In this study, tensile splitting tests and corresponding numerical simulations are performed on high strength concrete, low strength concrete, the interface between the two types of concrete, and the interface reinforced by steel fiber, respectively. The tensile splitting strength, crack initiation load, and damage characteristics are analyzed based on experiment data and fracture surface of samples. It can be concluded that tensile splitting strength and crack initiation load have the descending order of ‘HT’ (high strength concrete sample) > ‘LT’ (low strength concrete sample) > ‘FT’ (interface reinforced by steel fiber) > ‘OT’ (interface). The tensile splitting strength is related not only to the roughness of the fracture surface, but also to the percentage of fractured aggregates. The steel fiber can increase initiation cracking load, peak load and residual strength of the interface. The existing of interface composited by two types of material can cause asymmetric distribution of the stress state, even if geometry and the load are symmetrical for samples.

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.

This study investigates the dynamic response and local stability of gravity-anchored foundations constructed in layered argillaceous sandstone under the coupled effects of wet–dry cycling degradation and cyclic vehicle loads. Based on in situ direct shear tests and FLAC3D 7.0 numerical simulations, a concrete–rock interface model, a rock mass direct shear model, and a three-dimensional dynamic model of the anchored foundation were developed. The parameters of the interface model were validated using the results of the direct shear tests. Wet–dry cycling degradation was subsequently incorporated to analyze the cyclic shear response of the interface and rock mass under different numbers of cycles. Cyclic vehicle loads were modeled as increments in main cable tension with an equivalent sinusoidal waveform. The results indicate that as the number of wet–dry cycles increases, the cyclic shear hysteresis loops shift overall toward lower shear stress levels. Peak shear stress decreases by approximately 49.26–51.64% compared to the natural state, and the hysteresis loop area decreases significantly. This indicates that wet–dry cyclic degradation weakens the cyclic shear resistance and energy dissipation capacity of the contact surface and rock structural planes. Dynamic analysis results for the anchor foundation indicate that wet–dry cycling degradation significantly increases the displacement response levels of the rock mass near the front toe and rear heel. Specifically, under the n = 20 condition, the displacement at the last peak increased by approximately 109.3–123.9% compared to the undisturbed state; simultaneously, the local plastic zones in the rock mass surrounding the anchorages gradually expanded, and the local safety factors of the rock mass near the toe and heel decreased overall. This study elucidates the degradation mechanisms and dynamic behavior of gravity anchors under the combined action of environmental and operational loads, providing a basis for the design and safety assessment of foundations for long-span suspension bridges.

Polyurethane-solidified ballast (PSB) technology and geogrid reinforcement can be used to strengthen ballast beds. However, there has been little research on the mechanical properties of ballast beds under the combined effects of these two methods. This study addresses this research gap through an integrated experimental–numerical investigation of the cooperative reinforcement mechanisms. Uniaxial compression and oblique shear tests were conducted on standard and geogrid-reinforced PSB specimens, and a three-dimensional random aggregate finite element model was developed to represent the actual ballast gradation. The results show that geogrid reinforcement increased the compressive strength by 42.1% and reduced the maximum lateral deformation by 14.2%, while having little influence on the elastic modulus and Poisson’s ratio. In the oblique shear tests, geogrid reinforcement increased the peak loads by 14.2–23.0% compared with the standard specimen. A directional dependence was also observed, with the highest peak load occurring when the shear plane corresponded to the unreinforced surface. These findings clarify the cooperative reinforcement mechanism between polyurethane solidification and geogrid confinement and provide a mechanical basis for designing reinforced ballasted trackbeds in heavy-haul railways subjected to high axle loads and pronounced lateral deformation, as well as in high-speed railway lines requiring enhanced track stability.