Rockbursts pose a significant threat to human safety and environmental stability. This paper aims to predict rockburst intensity using a machine learning model. A dataset containing 344 rockburst cases was collected, with eight inducing features as input and four rockburst grades as output. In the preprocessing stage, missing feature values were estimated using a regression imputation strategy. A novel approach, which combines feature selection (FS), t-distributed stochastic neighbor embedding (t-SNE), and Gaussian mixture model (GMM) clustering, was proposed to relabel the dataset. The effectiveness of this approach was compared with common statistical methods, and its underlying principles were analyzed. A voting ensemble strategy was used to build the machine learning model, and optimal hyperparameters were determined using the tree-structured Parzen estimator (TPE), whose efficiency and accuracy were compared with three common optimization algorithms. The best combination model was determined using performance evaluation and subsequently applied to practical rockburst prediction. Finally, feature sensitivity was studied using a relative importance analysis. The results indicate that the FS + t-SNE + GMM approach stands out as the optimum data preprocessing method, significantly improving the prediction accuracy and generalization ability of the model. TPE is the most effective optimization algorithm, characterized simultaneously by both high search capability and efficiency. Moreover, the elastic energy index W_et, the maximum circumferential stress of surrounding rock σ_θ, and the uniaxial compression strength of rock σ_c were identified as relatively important features in the rockburst prediction model.

Loess tunnels often undergo large-scale deformation with complex spatial and temporal distribution. Mastering the characteristics of spatial and temporal deformation is conducive to precise policy implementation and the control of large deformation of the tunnel. In this study, relying on the Yulinzi Tunnel in Gansu Province, China, based on 3D laser scanning technology, the tunnel was monitored for a short period of 24 h and a long period of 36 days. The refined characteristics of the temporal and spatial deformation of the representative points of the interrupted surface, the tunnel face, and the excavation mileage during the excavation process of the three-bench and seven-step method of the tunnel were analyzed. The results show that the tunnel’s arch has large deformation, and there is twisting deformation. The distribution of the overall deformation of the tunnel is related to the excavation sequence, showing a stepped deformation law. With the construction of the following excavation process, the deformation rate of the tunnel always indicates the characteristics of significant in the early stage and small in the later stage, and the overall deformation changes with time in accordance with the distribution law of the exponential function. The research results provide a reference for predicting the deformation development trend of loess tunnels and providing reasonable deformation control methods.

(1) Background: Lining voids and macroscopic water freezing are both prominent issues that threaten driving safety in high-speed railway tunnels. With the continuous expansion of the scale of high-speed rail tunnels in extremely cold areas in China, the issues of lining voids, water accumulation, and frost heaving have triggered heated discussions. The need to reveal the frost-heave mechanism is urgent. (2) Methods: Firstly, a simulation experiment of the frost heaving of accumulated water was carried out based on the discharge conditions of accumulated water. Secondly, a numerical model was used to study the evolution process of frost heaves caused by accumulated water in voids. Finally, a drainage coefficient was introduced to propose a method for calculating the frost-heave force of accumulated water in voids. (3) Results: The blockage of the drainage channel leads to the generation of frost-heave force; the freezing/thawing process of the void water develops from the thinnest part to the thickest part of the void edge, and the freeze/thaw cycle of the water body causes the frost-heave force to become greater and greater in the evacuated cavity. The higher the height and the closer the drainage channel is to the hollow bottom, the greater the frost-heave force when the accumulated water freezes. (4) Conclusions: When the evacuated water freezes, it develops from the thinner edge to the thicker center. The magnitude of the frost-heave force is affected by the freeze/thaw cycle, the height of the evacuated cavity, and the position of the drainage channel.

Rocks present complex deformation behaviours and damage processes under triaxial cyclic loading—a subject not yet sufficiently researched. This paper performed triaxial multistage constant-amplitude cyclic loading experiments under different confining stresses on carbonaceous phyllite. The degradation process is analysed by investigating the variation of elastic modulus E_S, Poisson’s ratio υ, irreversible strain ε^irr and energy. Moreover, the rock’s failure mode is explored from both macro and micro perspectives. The results showed that the increase in stress level caused the decrease of E_S in a step-like form, and the constant-amplitude cyclic loading in each stress level caused a slow decrease of E_S, while the υ increased with stress level and constant-amplitude cycles in a similar form. ε^irr accumulated rapidly at first and then slowly at each stress level; the stress level and irreversible axial strain are related by an exponential function. In terms of energy evolution analysis, the damage to rock can be represented by the cumulative damage energy, there were deceleration accumulations and stability accumulation stages of damage at all stress levels, and an acceleration accumulation stage occurred when the rock was close to failure. The failures of rock under cyclic loading are mainly shear failures, accompanied by grain crushing.

In the present study, large-scale specimens based on the tunnel prototype were prepared and static load tests were carried out to investigate the damage caused by lining voids. Based on the strengthening scheme of the tunnel, the strengthened specimens were prepared to explore the strengthening effect on the strengthening structure. The strengthening structure is made of a steel plate fixed with chemical anchor bolts and two-component epoxy adhesive. By analyzing the failure mode, structural deformation, and the relationship between load and strain, the damage caused by vault void with various void heights was analyzed and the obtained results were verified through the experiment. Moreover, the enhancement of the bearing capacity and stiffness of the structure strengthened by surface bonding steel was studied. The obtained results show that the damage caused by the lining void mainly occurs at the void boundary. The damage appears as multiple longitudinal cracks. The crack starts from the lower surface and develops radially. Using chemical anchor bolts and two-component epoxy adhesive to bond the steel plate on the lining surface, the damage can be reduced, and the bearing capacity of the structure can be improved effectively when the void height is a quarter of the second lining thickness, the number of cracks is reduced from 14 to 5 after steel plate strengthening, and the length of the longest crack is reduced from 13.2 cm to 8.3 cm, reduced by 37.12%. The steel plate strengthening also reduces the strain of the lower steel bar at the void boundary from 1130.58 με to 555.12 με, and the strain decreases by 50.89%. The experimental results show that the position where the void has the greatest impact on the lining is at the void boundary. Therefore, when steel plates are used to strengthen the void lining, the void boundary should be emphasized, which makes the strengthening more accurate and saves the cost of treatment.

Foot steel pipe was the main arch foot supporting structure to control large deformation of loess tunnels, but the supporting effect was not ideal. Taking Yulinzi Tunnel in Qingyang, Gansu Province, as the engineering background, the design concept and implementation scheme of a foot steel pipe active bearing was put forward. The purpose was to solve the problem that it was difficult to control the surrounding rock settlement with the foot steel pipe. Numerical simulation and field experiments were used to verify the effect of the active bearing technology of the foot steel pipe. The main conclusions were as follows: (1) The effect of increasing the diameter of the foot steel pipe is better than that of increasing the number of foot steel pipes. (2) The active bearing mode of exerting its bearing capacity in advance by prepressing the foot steel pipe can effectively reduce the settlement of the vault. The settlement rate of the vault can be reduced by about 70% in 1–2 days and more than 50% in 1–3 days. (3) At the initial stage of surrounding rock deformation, this technology can provide a large bearing capacity, thereby reducing the overall deformation of the surrounding rock, slowing down the release of the surrounding rock pressure, and playing a positive role in the settlement control of the vault.