Abstract:To study the stability performance of rectangular tubular flange composite beams with torsional bracing set at the mid-span under concentrated loads, the "plate-beam theory" was adopted, and the elastic lateral-torsional buckling total potential energy equation of the rectangular tubular flange composite beam was established. The displacement functions were selected, and the energy variation method was used to solve the buckling equation, and the analytical solution for the elastic lateral-torsional buckling critical moment of the rectangular tubular flange composite beam was obtained. Then, using the 1stOpt soft ware, considering the influence of multiple parameters, the elastic critical moments of the composite beams under different cross-sectional dimensions, spans, and torsional bracing stiffness were fitted, and the calculation formulas for the elastic critical moments of the composite beams with and without torsional bracing were obtained. These formulas were compared with the finite element analysis results, and the error was within 5%, verifying the accuracy of the calculation formulas. Finally, the influence laws of parameters such as concrete strength, web height-thickness ratio, span-height ratio, and upper flange steel content ratio on the stability performance of the rectangular tubular flange composite beam were analyzed. The study found that as the torsional bracing stiffness increased, the elastic lateral-torsional buckling critical moment of the composite beam gradually increased, and the stability of the beam was improved; when the torsional bracing stiffness reached the threshold stiffness, the elastic lateral-torsional buckling critical moment of the composite beam no longer increased; changing the concrete strength or web height-thickness ratio would not cause significant changes in the elastic lateral-torsional buckling critical moment of the beam, while reducing the span-height ratio or increasing the upper flange steel content ratio would significantly increase the elastic lateral-torsional buckling critical moment of this type of composite beam.

Abstract:Crack detection is one of the important aspects of structural health monitoring. To achieve qualitative analysis of cracks in steel beams in complex backgrounds, a two-stage detection method based on improved Convolutional Neural Network (CNN) and digital image processing is proposed for crack damage location and crack image segmentation. The first stage uses a multi-scale convolutional neural network to identify crack images in complex backgrounds. This network consists of a multi-scale convolution module Inception and a residual module. The multi-scale convolution module Inception contains three different-sized convolution kernels (1×1, 3×3, 5×5) for multi-scale feature extraction of the image. In the residual module, convolution layers and nonlinear activation functions are introduced to enhance cross-layer fusion ability and extract deeper features. The Grad-CAM visualization analysis highlights the prediction basis of the multi-scale convolutional neural network, proving its classification performance and discrimination basis. In the second stage, for the identified crack images, a combined process of image filtering denoising, threshold segmentation to separate crack pixels, and morphological processing to optimize the segmentation result is proposed for pixel-level segmentation and extraction of cracks. The pixel marking results manually annotated are used as the true labels to evaluate the recognition effect of image segmentation. The training results on the dataset show that the multi-scale convolutional neural network has an identification accuracy of 98.8% for steel beam crack images. The proposed image processing combination process has a maximum intersection-over-union (IOU) of 0.819, which can better classify and extract cracks.

Abstract:The pipe curtain pre-assembly technology, as a new type of underground shallow-buried excavation method, has gained increasing attention and development due to its excellent safety and practicality. However, there is a lack of comprehensive and systematic research on its mechanical properties, especially in the aspect of bending performance. To further explore the mechanical properties of the pipe curtain pre-assembly structure, two bending components were designed for two-point loading tests. By treating the welded reinforcing bars in the specimen as steel supports between the precast structural steel plates of the steel tube and setting the steel struts as the parameter, the bending performance of the pipe curtain pre-assembly structure was tested and studied. In order to fully investigate the influence of the steel supports on the bending performance of the pipe curtain pre-assembly structure, the finite element software was used to establish the finite element model of the pipe curtain pre-assembly structure components. By comparing with the test results, the correctness of the finite element model was verified. Based on this model, the influence of the diameter and position of the steel struts on the bending performance of the pipe curtain pre-assembly structure was analyzed. The research results show that the firm welding between the steel struts and the steel curtain can effectively prevent the separation of the steel plate from the concrete, improving the mechanical performance of the test specimen; the diameter of the steel struts has a significant impact on the bending bearing capacity of the test specimen; the position of the steel struts has a significant impact on the bending bearing capacity of the test specimen. In actual engineering, the steel supports should be set within the range of equal sections and at positions close to the sudden change in the section height.

Abstract:The masonry structure has significant characteristics such as poor tensile, shear, flexural and seismic performance. To improve the seismic performance of the masonry structure, the experimental team used mechanism sand high toughness concrete (hereinafter referred to as MSHDC) as a special material for structural reinforcement. This material has remarkable properties such as high ductility and high strength. Through the method of pressing and coating a surface layer on the surface of the masonry specimens for reinforcement, the bearing capacity and seismic performance of the reinforced masonry structure were improved. This experiment produced four masonry walls. The wall material was ordinary concrete bricks+cement mortar. The blank test group and the test condition group were respectively un-reinforced, 35mm thick reinforced mesh cement mortar single-sided reinforcement, and 15mm thick MSHDC single-sided and double-sided reinforcement methods were used to reinforce the brick walls made at the same time. Through the low-cycle repetitive load test, the effect of the MSHDC surface layer on the seismic reinforcement of the masonry structure was explored. The results showed that the unreinforced wall exhibited a non-signaled, poor ductility brittle shear failure under low-cycle repetitive load, while the wall reinforced with the MSHDC surface layer exhibited a signaled, better ductility shear-compression failure. At the same time, the MSHDC surface layer reinforcement could effectively enhance the compressive bearing capacity, deformation capacity, stiffness, ductility and energy dissipation capacity of the wall. Compared with the un-reinforced specimens, the ultimate load and ductility of the MSHDC double-sided reinforced specimens increased by 44% and 49.4% respectively, and the enhancement effect was the strongest. In summary, the MSHDC double-sided reinforcement method had the strongest effect on enhancing the seismic performance of the wall, while the MSHDC single-sided reinforcement method was the weakest.

