Capsule networks can encode the properties and spatial relationships of skin cancer image features, thereby overcoming the disadvantage of information loss in the pooling process of convolutional neural networks. Aiming at the problem that only shallow features can be extracted and the convergence performance of the squash function in capsule networks, a ResNeXt cascaded with capsule networks was proposed for Rs-Capsnet networks. Firstly, the complex features of the image were learned using the ResNeXt network. The Inception module and the residual connection were used to extract the deep features, and the weights of the feature map were adjusted and delivered to the capsule module through the CBAM attention module. Then, an improved squash function capsule network was used to complete the classification. Finally, the improved network was compared with mainstream models. The results show that Rs-Capsnet exhibits better performance in skin cancer image classification.

The bridge health monitoring system based on sensor data acquisition has become standard for new bridge construction. However, this scenario presents challenges due to the massive volume of monitoring data that is difficult to store. Therefore, focusing on the time-series characteristics of bridge monitoring data, compression schemes were explored for bridge monitoring data. Differential compression was investigated based on the arithmetic progression properties of bridge monitoring timestamps and floating-point exclusive OR(XOR) compression based on the low frequency of changes in monitoring value data. Compared to the Gorilla time series database algorithm, the XOR compression method added control bits to avoid degradation of compression results. Experimental analysis reveals that both algorithms exhibit varying degrees of superiority over common compressors. The differential compression of timestamp sequences demonstrates superior compression rates compared to common compressors, achieving a compression rate of 0.015 6 for timestamp sequences that conform to arithmetic progression characteristics, approaching the compression limit value. Compression and decompression speeds are above average, and the method is insensitive to monitoring type. On the other hand, the XOR compression method performs well on datasets with low frequency of change, achieving compression rates of 0. 302 8 for bridge data and 0. 662 8 for non-bridge data, indicating sensitivity of the XOR compression method to monitoring type. In practical applications of bridge monitoring, suitable compression storage schemes can be selected based on the characteristics of the bridge monitoring dataset.

low-temperature fracture is the normal distress of the thin-layer overlay asphalt mixture. To reveal the low-temperature cracking behavior, semicircular bending tests combined with crack observation, digital image correlation and a finite element numerical simulation based on the meso-structure cohesive zone model were carried out. The applicability of the model was verified by the load-displacement curve and crack paths. The results show that the low-temperature cracking behavior of asphalt mixture can be well demonstrated by digital image processing. The simulation of asphalt mixture meso-structure is suitable for analyzing the cracking behavior. Furthermore, the maximum tensile stress and neutral axis positions on the mid-span section are correlated with the properties of the materials.

To solve the problems of multiple influencing factors, different focuses of participating parties, and inability to quantitatively describe the comparison and selection schemes in the current process of selecting routes for mountainous expressways, based on thorough research on the influencing factors of expressway and railway route selection at home and abroad, the influencing factors of expressway route selection in mountainous areas of China were systematically analyzed and summarized, and four aspects : economy, technology, safety and environment were summarized, totaling 16 specific influencing factors. Based on the Xihe to Tanchang Expressway project, a group of 164 experts from the local government, industry regulatory departments, owners, design, construction, supervision, third-party testing units, and research institutes in the project area were organized to form an expert group. The analytic hierarchy process (AHP) was used to evaluate the importance of 16 route selection influencing factors. The results show that the comprehensive weight average score of the three influencing factors in the environmental indicators is high, reflecting the further strengthening of environmental awareness in the field of engineering construction. To further establish a optimization model for scheme comparison, taking the scheme comparison of the route from the project Dengta Village to Haolin Village as an example, the technique for order preference by similarity to idea solution (TOPSIS) method based on the AHP was used to compare and select schemes. The superiority of the four comparison schemes is 2.84, 0.56, 0.72, and 2. 15, respectively, indicating that scheme 1 is optimal. Based on a comprehensive analysis of the factors affecting the selection of mountainous expressway routes, the research results adopt the AHP-TOPSIS comprehensive evaluation index system model. While fully utilizing expert experience, it can objectively and scientifically

In order to reduce the vibration of fracturing branch pipe during fracturing operation, the fluid-structure coupling analysis method was used to carry out the modal analysis and harmonious response analysis of fracturing branch pipe, considering the impact of fracturing pump vibration and high-pressure fracturing fluid on fracturing branch pipe. The influence of angle of bend and number of fracturing branch pipe supports on fracturing branch pipe vibration was studied. The results show that under the current layout, the main vibration positions of the first six modal modes of the manifold appear at double elbows 2, 3, and 4, indicating that the main vibration deformation of the fracturing branch pipe occurs at its lower end, that is, at the double elbows far away from the fracturing truck. Considering the weak points at double elbows 2, 3, and 4, the smaller displacement response amplitudes of each double elbow at each connection angle are obtained based on the maximum displacement response amplitude. Indicating that each connection angle is the best choice for the actual fracturing operation. When four elastic supports are used, the maximum amplitude is much smaller than that of two or three elastic supports, indicating that adding elastic supports at the lower end of the fracturing branch pipe can weaken the vibration amplitude, and the optimal scheme can be obtained under the specific working range. The research results can provide theoretical guidance for the vibration characteristics and vibration reduction of fracturing branch pipe.

