Search

The peak dilation angle is an important mechanical feature of rock discontinuities, which is significant in assessing the mechanical behaviour of rock masses. Previous studies have shown that the efficiency and accuracy of traditional experimental methods and analytical models in determining the shear dilation angle are not completely satisfactory. Machine learning methods are popular due to their efficient prediction of outcomes for multiple influencing factors. In this paper, a novel hybrid machine learning model is proposed for predicting the peak dilation angle. The model incorporates support vector regression (SVR) techniques as the primary prediction tools, augmented with the grid search optimization algorithm to enhance prediction performance and optimize hyperparameters. The proposed model was employed on eighty-nine datasets with six input variables encompassing morphology and mechanical property parameters. Comparative analysis is conducted between the proposed model, the original SVR model, and existing analytical models. The results show that the proposed model surpasses both the original SVR model and analytical models, with a coefficient of determination (R2) of 0.917 and a mean absolute percentage error (MAPE) of 4.5%. Additionally, the study also reveals that normal stress is the most influential mechanical property parameter affecting the peak dilation angle. Consequently, the proposed model was shown to be effective in predicting the peak dilation angle of rock discontinuities. [ABSTRACT FROM AUTHOR]

It is well known that joints or fissures have an important effect on the failure mechanism of natural rocks. Previously, many numerical and experimental papers have been carried out to study the strength anisotropy and failure characteristics of jointed rocks. However, few studies have been carried out on the failure mechanism of nonpersistent jointed rock masses with different persistence, especially for nonpersistent joints in three dimensions. In the present study, the failure characteristics of a 3D nonpersistent jointed rock mass with different inclinations (θ) and persistence (K) are studied by numerical simulation. For the 3D digital elevation model (DEM), the linear parallel bond model (LPBM) and smooth-joint model (S-J) were used to model the rock-like material and joint interface, respectively. The connections between the geometric parameters of joints and peak strength are revealed. For the peak strength, the joint persistence only plays a minor role in specimens with inclinations of 0° and 90°, and its influence on strength is mainly reflected in the specimens with shear failure (θ = 45°, 60°, and 75°). Based on microcrack accumulation and evolution, four typical failure processes (shear failure, split failure, mixed failure, and intact failure) are analysed from the micro perspective. The shear stress evolution process on the 3D nonpersistent joint of the specimen with different inclinations under K1 = 0.42 was monitored by the measurement circle, and it was found that the distribution of shear stress inside the rock bridge is related to the failure mode of the specimen. For the specimens with θ = 0° and 90°, the shear stress had little change, indicating that there is slight shear slip behaviour on the joint surface. When the inclination is 45°, 60°, and 75°, the shear stress changes obviously during loading, indicating that the shear action is strong in this failure mode. [ABSTRACT FROM AUTHOR]