To investigate the degree of stability deterioration in toppling dangerous rock mass under different action times, firstly, the vibration load generated by train travel in a single action and its decay law were studied, and the theoretical relational equation of rock strength deterioration under the action of vibration loading was obtained by analyzing the increase in half axle expansion of micro fissures inside the rock. Secondly, the freezing stress of the microfissures was determined according to the rock freezing and swelling theory, and the theoretical relational equation of rock strength deterioration under the action of freeze-thaw cycles was obtained by integrating the rock confinement and moisture migration conditions of the fissures wall. The stability coefficient of the toppling dangerous rock mass exhibited a rapid and then slow decreasing trend with increasing number of actions, which was non-linearly negatively correlated with the train running speed and the initial porosity of the rock, and non-linearly positively correlated with the freezing temperature. Controlling the freezing temperature and initial porosity of rock can significantly reduce the degree of deterioration in the stability of dangerous rock mass.

The shear strength and resistance of granular materials are critical indicators in geotechnical engineering and infrastructure construction. Both sliding and rotation influence the energy evolution of soil granular motion during shear. To examine the effects of particle rotation on shear damage and energy evolution in granular systems, we first describe the transformation of irregularly shaped particles into regular shapes via geometrical parameters, ensuring the invariance of energy density and density. We then analyze the impact of particle rotation on shear-stress variation and energy dissipation through a shear energy evolution equation. Additionally, we establish the relationship between the shear-stress ratio and normal stress, considering particle rotation. Finally, we verify the influence of particle rotation on energy evolution and shear damage through shear tests on irregular calcareous sand and regular silica-bead particles. The results indicate that granular materials do not fully comply with the Coulomb strength criterion. In the initial shear stage, most of the external work is converted into granular rotational-shear energy, whereas in the later stage, it primarily shifts to granular sliding-shear energy. Notably, the sensitivity of the granular rotational energy to a vertical load is significantly greater than that of the granular sliding energy.

The settlement values of subway tunnels during the construction period exhibit significant nonlinear and spatial–temporal variation characteristics. To overcome the problems of historical data interference and spatiotemporal characteristics in tunnel settlement prediction models, this paper proposes a tunnel settlement prediction method based on data decomposition, reconstruction, and optimization. First, the original data are optimized via the SSA, which has global optimization capability, high noise immunity, and high adaptivity. The original signal is subsequently decomposed into multiple subsignal sequences via a variational modal decomposition (VMD) algorithm combined with a rolling decomposition technique. Finally, the decomposed signals are fed into the machine learning model to construct a high-precision settlement prediction model based on rolling decomposition. The prediction accuracy of different models was analyzed via the measured settlement data during the construction period of the Beijing subway as an example. The results show that the prediction model with the integrated decomposition algorithm reduces the RMSE and MAE by 33% and 37%, respectively, which significantly improves the prediction accuracy and generalization ability of the neural network to meet the demand of practical engineering prediction and simultaneously enhances the risk warning ability of the model.