Chaoyang Wu

To investigate the slope instability caused by differential frost heaving mechanisms from the slope crest to the toe during frost heave processes, this study takes a typical silty clay slope in Xinjiang, China, as the research object. Through indoor triaxial consolidated undrained shear tests, eight sets of natural and frost-heaved specimens were prepared under confining pressure conditions ranging from 100 to 400 kPa. The geotechnical parameters of the soil in both natural and frost-heaved states were obtained, and a spatiotemporal thermo-hydro-mechanical coupled numerical model was established to reveal the dynamic evolution law of anchor rod axial forces and the frost heave response mechanism between the frame and slope soil. The analytical results indicate that (1) the frost heave process is influenced by slope boundaries, resulting in distinct spatial variations in the temperature field response across the slope surface—namely pronounced responses at the crest and toe but a weaker response in the mid-slope. (2) Under the coupled drive of the water potential gradient and gravitational potential gradient, the ice content in the toe area increases significantly, and the horizontal frost heave force exhibits exponential growth, reaching its peak value of 92 kPa at the toe in February. (3) During soil freezing, the reverse stress field generated by soil arching shows consistent temporal variation trends with the temperature field. Along the height of the soil arch, the intensity of the reverse frost heave force field displays a nonlinear distribution characteristic of initial strengthening followed by attenuation. (4) By analyzing the changes in anchor rod axial forces during frost heaving, it was found that axial forces during the frost heave period are approximately 1.3 times those under natural conditions, confirming the frost heave period as the most critical condition for frame anchor design. Furthermore, through comparative analysis with 12 months of on-site anchor rod axial force monitoring data, the reliability and accuracy of the numerical simulation model were validated. These research outcomes provide a theoretical basis for the design of frame anchor support systems in seasonally frozen regions.

This study developed a dynamic model for the Traffic Speed Deflectometer (TSD) on inverted asphalt pavement structures. It is aimed at evaluating the deflection slope and quantifying measurement errors. First, the reliability of the ABAQUS model in simulating the dynamic response of asphalt pavements was validated by comparing with previous studies. The deflection slope curves of inverted and semi-rigid base pavements with varying thicknesses were compared, revealing that the inverted pavement exhibited complex deflection slope trends in TSD measurements. A significant decrease in peak deflection was observed at 0.15 m from the load gap center with increasing surface thickness. The deflection velocity measurement value of the TSD calibration sensor (S3500) on the inverted asphalt pavement is not zero, which causes the road surface deflection to be lower than the actual deflection, with an error as high as 80.1%, which overestimated the pavement’s structural capacity. These findings suggest that the sensor configuration and measurement strategy of TSD should be reconsidered when applied to inverted asphalt pavement structures. The results provide useful insights that may support the refinement of TSD application strategies.

Permeability characterized by permeability coefficient is one of the important engineering properties of clay, and its permeation mechanism depends on the permeation process of water in clay. To clarify the characteristics and influencing factors of the anisotropy of stress in red clay, indoor variable head permeability tests were conducted to obtain the permeability coefficients of undisturbed red clay in vertical and horizontal directions under different overlying loads. The research results indicate that under the same overburden load, the vertical permeability coefficient (k_V) is greater than the horizontal permeability coefficient (k_H), and on semi-logarithmic coordinates, both permeability coefficients exhibit a linear negative growth relationship with the overburden load. Moreover, with a gradual increase in the overburden load, the permeability coefficient ratio kV/kH shows a decreasing trend. When the overburden load does not exceed 150 kPa, the ratio of permeability coefficients (k_V/k_H) shows a linear decreasing trend, and when the overburden load exceeds 150 kPa, the rate of decrease gradually slows and eventually stabilizes. It is also found that the void ratio (e) is linearly related to lgk_V and lgk_H, and based on this, a prediction model for the anisotropy of stress in undisturbed red clay based on permeability coefficients is established. Under increasing osmotic pressure, the permeability coefficients in both directions exhibit linear growth. Essentially, it is the osmotic pressure that alters the stress pathways of red clay, and the smaller the overburden load, the greater the effect of osmotic pressure. With the increase in the overburden load, the anisotropy of stress gradually weakens.

To accurately identify the deflection data collected by a traffic speed deflectometer (TSD) and eliminate the noise in the measured signals, a TSD signal denoising method based on the partial swarm optimization–variational mode decomposition (PSO–VMD) method is proposed. Initially, the VMD algorithm is used for modal decomposition, calculating the correlation coefficients between each decomposed mode and the original signal for modal selection and signal reconstruction; Then, the particle swarm optimization algorithm is utilized to optimize the number of modes K and the value α for the VMD algorithm, adopting fuzzy entropy as the affinity function to circumvent effects from sequence decomposition and forecasting accuracy, thus identifying the optimal combination of hyperparameters. Finally, the analysis on simulated signals indicates that the PSO–VMD method secures the best parameters, showing a clear advantage in denoising. Denoising real TSD data validates that the approach proposed herein achieves commendable outcomes in TSD deflection noise reduction, offering a feasible strategy for TSD signal denoising.

