The high-speed railway subgrade compaction quality is controlled by the compaction degree (K), with the maximum dry density (ρ_dmax) serving as a crucial indicator for its calculation. The current mechanisms and methods for determining the ρ_dmax still suffer from uncertainties, inefficiencies, and lack of intelligence. These deficiencies can lead to insufficient assessments for the high-speed railway subgrade compaction quality, further impacting the operational safety of high-speed railways. In this paper, a novel method for full-section assessment of high-speed railway subgrade compaction quality based on ML-interval prediction theory is proposed. Firstly, based on indoor vibration compaction tests, a method for determining the ρ_dmax based on the dynamic stiffness K_rb turning point is proposed. Secondly, the Pso-OptimalML-Adaboost (POA) model for predicting ρ_dmax is determined based on three typical machine learning (ML) algorithms, which are back propagation neural network (BPNN), support vector regression (SVR), and random forest (RF). Thirdly, the interval prediction theory is introduced to quantify the uncertainty in ρ_dmax prediction. Finally, based on the Bootstrap-POA-ANN interval prediction model and spatial interpolation algorithms, the interval distribution of ρ_dmax across the full-section can be determined, and a model for full-section assessment of compaction quality is developed based on the compaction standard (95%). Moreover, the proposed method is applied to determine the optimal compaction thicknesses (H₀), within the station subgrade test section in the southwest region. The results indicate that: (1) The PSO-BPNN-AdaBoost model performs better in the accuracy and error metrics, which is selected as the POA model for predicting ρ_dmax. (2) The Bootstrap-POA-ANN interval prediction model for ρ_dmax can construct clear and reliable prediction intervals. (3) The model for full-section assessment of compaction quality can provide the full-section distribution interval for K. Comparing the H₀ of 50~60 cm and 60~70 cm, the compaction quality is better with the H₀ of 40~50 cm. The research findings can provide effective techniques for assessing the compaction quality of high-speed railway subgrades.

This study proposes a new waterproof sealing layer to reduce the impact of water on subgrade beds. The proposed waterproof sealing layer is composed of a polyurethane adhesive (PA) mixture, which aims to control interlaminar deformation and prevent seepage. A variety of laboratory tests were first performed to analyze the attenuation characteristics and mechanical properties of various polyurethane polymer (PP)-improved gravel mixtures under thermohydraulic coupling effects. In addition, a waterproof performance model test of the PP-improved gravel layer was conducted to investigate its waterproof and drainage performance and hydraulic damage mechanism. Finally, the feasibility and effectiveness of the surface structure of the waterproof drainage subgrade bed containing the PA mixture was tested in combination with the treatment project of the Ciyaowan station of the Baoshen heavy-haul railway. According to the experimental and model results, (1) the waterproof layer containing the polyurethane mixture exhibited satisfactory stiffness, elasticity and flexibility. The waterproof layer containing the polyurethane mixture also controlled the deformation between layers, and its mechanical properties remained stable. (2) The waterproof layer with the dense polyurethane mixture performed well in terms of the waterproof aspect, and no infiltration occurred under cyclic load (3). Long-term field monitoring revealed that the effect of the implementation of a PP-improved gravel layer to treat mud pumping was remarkable. The settlement of the PP-improved gravel layer only reached 13.21 mm, and the settlement remained stable in the later stage.