About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Alloys and Compounds for Thermoelectric and Solar Cell Applications XII
|
Presentation Title |
Predictions via Machine Learning of the Thermoelectric Properties of Doped SnSe Materials with Experimental Validation |
Author(s) |
Holger Kleinke |
On-Site Speaker (Planned) |
Holger Kleinke |
Abstract Scope |
In recent years, machine learning (ML) studies to predict or suggest efficient thermoelectric material have become increasingly important. Previous ML studies have used different literature sources or density functional theory calculations as input. As quality and consistency of data play a crucial role in the success of the models’ outputs, we developed an ML model using data exclusively from our lab experiments with various doped SnSe materials to predict their total thermal conductivity.
The model demonstrated remarkable validation using its simulated outputs via new measurements on hitherto unprepared variants of differently doped SnSe samples.
We anticipate further developments in this project for the discovery of previously unidentified state-of-the-art thermoelectric materials. |
Proceedings Inclusion? |
Planned: |
Keywords |
Energy Conversion and Storage, Sustainability, Machine Learning |