The occurrence of rockburst can seriously impact the construction and production of deep underground engineering. To prevent rockburst, machine learning (ML) models have been widely employed to predict rockburst based on some related variables. However, due to the costs and complicated geological conditions, complete datasets to evaluate rockburst cannot always be obtained in rock engineering. To fill this limitation, this study proposed an ensemble tree model suitable for incomplete datasets, i.e., the histogram gradient boosting tree (HGBT), to build intelligent models for rockburst prediction. Three hundred fourteen rockburst cases were employed to develop the HGBT model. The hunger game search (HGS) algorithm was implemented to optimize the HGBT model. The established HGBT model had an excellent testing performance (accuracy of 88.9%). An incomplete database with missing values was applied to compare the performances of HGBT and other ML models (random forest, artificial neural network, and so on). HGBT received an accuracy of 78.8% in the incomplete database, and its capacity was better than that of other ML models. Additionally, the importance of input variables in the HGBT model was analyzed. Finally, the feasibility of the HGBT model was validated by rockburst cases from Sanshandao Gold Mine, China.

The construction of deep underground engineering is greatly influenced by complex geological conditions such as high stress, faults, and fracture zones, which significantly affects the stability of the project. Taking the construction of 1,915-m-deep shaft in Sanshandao Gold Mine as the engineering background, which passes through many different strata and multiple fracture zones, the stability evaluation and failure zone prediction during its excavation under the influence of high stress, uneven strata, and fault structure are studied. Results show that the range of the failure zone increases significantly when the shaft passes through the fracture zone or different lithologies, and the maximum depth is 5.28 m. When the distance between the rock mass in the borehole and the excavation face is greater than 48 m, the disturbance superposition effect basically disappears. This paper provides theoretical and data support for the design and construction of the kilometer-deep shaft in Sanshandao Gold Mine.

Jointed rock mass is a material with anisotropic strength and coexistence of continuous and discontinuous media. In order to study the mechanical behavior of rock mass impacted by the spatial distribution of joints, RFPA^2D, which is a discrete element geotechnical engineering simulation, is employed in this paper. Based on the basic rock mechanical parameters obtained from uniaxial compression experiment, the influence of joint inclination angle on rock strength and strain path is explored. The simulation results show that: Cracks will cause large initial damage to the rock mass, resulting in a significant decrease in the peak strength of the specimen compared with the intact rock mass. Meanwhile, it decreases the stage of elastic deformation, and it has the strongest impact to strain path when joint angle equals to 90°.

In order to study the present state of the tectonic stress field in northwest Jiaodong Peninsula, the characteristics of the ground stress distribution along depth were analyzed with the method of regression analysis. A total of 164 items of in situ stress measurement data were collected. There are mainly two types of in situ stress states, one is σ_H>σ_h>σ_v and the other is σ_>σ_v >σ_h, and the type of in situ stress state is related to the depth. The principal stresses σ_H, σ_h, and σ_v increase approximately linearly with depth, and the stress gradients are 0.0556, 0.0248, and 0.0346, respectively. The lateral pressure coefficients KH, Kh, and Kav vary approximately hyperbolically with the increase in depth and approach 1.99, 1.12, and 1.56, respectively. The ratio of half of the maximum horizontal differential stress to the average horizontal principal stress μ_d varies linearly with the increase of depth and approaches the value of 0.33. The maximum shear stress τ_m increases approximately linearly with depth, and the stress gradients are 0.0146. The ratio of horizontal principal stress difference with vertical stress on 1000 m–2000 m is 0.5– 0.7. In addition, the maximum level of principal stress advantage direction in the northwest Jiaodong Peninsula is nearly E–W.

The problem of high stress conditions is a critical challenge for mining activity in deep metal mines, and the difficulty and economic cost of maintaining stope stability also increases. In order to ensure stope stability in the mining process, it is necessary to evaluate stope stability for a more stable stope structure. However, there are many kinds of methods for rock stability assessment, and most of them are based on empirical and analytical methods, which lead to uncertain results. In this paper, a comprehensive evaluation method including factors of stope stability, self-stabilization time, maximum safety span, and support measure is determined, which is more likely to provide a comprehensive result. Finally, it was verified by the field application in Jiaojia Gold Mine. Results show that it can accurately evaluate the stope stability from different aspects compared with the traditional single evaluation method, and it leads to a more comprehensive and reliable result.

Crustal stress is a critical parameter utilized to analyze the stability of the tunnel in underground hard‐rock mining sites. In order to analyze the distribution law of crustal stress and its influence on the stability of the deep tunnel in the Huanaote mining area, the in‐situ stress field model of this area was constructed firstly by using the borehole stress relief method. Moreover, the numerical model is established based on three real engineering conditions: the excavation direction is parallel, vertical, and intersection to the measured maximum horizontal principal stress direction, respectively. Results show that, compared with the other two layout schemes, the stress of two side walls reaches peak at 27 MPa, when the excavation direction is intersected to the measured maximum horizontal principal stress direction, which indicates that the support and maintenance of two side walls should be strengthened to ensure the stability of the tunnel and the safety of personnel and equipment in real projects.