The research investigated the mechanical behaviors of ice- soil interface under different temperatures (- 1 degrees C, - 2 degrees C, and - 3 degrees C) and initial soil moisture contents (14%, 16% and 18%) using direct shear tests. The shear stresses of interface were described by generalized Duncan-Chang model and the shear stiffness expression of interface was derived. The results showed that at temperature of - 1 degrees C, the effect of initial moisture content on the interface strength was minimal. As the temperature drops to - 2 degrees C and - 3 degrees C, the higher initial moisture content in the soil enhances the bonding strength of interface. The shear stiffness and strength were sensitive to the initial moisture content at temperature of - 3 degrees C, as initial moisture content increases from 14 to 18%, the shear stiffness increases by 55.5% from 9.09 x 104 kPa/m to 2.04 x 105 kPa/m, the peak strength increases by 43.5% from 156.8kPa to 277.3kPa, and the residual strength increases by 51.1% from 114 to 233kPa. The finite element model of the ice- soil interface was established using COMSOL Multiphysics and the model parameters were assigned based on experimental results, the variations of Mises stress, displacement, friction stress, and adhesive stress of interface during shear process were analyzed.

Different compaction conditions (water content and density) may induce various soil structures. The influence of these structures on small strain shear stiffness G seems contradictory and is not understood (e.g., denser specimens may have larger or smaller G than looser specimens after compression). Furthermore, the influence of compaction condition on stiffness anisotropy remains unclear. This study investigated the evolution of structure and anisotropic stiffness of saturated and compacted loess during isotropic compression. Specimens compacted at different water contents and densities were explored. The measured G was normalised by a void ratio function (f (e)) to eliminate density effects. Before yielding, G/f (e) increases with decreasing compaction water content and increasing density. These two trends are reversed at large stresses (2 to 3 times yield stress), implying that an initially softer structure becomes stiffer. Based on mercury intrusion porosimetry, stereomicroscope, and scanning electron microscope results, the trend reversal is likely because interparticle contacts are more strengthened and pores are more compressed in the initially softer specimens. Furthermore, the stiffness anisotropy becomes more significant with decreasing compaction water content and increasing density because of more orientated fabrics, as evidenced by the particle/aggregate directional distribution results.

The fabric anisotropy in granular soils is a very important character in soil mechanics that may directly affect many geotechnical engineering properties. The principal objective of this study is to develop an efficient approach for assessing the degree of fabric anisotropy as a function of grading, particles shape and weighting specifications. By assuming cross-anisotropy, the anisotropic shear stiffness values of 1042 implemented tests on 200 various sandy and gravelly soil specimens from 43 different soil types were collected from the literature. Those were combined with their corresponding void ratios, stress conditions, grading parameters, particles shape and weighting attributes to generate a global database of anisotropic shear moduli in terms of testing conditions. The magnitudes of fabric anisotropy ratio were obtained using a well-known empirical equation, and they were plotted against the relevant soil grading and particles information to examine the dependency level of this ratio to the particularities. A series of multiple regression analyses were carried out to develop a global correlation for evaluating fabric anisotropy ratio in granular soils from grading, particles shape and weighting characteristic. The results showed that reliable quantities of fabric anisotropy ratio can be estimated using the surface appearance soil specifications. The findings may serve as an appropriate technique to yield good approximations for fabric and shear stiffness anisotropies using soil grading and particle properties.

In order to study the impact of surface roughness on the cyclic shear characteristics of the Soil-Rock Mixture and concrete interface, a series of cyclic shear tests were conducted using a large indoor direct shear apparatus. The effects of three concrete surface roughness coefficients JRC (0.4, 9.5, 16.7), five rock content levels (0%, 25%, 50%, 75%, 100%), and three cyclic shear displacement amplitudes (1, 3, 6 mm) on interface cyclic shear stress and Soil-Rock Mixture shear deformation were analyzed. A Bidirectional Long Short-Term Memory (BoBiLSTM) model was proposed, utilizing Bayesian optimization and k-fold cross-validation for hyperparameter tuning to streamline the model parameter selection process and enhance the prediction accuracy of the stress-strain relationship under cyclic loading. The experimental results show that, under five rock content levels, as the concrete surface roughness coefficient and cyclic shear displacement amplitude increase, the interface average peak shear stress increases accordingly. The interface average peak shear stress of the sample with 75% rock content is the highest; in terms of vertical displacement, the sample with 50% rock content has the maximum displacement, while the sample with 25% rock content has the minimum. The two types of samples show different soil deformation patterns in the two shear directions during the cyclic shearing process; as the shear displacement amplitude increases from 1 mm to 3 mm and 6 mm, the greater the concrete surface roughness, the smaller the change in shear stiffness and damping ratio. Compared to traditional Long Short-Term Memory (LSTM) models, the BoBiLSTM model demonstrated improvements in the average metrics of R2, RMSE, and MAPE by 0.32%, 57.25%, and 72.32%, respectively.