoelectric Materials via Transfer Learning

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相关主题

摘要

Thermoelectric (TE) technology can realize direct conversion of widely distributedheat into useful electricity, which provides a promising route to solve theglobal energy crisis that is increasingly severe. However, it is extremely complexand time-consuming to discover advanced TE materials via conventionaltrial-and-error approaches. In this work, using a pre-trained neural networkarchitecture for the electronic bandgap, a transfer learning (TL) strategy thatallows ready and accurate prediction on the ZT values of any TE materials atarbitrary temperature is proposed. Compared with direct machine learningalgorithms, the TL-driven model exhibits significantly enhanced predictivepower beyond the initial dataset, as characterized by improved Pearson correlationcoefficient (reduced mean absolute error) from 23% to 95% (0.35 to 0.07)for the p-type systems, and 46% to 94% (0.23 to 0.06) for the n-type systems.By screening 6353 possible candidates in the AFLOW repository that havingrelatively smaller gaps, 925 p-type and 788 n-type systems are quickly identifiedto exhibit ZT exceeding 2.0. Equally importantly, the established TL modelis highly adaptable to ZT prediction in an even larger search space, where theconstituent atoms and/or stoichiometric compositions of the screened systemsmay be variously tuned to further optimize their TE performance.

著录项

来源
《Advanced energy materials》 |2023年第23期|2300049.1-2300049.7|共7页
作者
Mengke Li; Hongmei Yuan; Yufeng LuoXiaoyi MaHuijun LiuWentao JiangYing Fang;
展开▼
作者单位

Key Laboratory of Artificial Micro- and Nano-Structures ofMinistry of Education and School of Physics and TechnologyWuhan UniversityWuhan 430072, China;

School of Computer ScienceWuhan UniversityWuhan 430072, China;

展开▼
收录信息
原文格式 PDF
正文语种英语
中图分类
关键词
bandgap; figure-of-merit; high-throughput screening; thermoelectric materials; transfer learning;

Accelerated Discovery of Advanced Thermoelectric Materials via Transfer Learning

摘要

著录项

相关主题

期刊订阅