Machine learning based prediction for bulk porosity and static elastic modulus of Yungang Grottoes sandstone

A new machine learning model, specifically an Artificial Neural Network (ANN), offers highly accurate predictions for the bulk porosity and static elastic modulus of weathered Yungang Grottoes sandstone. This ANN outperformed other regression models like Multiple Linear Regression (MLR) and Support Vector Regression (SVR) with a coefficient of determination (R²) between 0.9537–0.9641 during testing, demonstrating superior precision and generalization. The model's error margins were minimal—no more than 0.7% for bulk porosity and 0.15 GPa for static elastic modulus—while maintaining a stable performance across training and testing phases. These findings enable faster, more reliable assessments of sandstone weathering degrees, with validated predictions for bulk porosity (7–14%) and static elastic modulus (0.7–1.4 GPa).

Zhang, R., Huang, J., Cheng, Y., & Zhang, Y. (2024). Machine learning based prediction for bulk porosity and static elastic modulus of Yungang Grottoes sandstone. Heritage Science, 12(1). https://doi.org/10.1186/s40494-024-01459-z