| 摘要: |
| 在追求可持续发展的未来，热电材料是不可或缺的。它在全固态发电和制冷方面具有十分广泛的应用前景。在过去的几十年间，人们一直致力于寻找新型高性能热电材料。然而，传统的实验试错法效率较低，限制了新材料的研究步伐。机器学习作为一种具有强大数据分析能力的方法，近年来已越来越多地应用于热电材料的研究。这篇综述总结了热电材料研究领域常用的机器学习方法，系统地介绍了它们在材料结构、电子输运和热电输运等性质上的应用案例和相关研究进展，并对该领域的发展前景进行了展望。 |
| 关键词: 机器学习 热电材料 高通量筛选 |
| DOI： |
|
| 基金项目: |
|
| Machine Learning for the Studies of Thermoelectric Materials |
|
LIU Jianghui, LIU Huijun
|
| (School of Physics and Technology, Wuhan University, Wuhan 430072, China) |
| Abstract: |
| Thermoelectric materials offer a unique opportunity in all-solid-state power generation and refrigeration, which are highly desirable in a sustainable future. During the past few decades, considerable efforts have been devoted to search for high performance thermoelectric materials, which are however limited by the low efficiency of conventional trial-and-error experiments. Machine learning, an approach known for its powerful data-analysis capability, has been widely applied in the study of thermoelectric materials recently. This review first summarizes the machine learning methods commonly used in the field of thermoelectrics, and then focuses on their applications in the studies of structural, electronic, and thermoelectric transport properties. A brief conclusion and perspective are given in the end. |
| Key words: machine learning thermoelectric materials high-throughput screening |
