| Abstract Scope |
The composition of elements in alloys and metal-matrix composites plays a crucial role in defining their functional properties. Thermoelectricity, the conversion of heat into electricity, is a key characteristic for materials used in energy conversion applications. While thermoelectric properties are traditionally predicted using computationally expensive methods such as density functional theory (DFT), machine learning (ML) offers a more cost-effective alternative when paired with appropriate predictors. This study explores the use of physical properties of elements to develop ML models for predicting the thermoelectric performance of Cu-based alloys and composites. By employing data-driven techniques, this study aims to expedite the discovery and design of novel, low-cost, and environmentally sustainable thermoelectric materials. |