To explore the process and evolution of large-scale flight delay propagation, and avoid previous research mainly focusing on observing real data and the distribution of delay propagation networks. Drawing inspiration from the classic susceptible-explored-infected-recovered(SEIR) model and taking into account the impact of node closure on large-scale flight delays. Based on this, the state of airport nodes was added to five categories, and a large-scale flight delay propagation model based on susceptible-explored-infected-death-recovered (SEIDR) was constructed and applied to air traffic networks. Using a combination of phase trajectory analysis and related parameter analysis, the propagation threshold and propagation law of large-scale flight delays were obtained, and the influence of propagation parameters between airport nodes on the propagation law of large-scale flight delays was further analyzed. Finally, a large-scale flight delay in 2022 was taken as an example for analysis and verification. The results show that the established model can more accurately describe the evolution process and propagation law of large-scale flight delays.

Reinforcement and water sealing effect difficult to achieve by conventional stratum grouting method in strong seepage sandy soil stratum. In order to study the mechanical characteristics of freezing-grouting combination in water-rich sand stratum, the stress-strain relationship and its influencing factors of artificially frozen cement sand were studied by triaxial tests, and the variation law and strength mechanism of stress-strain relationship of samples under different freezing temperature, curing age and confining pressure were discussed. The results show that the stress-strain curve of frozen cement-sand has a certain strain-hardened nonlinear ductility stage including compaction stage linear elasticity stage nonlinear ductility stage and strain-hardening stage. The non-linear increase of cohesion and internal angle of frozen cement-sand due to the curing age of freezing temperature increases the shear strength of frozen cement-sand, and the non-linear ductility phase strain ratio enhances the brittleness of frozen cement-sand. The increase of confining pressure increases the shear strength of frozen cement-sand, and the proportion of strain in nonlinear ductility stage improves the ductility of frozen cement-sand. Based on Mohr-Coulomb strength criterion, a non-linear strength prediction model of frozen cement sand was established, which considered the influence of freezing temperature and curing age. The error between the predicted results and the measured values is less than 5% The research results can provide parameter support for the fine design of freezing-grouting combined reinforcement scheme for water-rich sandy soil stratum.

Considering the technical issues related to the construction of extremely small radius and ultra large diameter shield tunnels in stratified strata, taking the shield tunnel project of the Pazhou Station to Beigang Park Station section of the Pearl River Delta intercity rail transit in Guangzhou as the research object, key construction technologies of shield excavation were elaborated in detail. Solutions to outward deviation of the excavation axis because of the small curve radius and strata disturbance caused by the excavation were provided. Built upon the analysis of on-site real-time monitoring data, variation patterns of vertical and horizontal displacements, pore water pressure and soil pressure during the construction of the small radius shield tunnel were revealed. Monitoring data shows that the shield excavation of the turning section in rock layers has little impact on surface settlements. The influence of shield tunneling on the surrounding soil decreases with the increasing distance, the maximum horizontal displacements occur in the area above the tunnel. There is a high similarity in the changing trend of pore pressure and soil pressure and the impact on the soil outside the turning is greater when the shield turns.

In practical production processes, equipment performance gradually degrades over time, leading to extended processing durations. To address this issue, an improved nondominated sorting genetic algorithm II (NSGA-II) was proposed for a re-entrant hybrid flow-shop scheduling problem that considers deterioration effects. Firstly, a mathematical model was formulated with the optimization objectives of minimizing makespan and reducing processing energy consumption. Secondly, a job-sequence-based encoding method was employed, and an energy-efficient scheduling decoding method that accounts for deterioration effects was designed. Additionally, to enhance population diversity, various mutation operators were introduced, and algorithm parameters were adaptively adjusted to prevent convergence to local optima. A variable neighborhood search strategy was also integrated to reinforce the local search capability of the algorithm. Finally, comparative experiments with other algorithms on ten different scale test instances demonstrated that the proposed algorithm delivers superior solution quality, along with better diversity and convergence properties.

To explore the impact of the spatial distribution of material sources within a watershed on the susceptibility to debris flows. The nearest neighbor index was adopted, based on the principles of mathematical statistics, to quantify the spatial clustering of material sources. Using 2243 small watersheds as evaluation units, the longitudinal gradient, area-elevation integral, topographic wetness index, peak ground acceleration of earthquakes, and rock hardness were taken as disaster-prone indicators, and the aggregation index of material sources, connectivity index, and material reserves were taken as the core material source indicators. The LightGBM model was relied on to investigate the susceptibility to debris flows in the Shigu-Gangtuo section of the upper reaches of the Jinsha River. The research process calculated the index system of material source factors and the index system without material source factors. Both results indicate that the high and very high susceptibility areas are mainly concentrated in the Benzilan-Batang section. The receiver operating characteristic curve (ROC) curve analysis shows that after incorporating the material source characteristic indicators, the area under the curve (AUC) value increases by 6% compared to the AUC value without material source characteristics, indicating that the model performs well and has high predictive accuracy after the inclusion of material source indicators. It also proves that the material source characteristic indicators are highly correlated with the probability of debris flow occurrence.