.png){kind=icon}
.png){kind=icon}
Return
Rock Strength Estimation Using Drilling Vibration Signal
.png){kind=icon}
PRE
AI
M
Z
W
S
W
Z
DOI：10.1007/s00603-026-05857-6.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The uniaxial compressive strength (UCS) of rock is a critical parameter in underground engineering design, where rapid acquisition enables intelligent and efficient excavation. This study investigates the relationship between drilling mechanical parameters (penetration rate, thrust force, rotation speed, and torque), vibration signals, and UCS. Drilling tests on mudstone, sandstone, and limestone were conducted using a real hydraulic drill rig from an anchor bolter, with time‑series mechanical parameters and vibration signals measured via non‑destructively installed sensors. An Extreme Gradient Boosting (XGBoost) algorithm was employed to develop an intelligent UCS prediction model. The results show that: (1) preprocessing raw vibration data with variational mode decomposition (VMD) threshold denoising significantly improves noise reduction metrics, confirming its effectiveness; (2) vibration signal frequency amplitude increases proportionally with UCS, and vibration features provide more granular insights into drilling progression than mechanical parameters; (3) the XGBoost‑based UCS prediction model, trained on time‑domain (mean, standard deviation) and time‑frequency (dominant frequency) vibration features, achieves an accuracy of 98.15% with a root-mean-square error of 7.35 MPa. Compared with RF, SVM, and BP models, the RMSE is reduced by 67.3%, 38.4%, and 49.2%, respectively, while the coefficient of determination (R2) is increased by 75.9%, 20.0%, and 12.0%. These findings help upgrade the vibration‑based drilling approach as a reliable in situ geotechnical testing method for rapid UCS estimation.
Keywords:
Measurement while drilling
Vibration signal
Uniaxial compressive strength
Machine learning
Journal
IF:
6.6
Papers:
6.0K
Citations:
3.0W
Organization
Cited Papers