Generalizable Prediction Model of Molten Salt Mixture Density with Chemistry-Informed Transfer Learning
This work addresses the need for accurate thermophysical property predictions in molten salt applications, such as energy systems, but is incremental as it builds on existing models with a hybrid method.
The paper tackled the problem of predicting molten salt mixture density, which is challenging due to incomplete databases and difficult experiments, by proposing a transfer learning approach using deep neural networks that achieved high accuracy with r² > 0.99 and MAPE < 1%.
Optimally designing molten salt applications requires knowledge of their thermophysical properties, but existing databases are incomplete, and experiments are challenging. Ideal mixing and Redlich-Kister models are computationally cheap but lack either accuracy or generality. To address this, a transfer learning approach using deep neural networks (DNNs) is proposed, combining Redlich-Kister models, experimental data, and ab initio properties. The approach predicts molten salt density with high accuracy ($r^{2}$ > 0.99, MAPE < 1%), outperforming the alternatives.