Early freezing of concrete is common in the construction of water conservancy projects in northern China. Early freezing damage induces the deterioration of the mechanical properties of concrete structures, which seriously affects the safety, stability, and service life of engineering structures. Through a laboratory uniaxial compression test and a computed tomography (CT) test, the influence law of freezing time and freezing temperature on the mechanical properties of concrete is analyzed herein. The three-dimensional pore structure of concrete at different freezing times is reconstructed. The pore distribution and pore structure characteristic parameters of concrete at different freezing times are studied. The correlation between mechanical properties and the pore structure of early frozen concrete is determined. The results show that with the delay of freezing time, the porosity of concrete test blocks increases first and then decreases. The average pore surface area primarily decreases and then increases. The average pore diameter increases with the trend of a quadratic parabola. The average pore form factor primarily decreases, then increases, and then decreases. The average pore surface area has the best correlation with the compressive strength and elastic modulus of early frozen concrete. The outcomes suggest that the average pore surface area should be preferred when constructing the mesoscopic damage index of early frozen concrete. Relevant results provide support for revealing the macro and micro damage mechanisms of early frozen concrete.

The stress state and principal stress axis changes of the stress‐field tensor are analyzed during the advancement of a tunnel face on the basis of a given case study of the Jinping II Hydropower Station in China. First, the prevailing pole diagram in geology is used to illustrate the rotation of the stress axes as the tunnel face advances. The results show that the orientation adjustments of principal stresses in different positions near the tunnel boundary share common characteristics. The major and minor principal stress axes ahead of the tunnel face will rotate to intersect with the excavation surface at an angle, with the intermediate principal stress axis being almost parallel to the excavation surface. Furthermore, the stress triaxiality that is commonly used to indicate the deformation and damage of metal materials is introduced to describe the stress state change of the excavation‐induced stress. The stress triaxiality is found to represent the stress state change due to the variation in both the magnitude and orientation of the stress‐field tensor. According to the physical meaning and the change law of the stress triaxiality, stress disturbance during tunnel excavation can be divided into four stages, and the stress disturbance zone is divided into a strong disturbance zone and a weak disturbance zone. The disturbance characteristics of different stages and the distribution patterns of various zones are analyzed, which may be useful for practical application in the design and construction of rock tunnels.

Surrounding rock and lining are composite structures with asymmetric mechanical properties. Understanding the mechanical properties and failure characteristics of rock–concrete composites is crucial for gaining insights into the mechanisms that induce disasters in deep-underground environments. Uniaxial compression and acoustic emission tests were conducted on rock–concrete composite specimens cured at temperatures of 20 °C, 40 °C, 60 °C, and 80 °C, with interface angles of 15°, 30°, 45°, 60°, 75°, and 90° respectively. The results indicated that the specimens’ strength decreased at increasing geothermal temperatures. The composites with an 80 °C curing temperature and a 60° interface angle exhibited the lowest strength. A higher geothermal temperature significantly reduced the number of cracks in the concrete component during composite failure and mitigated the influence of the inclined interface angle. The failure modes of the specimens include axial penetration splitting, interface shear, Y-shaped fracture, and interface splitting–concrete shear failure. Finally, a model relating the strength of the rock–concrete composite to the inclined interface angle and the geothermal temperature was derived and verified against the experimental results with a relative error of 9.8%. The findings have significant implications for the safety and stability of tunnels in high-temperature conditions.

In the risk assessment of water inrush from coal floors, the amount of measured data obtained through on-site testing is small and random, which limits the prediction accuracy and generalizability of a model based on measured data. Using the distribution characteristics of the measured data and mega-trend diffusion theory, we propose a virtual sample enhancement method based on class distribution mega-trend diffusion technology (CDMTD) and introduce constraints on the class distribution of influencing factors. This method was used to generate virtual samples and enhance the measured database. A prediction model of the water inrush risk for the coal seam floor was established using a coupled algorithm of extreme learning machines, self-adaptive differential evolution, and CDMTD (PCA-CDMTD-SaDE-ELM) and was used to evaluate the water inrush risk in the 19,105 working face of the Yunjialing Mine. The CDMTD method could effectively solve the problem of virtual sample distribution variation in the overall trend diffusion theory and enhance the measured database, reducing the impact of small sample sizes. Compared to other optimization models, our model showed the best prediction performance, with an error reduction of 42.95–51.27% and results biased towards safety. Our results support safe and efficient coal mining above Ordovician limestone-confined water.