Cement-emulsified asphalt mortar (CA mortar) is an organic–inorganic composite material composed of cement, emulsified asphalt, fine sand, water, and various admixtures. It is mainly used as the cushion layer for high-speed railway ballastless tracks. CA mortar cushion layers in North China often have to withstand the coupling effects of humidity and freeze–thaw, which has a very important impact on the fatigue performance of CA mortar. Based on the big data statistical results, the temperature conditions and cycle times of the CA mortar layer Freeze–Thaw cycle in North China were determined. Also, a fatigue performance test under humidity–freeze–thaw coupling conditions was designed and carried out. The fitting curve equations of fatigue stress and fatigue life under different humidity conditions and freeze–thaw coupling were established. The relationship between fatigue performance parameters K and n and humidity conditions was analyzed. This study shows that with the increase in humidity, the fatigue life of CA mortar under different humidity conditions shows an overall downward trend. The fatigue performance and fatigue life stress level sensitivity of CA mortar decrease with increasing humidity. The proportion of water damage and freeze–thaw damage to total damage increases with increasing humidity, which means that the humidity and freeze–thaw have a more significant impact on the fatigue properties of CA mortar. When the humidity is low, the fatigue cracks of CA mortar are mostly generated across the cement paste, and the macroscopic damage presents as longitudinal cracking. When the humidity is high, the fatigue cracks of CA mortar are mostly generated at the interface between aggregate and paste, and the macroscopic damage presents as oblique cracking. Based on the analysis of the damage mechanism, it is suggested that the humidity of CA mortar should be controlled below 25% in the actual project to ensure its durability.

Subgrade differential settlement can compromise ride safety and operational stability by intensifying wheel–rail interaction and vehicle–track vibrations. Early-grade identification remains challenging due to weak onset response variations and the nonstationarity of acceleration signals. Conventional data-driven models often provide neither physics-consistent indicators nor well-calibrated confidence for threshold-based alarms and risk grading. Physics-Aware Time-Series Mixer with feature-wise linear modulation priors (PATS) is developed for ordinal identification of settlement amplitude A, wavelength L, and lateral coefficient K using carbody lateral and vertical accelerations. PATS integrates (1) improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) to separate low-frequency geometric trends from mid-to-high-frequency impacts, (2) physics-informed descriptors, namely sensitive intrinsic mode function (IMF) energy priors and statistics-based cumulative characteristic function (SCCF) features, injected into a lightweight Time-Series Mixer (TSMixer) backbone via feature-wise linear modulation (FiLM), and (3) a multi-group ordinal head based on consistent rank logits (CORAL) to output reliability-aware probabilities. Experiments show 97.41% accuracy with 0.0187 expected calibration error (ECE) on K at 1.050 ms/sample, 96.30% accuracy with 0.0350 ECE on A at 0.964 ms/sample, and 100.00% accuracy with 0.0011 ECE on L. The artificial intelligence contribution lies in an interpretable physics-guided prior injection mechanism for nonstationary multivariate time-series grading, coupled with a reliability-oriented ordinal output design. The engineering application contribution is a deployable onboard monitoring framework for early-grade identification of high-speed railway subgrade differential settlement from onboard accelerations, enabling threshold-based alarming and risk grading.