Long-term prediction modeling of shallow rockburst with small dataset based on machine learning
Rockburst present substantial hazards in both deep underground construction and shallow depths, underscoring the critical need for accurate prediction methods. This study addressed this need by collecting and analyzing 69 real datasets of rockburst occurring within a 500 m burial depth, which posed...
Saved in:
Published in | Scientific reports Vol. 14; no. 1; pp. 16131 - 12 |
---|---|
Main Authors | , , , , , , , , |
Format | Journal Article |
Language | English |
Published |
England
Nature Publishing Group
12.07.2024
Nature Portfolio |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Rockburst present substantial hazards in both deep underground construction and shallow depths, underscoring the critical need for accurate prediction methods. This study addressed this need by collecting and analyzing 69 real datasets of rockburst occurring within a 500 m burial depth, which posed challenges due to the dataset's multi-categorized, unbalanced, and small nature. Through a rigorous comparison and screening process involving 11 machine learning algorithms and optimization with KMeansSMOKE oversampling, the Random Forest algorithm emerged as the most optimal choice. Efficient adjustment of hyper parameter was achieved using the Optuna framework. The resulting KMSORF model, which integrates KMeansSMOKE, Optuna, and Random Forest, demonstrated superior performance compared to mainstream models such as Gradient Boosting (GB), Extreme Gradient Boosting (XBG), and Extra Trees (ET). Application of the model in a tungsten mine and tunnel project showcased its ability to accurately forecast rockburst levels, thereby providing valuable insights for risk management in underground construction. Overall, this study contributes to the advancement of safety measures in underground construction by offering an effective predictive model for rockburst occurrences. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 2045-2322 2045-2322 |
DOI: | 10.1038/s41598-024-64107-3 |