Thermal conductivity prediction and structure-property relationship of CaO-SiO2-Al2O3 ternary system: A combination of molecular dynamics simulations and machine learning

Abstract

Accurate knowledge of thermal conductivity of amorphous coal ash is essential to a mathematical model of the coal gasification process or to a design of a coal gasification system. However, the existing experimental techniques are largely limited in large-scale measurements or accurate predictions of the thermal conductivity of amorphous coal ash at high temperatures. Herein, molecular dynamics (MD) simulations combined with machine learning (ML) techniques was used to predict the thermal conductivity of CaO-SiO2-Al2O3 (CSA) ternary system. The results showed that the random forest (RF) algorithm has the highest level of accuracy and provides good and reliable predictions over the entire compositional domain. Further composition and structure analysis showed that the higher content of CaO, lower content of Al2O3 and SiO2 contribute to high thermal conductivity. Moreover, the function of CaO connecting NBOs promotes the heat transfer in the amorphous CSA slags. In general, this paper provides an efficient combinational strategy for thermal conductivity prediction of amorphous CSA ternary system as well as the mechanism research.