Strength prediction and drillability identification for rock based on measurement while drilling parameters

• Published in: Journal of Central South University Vol. 30; no. 12; pp. 4036 - 4051
• Main Authors: Wang, Shao-feng, Wu, Yu-meng, Cai, Xin, Zhou, Zi-long
• Format: Journal Article
• Language: English
• Published: Changsha Central South University 01.12.2023; Springer Nature B.V
• Subjects: Boreholes; Classification; Compressive strength; Construction accidents & safety; Drilling; Engineering; Excavation; Machine learning; Mechanical properties; Metallic Materials; Occupational safety; Parameter identification; Regression analysis; Specific energy; drillability identification; TOPSIS-RSR method; 机器学习; TOPSIS-RSR方法; 可钻性分类; measurement while drilling (MWD); 随钻测量(WMD); machine learning; 强度预测; 可钻性识别; drillability classification; strength prediction
• ISSN: 2095-2899; 2227-5223
• DOI: 10.1007/s11771-023-5492-4

Rapid acquisition of rock mechanical parameters and accurate identification of rock drillability are important to guide the safe construction of different scale drilling engineering (wells and boreholes) and high-efficient excavation of rock engineering. A database is established based on 281 sets of drilling parameters and rock mechanical parameters collected from four micro drilling experiments. The drilling parameters in the database include drilling force ( F ), torque ( T ), rotational speed ( N ), and rate of penetration ( V ), from which the specific energy (SE) and the drillability index ( I d ) are calculated. With these parameters as input, fitting regression analysis and machine learning regression are used to predict the uniaxial compressive strength (UCS) of rocks. Furthermore, TOPSIS-RSR method is used to achieve rock drillability classification, and machine learning classification methods are used to perceive and identify drillability. In the prediction and recognition process, the accuracies of different methods are compared to determine the optimal model. The research methods and findings can provide new approaches for real-time in-situ measurement of UCS and drillability classification of rock, providing a basis for improving the efficiencies of drilling and excavation and ensuring the construction safety.