This paper uses shake table tests to study tunnel landslide failures in earthquake zones under four conditions: (GK1) the tunnel intersects the sliding mass, (GK2) the tunnel is perpendicular to the sliding surface, (GK3) the tunnel is positioned below the sliding surface, and (GK4) the tunnel is situated above the bedrock. The dynamic responses under the four conditions are analyzed using time-domain strain analysis methods. Additionally, from an energy perspective, the amplified Arias intensity (MIa) is employed to characterize the cumulative deformation damage of the tunnel lining. The results indicate that under four working conditions, the upper landslide region of the tunnel landslide system exhibits a settlement-compression-shear type of sliding failure. However, in conditions GK1 and GK2, where the lining structure is present, the tunnel lining provides additional support to the landslide, resulting in less severe damage to the slope compared to conditions GK3 and GK4. However, under conditions GK1 and GK2, the left sidewall of the tunnel lining experiences more severe damage due to landslide pressure. The maximum soil pressure and bending moment on the left sidewalls in GK3 and GK4 are only 40-60% of those observed in GK1 and GK2. In addition, based on the trend of MIa, the cumulative deformation evolution of the tunnel lining can be categorized into three stages: the initial stage (0.1-0.2 g), the progressive deformation stage (0.2-0.4 g), and the failure deformation stage (0.4-0.6 g). Further research confirms that under seismic action, the slope experiences a significant progressive catastrophic evolution. This process is characterized by typical seismic cumulative damage effects, with sustained seismic loading causing deformation and damage to gradually expand from localized areas to the entire slope. This continuous fatigue effect progressively weakens the stability of the lining structure, ultimately leading to its failure. Therefore, the deformation and damage of the slope under seismic loading pose a serious threat to the safety of tunnel linings, highlighting the need for close attention to their long-term stability. The research results provide a scientific basis for reinforcing tunnel linings in earthquake-prone mountainous areas.

The influence of groundwater seepage on tunnel is not negligible, so it is very important to calculate the distribution of pore water pressure and the seepage volume accurately. The current analytical studies of seepage field in tunnels with grouting ring assume the head outside the grouting ring to be constant, which is accurate when there is a large difference between the permeability of the grouting ring and the soil body, but less accurate when there is a small difference in permeability. Accordingly, this paper combines a new conformal transformation method and the separated variable method to overcome the current problem of not being able to obtain an analytical solution after assuming the head at the outer boundary of the grouting circle as a nonconstant head. And according to the obtained analytical solution and the existing analytical solutions and numerical simulation results for comparison, comparison results show that: when 1 <= kr/kg <= 100, the absolute value of the maximum relative error between the calculation results of the external pore water pressure of the grouting circle of the CVM solution and those of the numerical solution is more than 1.5 times of that of the analytical solution of this paper, and the maximum value is nearly 3 times of that of the analytical solution of this paper. Therefore, the analytical solution obtained in this paper by assuming that the outer boundary of the grout ring is a non-constant head is more accurate and has some applicability in the case where the permeability of the soil and the grout ring do not differ much. Finally, extensive parametric analyses of the permeability and thickness of the grouting ring and the depth of the tunnel are also performed to demonstrate the capability of the proposed analytical solution. In addition, the proposed analytical solution is much less computationally demanding compared to numerical software, but the accuracy is comparable.