Moisture intrusion into the subgrade can significantly increase its moisture content, leading to a decrease in stiffness and strength, thereby compromising the serviceability performance of the pavement. Electro-osmosis has been used as an effective method for reducing moisture content and improving subgrade mechanical properties. However, its impact on mechanical properties has not been well understood. This study evaluated the mechanical behavior of electro-osmosis-treated subgrade soil through laboratory experiments that included bender element and cyclic triaxial tests. The study analyzed the effects of supply voltage and soil compaction degree on electro-osmosis treatment. The results showed that after treatment, the shear wave velocity increased by 26.0 to 59.2%, and the dynamic resilient modulus improved by a factor of three. Increasing the supply voltage and degree of compaction was found to lead to more significant improvements. Further analysis revealed that the reduction in moisture content alone was insufficient to contribute to the improvement. Cementation of colloids generated by the electrochemical reaction between soil particles also contributed to the improvement. It is worth noting that the nonuniform distribution of moisture and colloid in electro-osmosis-treated soils resulted in heterogeneity, with soil close to the anode being the weakest in terms of mechanical strength. Chemical injection or polarity reversal was suggested to enhance the uniformity of distribution and improve the overall treatment effectiveness. Overall, the study highlights the potential of electro-osmosis as a viable method for improving the mechanical properties of subgrade soil, but further research is required to investigate the heterogeneity of the distribution of moisture and colloid.

The dynamic resilient modulus (MR) of a subgrade soil is a fundamental parameter for evaluating the dynamic stability and service resilience of subgrade fillers and structures, as well as an instrumental input for calculating the mechanical response and fatigue life of a pavement structure. To accurately and reasonably characterise the MR of subgrade soils, machine learning (ML) models were established using the support vector machine, random forest, and extreme gradient boosting algorithms based on a large-scale dataset including 3533 records of MR tests conducted on subgrade soils. Meanwhile, the weighted plasticity index (WPI), initial moisture content (w), dry unit weight (gamma d), confining stress (sigma c), deviator stress (sigma d), and numbers of freeze-thaw cycles (NFT) were set as the input variables to predict the MR using ML models, which considered the effects of wheel loads, physical properties and climate fluctuation on the subgrade soils during the service period. Subsequently, the Shapley additive explanations method was developed to explain the prediction model for the MR of subgrade soils based on ML algorithms. The results quantitatively illustrated the explicit mapping relationship and internal influencing mechanism between the significant features of the influences and MR of subgrade soils, which was consistent with prior experimental and physical cognition. In summary, the study findings provide meaningful guidelines for the structural design and life evaluation of pavement subgrade engineering.

The dynamic resilience characteristics of aeolian sand subgrade are influenced by salt content and water content, exhibiting significant stress dependence and anisotropy. The resilient modulus(MR) M R ) of aeolian sand represents the stress-strain nonlinearity under cyclic loading, serving as an important parameter for the design of aeolian sand subgrade in desert areas. In order to investigate the variation of M R of aeolian sand subgrade with salt content and water content under traffic loading, as well as the M R characteristics under these conditions, three types of aeolian sand samples with varying water content and four sulphate contents were prepared. The variation of M R of aeolian sand under different confining pressures and deviator stress levels, as well as the influences of water content and salt content, was studied through indoor dynamic triaxial testing. Based on the pattern of the fitting parameters of the benchmark model, a prediction model suitable for the M R of aeolian sand was constructed. The results indicate a rise in aeolian sand's M R with increasing deviator stress and confining pressure, with confining pressure having a more significant impact than deviator stress. With the increase in water content, the M R of aeolian sand decreases nonlinearly, and with the increase in salt content, it exhibits a wave-shaped trend of increasing-decreasing-increasing, which is related to the dissolution state of sodium sulfate in the soil. Based on the experimental results, a prediction model of the M R of aeolian sand was established, derived from the benchmark model, which can reflect the influence of salt content and water content on the M R , introducing them as variables within the model.

Understanding the dynamic resilient modulus (MR) of a recycled carbonaceous mudstone soil-rock mixture (CMSRM) embankment under wet-dry cycles can provide a basis for CMSRM embankment design and evaluation. The effects of the stress states, rock contents, and wet-dry cycles on MR were analyzed by dynamic triaxial tests, and the prediction model for MR of CMSRM was studied also. The results show that the MR of CMSRM is negatively correlated with the deviation stress and positively correlated with the minimum bulk stress, with significant stress-dependent characteristics. With the increase in rock content, the MR of CMSRM increases at first and decreases later. Grey correlation analysis showed that the MR of the CMSRM is affected by the rock content, minimum bulk stress, wet-dry cycles, and deviator stress in order of priority. Considering the comprehensive effects of stress states, wet-dry cycles and particle gradation, a model predicting the MR was proposed. The research can provide useful information for the application of carbonaceous mudstone embankment in humid-heat regions.

Freeze -thaw (FT) cycles are regarded as a damage effect to the dynamic resilient modulus (MR) of subgrade soil in seasonal frozen regions and improving the resistance on freeze -thaw cycles must be concerned urgently. For this purpose, the bentonite superabsorbent polymer (BT -SAP) was developed as an available admixture to relieve the attenuation of MR of subgrade soil under FT cycles. In order to evaluate the improvement of resistance on FT cycles, a series of dynamic triaxial tests were conducted after various FT cycles. The main influences on MR were investigated including the numbers of freeze -thaw cycles, BT -SAP contents, deviator stress and confining stress. The experimental results indicate that the BT -SAP can delay the attenuation trend of MR that means the resistance on freeze -thaw cycles have been promoted. The ensuing discoveries are that with the increment of BT -SAP content, the sensitivity of the subgrade soil to FT cycles is gradually reduced. Nevertheless, results reveal that a proper content of BT -SAP is recommended as 0.5%, where if the BT -SAP content exceeds 0.5%, the improvement effect of resistance on FT cycles decreases. Moreover, the improved effect caused by BT -SAP can produce a coupling enhancement combining with confining stress, and it also can offset the softening effect due to the deviator stress. For better application and practicability, the high -accuracy empirical model for predicting MR is established and validated. Finally, based on the soil freezing temperature, unfrozen moisture content and moisture distribution characteristics, the mechanism of BT -SAP reducing the freeze -thaw cycle sensitivity is analyzed. The conclusions mentioned above show that employing BT -SAP material is acceptable and significant to improve resistance on freeze -thaw of subgrade soil.