Freezing depth prediction of surrounding rock in seasonally frozen tunnels based on bayes-optimized XGBoost
Jinglong Li, Jianqing Jia · 2026
Read Freezing depth prediction of surrounding rock in seasonally frozen tunnels based on bayes-optimized XGBoost by Jinglong Li, Jianqing Jia (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13458521.txt.
Description
Jinglong Li and Jianqing Jia predict how deep frost penetrates tunnel rock in cold regions in a 2026 study. Freeze-thaw cycles drive frost heave forces into tunnel linings, and freezing depth governs where those forces act, yet temperature, moisture and rock properties interact in complex ways. The team built a hydrothermal coupling model, generated a dataset by Latin Hypercube sampling plus numerical simulation, and trained a Bayes-optimised XGBoost model with SHAP explanations, checked against the Stefan solution.
The optimised model beat random forest, support vector regression and LightGBM, and proved sensitive to temperature, moisture and thermal conductivity but not specific heat. Higher thermal conductivity deepened freezing, while higher specific heat, moisture, initial rock temperature and coldest-month temperature shallowed it. Maximum error against field data was only 7.9 percent versus 22.4 percent for Stefan, giving cold-region designers a sharper frost tool.
Catalog record
| CREATOR | Jinglong Li, Jianqing Jia |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |