Machine Learning-Assisted Multi-Property Prediction and Sintering Mechanism Exploration of Mullite–Corundum Ceramics

Materials, Год журнала: 2025, Номер 18(6), С. 1384 - 1384

Опубликована: Март 20, 2025

Mullite–corundum ceramics are pivotal in heat transfer pipelines and thermal energy storage systems due to their excellent mechanical properties, stability, chemical resistance. Establishing relationships mechanisms through traditional experiments is time-consuming labor-intensive. In this study, gradient boosting regression (GBR), random forest (RF), artificial neural network (ANN) models were developed predict essential properties such as apparent porosity, bulk density, water absorption, flexural strength of mullite–corundum ceramics. The GBR model (R2 0.91–0.95) outperformed the RF ANN 0.83–0.89 0.88–0.91, respectively) accuracy. Feature importance partial dependence analyses revealed that sintering temperature K2O (~0.25%) positively affected density while negatively influencing porosity absorption. Additionally, temperature, additives, Fe2O3 (optimal content ~5% 1%, related strength. This approach provided new insight into between feedstock compositions process parameters ceramic it explored possible involved.

Язык: Английский

Underground hydrogen storage in geological formations: A review

Journal of Rock Mechanics and Geotechnical Engineering, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025

Язык: Английский

Hydrogen Energy System and Underground Hydrogen Storage in Depleted Reservoirs

Petroleum Research, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025