Journal of Rock Mechanics and Geotechnical Engineering最新文献

Vesicles can be of different sizes and shapes and can be randomly distributed within vesicular volcanic rocks. This study investigates the variation of engineering properties of vesicular rocks due to the changes in vesicle distribution characteristics for different cases of bulk porosity and vesicle diameter using a systematic numerical simulation program using the finite element method-based rock failure process analysis (RFPA) software. Models with uniform-size vesicles and combinations of different proportions of different-sized vesicles were considered to resemble natural vesicular rocks more closely, and ten different random vesicle distributions were tested for each case. Increasing bulk porosity decreased the uniaxial compressive strength (UCS) and elastic modulus of the specimens, and the specimens with the lowest bulk porosity showed the greatest range of UCS values in the case of uniform-size vesicles. The effect of vesicle diameter on UCS showed an unsystematic response which was understood to be a result of different vesicle distribution patterns, some of which facilitated a shear failure. Specimens with multiple-size vesicles in different proportions revealed that the variation of UCS due to vesicle distribution characteristics is minimum when the bulk porosity is equally shared by different size vesicles. In addition, when the proportion of smaller-sized vesicles is higher, UCS showed an increase compared to that of the equal proportion of different size vesicles case at low porosities, but a decrease at higher porosities. Variation of elastic modulus showed minor, unsystematic fluctuations as a function of vesicle diameter and different proportions of different-sized vesicles, and the range for different vesicle distribution patterns was narrow in general. Overall, the findings of this study recommend cautious use of the engineering properties determined through a limited number of laboratory tests on vesicular rocks.

Rock slopes are usually reinforced by a number of rock bolts due to the high efficiency and low price. However, where should the rock bolts be installed is still a troublesome issue. For anti-dip bedding rock slopes (ABRSs), the installation position of rock bolts is a controlling factor that determines the reinforcement effect. In this work, a theoretical method is firstly proposed for assessing the stability of ABRSs reinforced by rock bolts using a limit equilibrium model. A comparison of theoretical calculations and numerical results was conducted to test the correctness of the theoretical method. Based on the stability assessment of ABRSs, we introduce adaptive moment estimation method (Adam) to optimize the installation location of rock bolts. Using Adam optimizer, the optimal layout of rock bolts with the maximum factor of safety can be determined, and the factor of safety of the slope increases by about 25% using the same amount of rock bolts but with different installation locations. The proposed method enables the fast stability analysis and supporting design for reinforced ABRSs, which paves the way to smart supporting design of slopes.

For a special geological structure of columnar jointed rock mass (CJRM), its mechanical properties are strongly affected by the columnar joints. To describe the fracture behaviors of CJRM using the basic theories of interface mechanics for composite materials, the interface stresses of the vertical and horizontal joints, which are the two primary joints in the CJRM under triaxial compression, are studied, and their mathematical expressions are derived based on the superposition principle. Based on the obtained interface stresses of the vertical and horizontal joints in the CJRM, the crack initiation of the joint interface in the CJRM is studied using the maximum circumferential stress theory in fracture mechanics. Moreover, based on this investigation, the fracture behaviors of CJRM are analyzed. According to the results of similar material physical model tests for the CJRM, the theoretical study is verified. Finally, the influence of the mechanical parameters of the CJRM on the joint interface stress is discussed comprehensively.

During tunnel boring machine (TBM) excavation, lithology identification is an important issue to understand tunnelling performance and avoid time-consuming excavation. However, site investigation generally lacks ground samples and the information is subjective, heterogeneous, and imbalanced due to mixed ground conditions. In this study, an unsupervised (K-means) and synthetic minority oversampling technique (SMOTE)-guided light-gradient boosting machine (LightGBM) classifier is proposed to identify the soft ground tunnel classification and determine the imbalanced issue of tunnelling data. During the tunnel excavation, an earth pressure balance (EPB) TBM recorded 18 different operational parameters along with the three main tunnel lithologies. The proposed model is applied using Python low-code PyCaret library. Next, four decision tree-based classifiers were obtained in a short time period with automatic hyperparameter tuning to determine the best model for clustering-guided SMOTE application. In addition, the Shapley additive explanation (SHAP) was implemented to avoid the model black box problem. The proposed model was evaluated using different metrics such as accuracy, F1 score, precision, recall, and receiver operating characteristics (ROC) curve to obtain a reasonable outcome for the minority class. It shows that the proposed model can provide significant tunnel lithology identification based on the operational parameters of EPB-TBM. The proposed method can be applied to heterogeneous tunnel formations with several TBM operational parameters to describe the tunnel lithologies for efficient tunnelling.

Anthropogenic activity-induced sinkholes pose a serious threat to building safety and human life nowadays. Real-time detection and early warning of sinkhole formation are a key and urgent problem in urban areas. This paper presents an experimental study to evaluate the feasibility of fiber optic strain sensing nerves in sinkhole monitoring. Combining the artificial neural network (ANN) and particle image velocimetry (PIV) techniques, a series of model tests have been performed to explore the relationship between strain measurements and sinkhole development and to establish a conversion model from strain data to ground settlements. It is demonstrated that the failure mechanism of the soil above the sinkhole developed from a triangle failure plane to a vertical failure plane with increasing collapse volume. Meanwhile, the soil-embedded fiber optic strain sensing nerves allowed deformation monitoring of the ground soil in real time. Furthermore, the characteristics of the measured strain profiles indicate the locations of sinkholes and the associated shear bands. Based on the strain data, the ANN model predicts the ground settlement well. Additionally, micro-anchored fiber optic cables have been proven to increase the soil-to-fiber strain transfer efficiency for large deformation monitoring of ground collapse.