AI and knowledge driven computation of rock mass characteristic parameters across engineering projects
Zewang Zheng, Yunpei Zhang, Quan Xu, Kezhong Wang, Changchun Ren, Hao Cai · 2026
Read AI and knowledge driven computation of rock mass characteristic parameters across engineering projects by Zewang Zheng, Yunpei Zhang, Quan Xu, Kezhong Wang, Changchun Ren, Hao Cai (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-13493807.txt.
Description
Zheng and colleagues address a practical tunnelling problem: knowing rock quality ahead of a tunnel boring machine helps crews build safely and quickly. Each new project, though, starts with little data of its own. Earlier methods leaned on physics definitions that wobble when sensor data get noisy, or on fitted curves that fail when geology changes. Their 2026 approach computes features from boring machine data in a knowledge-guided way, then feeds them to learning models trained across projects.
The authors compared five algorithms and found CatBoost strongest on accuracy and F1-score, a combined measure of precision and recall. Tested on three separate tunnel projects, the method predicted rock class with accuracies in the mid-to-high eighties. Because the features transfer between sites, a new tunnel can borrow strength from finished projects instead of gathering its own long history. That supports real-time decisions at the tunnel face.
Catalog record
| CREATOR | Zewang Zheng, Yunpei Zhang, Quan Xu, Kezhong Wang, Changchun Ren, Hao Cai |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |