Gongcheng Kexue Yu Jishu/Advanced Engineering Science

The landslide—blocking river—dammed lake disaster chain is a typical disaster in mountain and gorge areas. The interaction between landslides and rivers, as the linking process in this disaster chain, often determines the scale of continuation of subsequent disasters in time and space. It is critical to reproduce the complex fluid-solid coupling process in the assessment of the landslide-blocking river disasters. By using the Drucker–Prager model to simulate the sliding of the landslide body, along with an artificial equation of state to model the flow of water, the two-phase double-point material point method (TPDP–MPM) program was developed. The simulation of the landslide-water body interaction process on an experimental scale was conducted and verified the reliability of TPDP–MPM in multiphase coupling problems. On this basis, a landslide—dammed lake disaster chain, namely the Pingdu River being blocked by the Erhuang Village landslide, Shiyan City, Hubei in 2011, was reproduced by the TPDP–MPM program. According to the TPDP–MPM simulation results, this disaster could be divided into four stages: unstable initiation, high-speed sliding, river braking, and accumulation. Based on the velocity field of the landslide mass and the evolution of the kinetic energy, the sliding pattern of the Erhuang Village landslide and the mechanism of river braking were revealed. The research results provided a theory and method for the whole process analysis and risk assessment of the landslide—dammed lake disaster chain

A circular carbon steel tube was added to the core concrete inside square stainless-steel tubes, the constraint effect coefficient to the core concrete and the ductility of the composite CFST sections could be improved remarkably, as a result, the anti-corrosion protection and fire protection of traditional steel structures could be reduced, furthermore, the disadvantageous influence of the demolished concrete lumps (DCLs) in new concrete could also be improved. Sixteen composite-sectioned concrete-filled stainless steel tubular (CS–CFSST) stub columns filled with high-strength demolished concrete lumps (DCLs) were tested. The influence of factors such as the replacement ratio of the demolished concrete lumps, confinement factor, and different concrete strength of new or demolished concrete on the mechanical performance was investigated. According to the testing results, the axial bearing capacity of the CS–CFSST stub columns decreased with the increase of the mix ratio of DCLs, the maximum decreasing amplitude of the bearing capacity was 6.8% with the replacement ratio of 20%. The failure modes of the CS–CFSST stub columns were quite familiar with those of normal CS–CFSST stub columns filled with normal fresh concrete. Two types of failure modes, drum-shaped and shear failure modes were found in the inner circular CFST stub columns, and excellent ductility and residual bearing capacity were also found as the load-deformation curves kept horizontally or ascended slowly, the residual bearing capacity of all the specimens were higher than 63% of its bearing capacity. The simplified method of constraint effect coefficient in the CS–CFSST stub column was proposed, and four calculation methods of the bearing capacity of the CS–CFSST stub columns infilling with DCLs were compared, coincident results were obtained using the calculation method considering the ultimate bearing capacity reduction of the circular CFST section

In order to solve three problems of traditional recommendation models, i.e., data sparsity, low robustness and the lack of deep-level semantics among heterogeneous features, a novel correlation visual adversarial Bayesian personalized ranking (CVABPR) recommendation model was proposed. First, based on the movie titles in the original MovieLens datasets, the corresponding movie posters were downloaded from Internet movie database (IMDB) to construct two multimodal datasets named MovieLens-100k-WMI and MovieLens-1M-WMI, respectively. Second, a group of heterogeneous but complementary image features were extracted using the SENet model to describe movie posters accurately. Then, the cluster canonical correlation analysis model was improved to mine the implicit cluster canonical correlation between the heterogeneous features. Afterwards, the correlation was used to optimize the visual Bayesian personalized ranking (VBPR) model to better depict the movies to be recommended. Finally, a perturbation factor was absorbed into the recommendation model to enhance the robustness of the CVABPR model through adversarial learning, making the recommendation model more stable and generating high-quality recommendation results. To verify the proposed CVABPR model, a set of experiments were carried out on two multimodal datasets. Evident performance improvements of the CVABPR model were observed on the two datasets. Specifically, a 3.802% performance improvement of the mean average precision (MAP) metric was obtained on the MovieLens-100k-WMI dataset, and a 4.609% performance improvement of the MAP metric was observed on the MovieLens-1M-WMI dataset. The mainstream baseline was defeated by the CVABPR model. Based on ablative analysis experiments, a more important role of the adversarial learning strategy was found compared with the cluster canonical correlation. Additionally, larger performance improvements were observed on the MovieLens-1M-WMI dataset with higher data sparsity. The key challenges of data sparsity and the lack of deep semantic among heterogeneous features were solved to a certain degree. Meanwhile, the CVABPR model has strong robustness

The bolted prefabricated concrete wall is widely used in construction due to its simple connection and rapid construction. The quasi-static tests of seven full-scale dry-type bolted prefabricated concrete walls (including four steel-plate-type bolted walls and three through-connection-type bolted walls) were conducted. The test results indicated that the bolted prefabricated wall would spread out at horizontal connection firstly and then the wall had rigid rotation, resulting in a lower lateral stiffness than the cast-in-place wall panel. Based on the elastic theoretical researches and finite numerical analyse, the horizontal deformation feature of the bolted prefabricated concrete wall was revealed, and the effect factors of lateral stiffness were analyzed. Moreover, the calculation formula of initial lateral stiffness of bolted prefabricated concrete wall was derived and its reliability was verified by numerical analyse and test results. In addition, through parameter analyse, a simplified calculation formula for initial lateral stiffness reducing the coefficient of bolted prefabricated concrete wall considering four effect factors (steel ratio of connection, height-width ratio of wall, axial compression ratio, and relative position of connection) was proposed. The error between the results by simplified formula and numerical analysis results is within 20%, indicating that the predicated results by the proposed simplified formula have high precision

The stability of coastal soft soil stratum is dramatically affected by external construction. To clarify the disturbance mechanism of the shallow shield tunnel engineering to the stratum, a refined numerical model for simulation of shield dynamic excavation was established based on the finite difference scheme, and the disturbance effect of every construction factor on the stratum was studied according to the modeling results. In the model, the combined effects of four factors on surrounding soil, including cutter head friction, excavation supporting force, grouting pressure, and shield friction, were considered. Referring to the behavior change of the grouting layer in the practical project, the relationship between pressure dissipation and slurry solidification during the shield tail grouting was divided into three stages to simulate the flowing state, the initial setting state and the final setting state of the slurry, respectively, and the corresponding parameters were sectionally assigned to the grouting layer of the newly-built tunnel and adjusted in real-time along with the excavation process. This model can realistically reflect the whole process of shield tunneling and realize the refined simulation of the construction. The numerical model was verified by comparison with the results from the field monitoring of the shield tunnel project with 1.0D (D is the outer diameter of the tunnel) burial depth in Xiamen Metro, and the influence of the above four construction factors on the soft soil stratum disturbance during the shallow excavation process was calculated and compared, and the effect of the factors on the disturbance deformation was summarized. The results show that the cutter head pushing causes the radial expansion of the forward soil. In the grouting process, a settlement groove is formed within 1.0D on both sides of the excavation space, and the near-field rebound and far-field expansion occur to the soil lateral displacements on two sides of the tunnel during the solidification of the grouting layer. The increase of the cutter head friction and shield friction will further aggravate the stratum disturbance deformation, and with the increase of the excavation supporting force and grouting pressure, the surface settlement slightly slows down and the stratum lateral displacement significantly increases. The results are of guiding significance for disturbance prediction and parameter selection in shallow shield excavation