Utilizing machine learning and digital twin technology for rock parameter estimation from drilling data

Rock parameters, including the uniaxial compressive strength (UCS), cohesion, and friction angle, are fundamental in geological and geotechnical engineering. The traditional methods for determining these parameters are often constrained by limitations such as complex operational procedures, time consumption, and difficulty in obtaining in situ parameters. This has led to an exploration of extracting rock parameters through drilling. This paper offers a comprehensive review of the current methodologies for estimating rock parameters derived from drilling tests, encompassing experimental, analytical, numerical, and machine learning (ML) approaches, as well as digital twin (DT) technologies. This review provides an in-depth analysis of recent advancements and research, highlighting the transformative impact of these innovations on the field. It emphasizes the growing application of ML algorithms such as artificial neural networks (ANNs), support vector regression (SVR), random forest (RF), and convolutional neural networks (CNNs) for rock property estimation, underscoring the diversity of techniques utilized. The integration of the DT technology with numerical methods, including finite element analysis (FEA) and discrete element model (DEM), facilitates real-time monitoring, simulation, and optimization of rock parameters extracted from drilling, which can enhance accuracy and robustness even with limited empirical data, representing a notable solution. This review aims to provide a detailed understanding of the application and effectiveness of these methodologies for extracting rock parameters from drilling data.