The excavation stability of tunnel is a key problem in tunneling engineering. A mechanical model is developed to determine the boundary curve of the plastic zone and the principal stress distribution of the surrounding rock of a circular tunnel under non-tectonic stress. Specifically, the effects of different influence factors on the shape of the plastic zone and the principal stress of the surrounding rock are quantitatively analyzed. Subsequently, the Mohr-Coulomb failure criterion is utilized to determine the yield failure of the surrounding rock mass. Furthermore, technical approaches and control measures for supporting the surrounding rock are put forward. In particular, the "long and short" coordinated hierarchical support technology with "anchor cables (or long bolts) + bolts" as the main support strategy has been recommended. Finally, the proposed support technology is validated for several typical tunnels by the numerical simulation method. The results indicate that the stability of the surrounding rock of a circular tunnel can be improved by restraining the malignant extension of the plastic zone, improving the principal stress environment, and allowing uniform distribution of the plastic zone within the controllable range of the support system. These results are helpful for optimization of support technology to stabilize the surrounding rock of a circular tunnel under non-tectonic stresses.
Traditional laboratory tests for measuring rock uniaxial compressive strength (UCS) are tedious and time-consuming. There is a pressing need for more effective methods to determine rock UCS, especially in deep mining environments under high <i>in-situ</i> stress. Thus, this study aims to develop an advanced model for predicting the UCS of rock material in deep mining environments by combining three boosting-based machine learning methods with four optimization algorithms. For this purpose, the Lead-Zinc mine in Southwest China is considered as the case study. Rock density, P-wave velocity, and point load strength index are used as input variables, and UCS is regarded as the output. Subsequently, twelve hybrid predictive models are obtained. Root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (<i>R<sup>2</sup></i>), and the proportion of the mean absolute percentage error less than 20% (A-20) are selected as the evaluation metrics. Experimental results showed that the hybrid model consisting of the extreme gradient boosting method and the artificial bee colony algorithm (XGBoost-ABC) achieved satisfactory results on the training dataset and exhibited the best generalization performance on the testing dataset. The values of <i>R<sup>2</sup></i>, A-20, RMSE, and MAE on the training dataset are 0.98, 1.0, 3.11 MPa, and 2.23 MPa, respectively. The highest values of <i>R<sup>2</sup></i> and A-20 (0.93 and 0.96), and the smallest RMSE and MAE values of 4.78 MPa and 3.76 MPa, are observed on the testing dataset. The proposed hybrid model can be considered a reliable and effective method for predicting rock UCS in deep mines.
Ore bodies incubating within fault zones are a common phenomenon in geological strata and pose a huge challenge for underground mining. To effectively exploit mineral resources, the layout of the mining roadway and the interaction between the roadway and geological fault must be considered. In this paper, a bonding fracture was formed on granite samples to simulate a closed fault, under which a circular hole was fabricated to simulate the roadway of the gold mine. We performed a biaxial compression test at a true-triaxial electrohydraulic servo testing system for granite samples with a combined fracture-hole structure. It is worth noting that the fracture inclination β and relative distance between fracture and hole L were taken into account. The digital image correlation (DIC) technique was used to observe the displacement and strain field evolution around the fracture-hole structure. Our results demonstrate that (1) the strength of the granite sample decreases with increasing bonding fracture dip angle β, and the displacement drops between the hanging wall and foot wall raised in both the horizontal and vertical displacement directions. Macroscopic cracks become dense, and the failure degree becomes severe around simulated fault areas. (2) With the increase in the distance L, the strength of the granite sample increases, the influence of the hole on the slip of the fracture plane is weakened, and the discontinuity of displacement becomes less obvious. (3) The maximum principal strain field quantitatively reveals the details of the crack initiation, propagation, and coalescence around the fracture-hole structure, and displacement nucleation is observed in the vertical displacement field.
Abstract Rockburst is a severe geological hazard that restricts deep mine operations and tunnel constructions. To overcome the shortcomings of widely used algorithms in rockburst prediction, this study investigates the ensemble trees, i.e., random forest (RF), extremely randomized tree (ET), adaptive boosting machine (AdaBoost), gradient boosting machine, extreme gradient boosting machine (XGBoost), light gradient boosting machine, and category gradient boosting machine, for rockburst estimation based on 314 real rockburst cases. Additionally, Bayesian optimization is utilized to optimize these ensemble trees. To improve performance, three combination strategies, voting, bagging, and stacking, are adopted to combine multiple models according to training accuracy. ET and XGBoost receive the best capabilities (85.71% testing accuracy) in single models, and except for AdaBoost, six ensemble trees have high accuracy and can effectively foretell strong rockburst to prevent large-scale underground disasters. The combination models generated by voting, bagging, and stacking perform better than single models, and the voting 2 model that combines XGBoost, ET, and RF with simple soft voting, is the most outstanding (88.89% testing accuracy). The performed sensitivity analysis confirms that the voting 2 model has better robustness than single models and has remarkable adaptation and superiority when input parameters vary or miss, and it has more power to deal with complex and variable engineering environments. Eventually, the rockburst cases in Sanshandao Gold Mine, China, were investigated, and these data verify the practicability of voting 2 in field rockburst prediction.