The Anton Paar MCR 52 intelligent high-speed rheometer was utilized in this paper to conduct rheological tests on the Jiangjiagou debris flow slurry via linear loading. The relationship curves between shear rate and shear stress at different solid volume concentrations were obtained, and predictive formulas for yield stress and viscosity coefficient variations with the shear rate are proposed. Furthermore, the evolution mechanisms of yield stress and viscosity coefficient with changes in solid volume concentration of the debris flow were thoroughly analyzed. The results indicated that the rheological curves of the Jiangjiagou debris flow conformed to the Bingham fluid model. An increase in solid volume concentration continuously promoted the upward trend of the rheological curves, with significant increments observed only at high volume concentrations. A predictive model for debris flow rheological parameters was established based on the linear relationship between the rheological parameters and the solid volume concentration. The rise in solid volume concentration inhibited turbulence in the debris flow, while clay minerals enhanced the debris flow slurry’s ability to capture pore water. Significant shear-thinning effects were observed within the debris flow, which were particularly more pronounced at lower shear rates. The study outcomes hold crucial engineering significance for a better understanding of debris flow rheological properties, the calculation of debris flow dynamic parameters, and disaster prevention and control.

With the powerful discriminative capabilities of convolutional neural networks, change detection has achieved significant success. However, current methods either ignore the spatiotemporal dependencies between dual-temporal images or suffer from decreased accuracy due to registration errors. Addressing these challenges, this paper proposes a method for remote sensing image change detection based on the cross-mixing attention network. To minimize the impact of registration errors on change detection results, a feature alignment module (FAM) is specifically developed in this study. The FAM performs spatial transformations on dual-temporal feature maps, achieving the precise spatial alignment of feature pairs and reducing false positive rates in change detection. Additionally, to fully exploit the spatiotemporal relationships between dual-temporal images, a cross-mixing attention module (CMAM) is utilized to extract global channel information, enhancing feature selection capabilities. Furthermore, attentional maps are created to guide the up-sampling process, optimizing feature information. Comprehensive experiments conducted on the LEVIR-CD and SYSU-CD change detection datasets demonstrate that the proposed model achieves F1 scores of 91.06% and 81.88%, respectively, outperforming other comparative models. In conclusion, the proposed model maintains good performance on two datasets and, thus, has good applicability in various change detection tasks.

With the increase of spatial resolution of remote sensing images, features of feature imaging become more and more complex, and the change detection methods based on techniques such as texture representation and local semantics are difficult to meet the demand. Most change detection methods usually focus on extracting semantic features and ignore the importance of high-resolution shallow information and fine-grained features, which often lead to uncertainty in edge detection and small target detection. For single-input networks when two temporal images are connected, the shallow layer of the network cannot provide the information of the individual original image to the deep layer features to help reconstruct the image, and therefore, the change detection results may be missing in detail and feature compactness. For this purpose, a twins context aggregation network (TCANet) is proposed to perform change detection on remote sensing images. In order to reduce the loss of spatial accuracy of remote sensing images and maintain high-resolution representation, we introduce HRNet as our backbone network to initially extract the features of interest. Our proposed context aggregation module (CAM) can amplify the convolutional neural network receptive field to obtain more detailed contextual information without significantly increasing the computational effort. The side output embedding module (SOEM) is proposed to improve the accuracy of small volume target change detection as well as to shorten the training process and speed up the detection while ensuring the performance. The method has experimented on the publicly available CDD dataset, the SYSU-CD dataset, and a challenging DSIFN dataset. With significant improvements in precision, recall, F1 score, and overall accuracy, the method outperforms the five methods mentioned in the literature.

This study developed two-and-half dimensional (2.5-D) finite element method (FEM) to predict viscoelastic pavement responses under moving loads and nonuniform tire contact stresses. The accuracy of 2.5-D FEM was validated with two analytical solutions for elastic and viscoelastic conditions. Compared to three-dimensional (3-D) FEM, the computational efficiency of the 2.5-D method was greatly improved. The effects of loading pattern and speed on pavement surface deflection and strain responses were analyzed for asphalt pavements with four different asphalt layer thicknesses. The analyzed pavement responses included surface deflections, maximum tensile strains in the asphalt layer, and maximum compressive strains on top of subgrade. The loading patterns have influence on the mechanical responses. According to the equivalent rule, the point load, rectangle type, and sinusoid-shape contact stresses were studied. It was found that the point load caused much greater pavement responses than that of the area-based loading. When the tire loading was simplified as uniform contact stress in rectangular area, the maximum tensile strains in the asphalt layer varied with the width/length ratio of contact area. Additionally, it was shown that the dynamic responses of pavement structure induced by the sinusoid-shape contact stresses and realistic nonuniform stresses were quite similar to each other in all the cases. The pavement strain responses decreased as the speed increased due to viscoelastic behavior of asphalt layer. The study results indicate that asphalt pavement responses under moving load can be calculated using the proposed 2.5-D FEM in a fast manner for mechanistic-empirical pavement design and analysis.