Abstract:To explore the influence laws of different mix ratios on the physical and mechanical properties of phosphogypsum-based foamed building gypsum, in order to improve the resource utilization rate of phosphogypsum, the experiment adopted single-factor experimental analysis and microscopic morpho-logy detection methods to systematically study the physical and mechanical properties, thermal properties and pore structure of phosphogypsum-based foamed building gypsum under three different mix ratios. The microscopic morphology of phosphogypsum-based foamed building gypsum under different mix ratios and the interaction mechanism between each component were deeply observed and analyzed using scanning electron microscopy (SEM), revealing the intrinsic relationship between the microstructure and macroscopic performance changes. The test results showed that the apparent density and mechanical properties of phosphogypsum-based foamed building gypsum decreased with the increase of foaming admixture content; the optimal mix ratio was: phosphogypsum content of 60%, cement content of 20%, slag powder content of 10%, gypsum whisker content of 10%, retarder content of 0.1% of the cementitious material, water-to-solid ratio of 0.5. Correspondingly, the optimal sodium dodecyl sulfate (K12) foaming agent content was 12.5%, and the phosphogypsum-based foamed building gypsum prepared with this foaming agent reached the optimal level of absolute dry strength; SEM tests indicated that gypsum whiskers filled the pores between crystal connections, increasing the density of the test block, and gypsum whiskers formed connections at the micro-fissures of the test block, preventing the generation and expansion of micro-cracks, improving the internal pore structure of the test block, and enhancing the comprehensive performance of phosphogypsum-based foamed building gypsum.

Abstract:Taking red clay as the research object and polyvinyl alcohol as the improved material, the effects of different polyrinyl alcohol (PVA) concentrations and curing ages on the strength of red clay were studied by direct shear and non-lateral compressive strength tests, and the microstructure of the improved soil was observed by scanning electron microscopy. The results show that under the same conditions of each curing age, the cohesion and unconfined compressive strength of red clay increase first and then decrease with the increase of PVA concentration. The internal friction angle does not show obvious variation with the increase of PVA concentration, but it is greater than the internal friction angle of red clay. The effect of improving the shear strength and unconfined compressive strength of red clay with 3% PVA is ideal. Microscopic tests show that PVA reacts with soil particles to form a three-dimensional mesh structure with high elasticity and high strength, which is lapped between pores. The network structure formed by low concentration of PVA is incomplete and the cohesive force is weak. With the increase of concentration, the network structure is gradually complete, which can effectively limit the movement of soil particles, so as to improve the shear and compressive strength of soil. However, the higher concentration of PVA leads to the increase of the viscosity of the hydrogel, which affects its uniform mixing with red clay, which is not conducive to the contact between PVA and soil particles, wea-kens the connection between the network structure and soil particles, and reduces the cohesion and unconfined compressive strength.

Abstract:To systematically study the aerodynamic characteristics of freight vehicles on high-pier bri-dges in mountainous areas under crosswind conditions, a numerical simulation method was developed to reveal the variation patterns of aerodynamic parameters under different crosswind intensities, bridge positions, and types of freight vehicles. The Catia software was used to complete the detailed three-dimensional modeling of the freight vehicle, restoring key structural features such as vehicle body size, cargo compartment shape, and center of gravity distribution. The Global Mapper software was used to import the terrain elevation data of the area where the high-pier bridges are located. Based on the bridge design drawings, a terrain-bridge model including the bridge piers, the bridge deck, and the surrounding mountains was developed. Finally, a "vehicle-topography-bridge" coupled calculation model was integrated, ensuring that the model can accurately reflect the spatial positional relationship between the complex terrain in mountainous areas and the bridges and vehicles. The results showed that in the crosswind environment, when the freight vehicle was stationary on the mountainous high-pier bridge, its safety was mainly affected by lateral forces, side-tipping moments, and yawing moments, so the safety of the freight vehicle was relatively low at a distance of 5 and 200 meters from the bridgehead. At the same position on the bridge, the aerodynamic load of the freight vehicle increased with the increase of wind speed.

Abstract:To evaluate the fatigue performance of concrete curved bridges under the influence of multiple working conditions, vibration equations for the vehicle and bridge subsystems are established. The coupling of the vehicle and bridge vibration equations was established by using the displacement coordination conditions and mechanical equilibrium conditions at the contact points of the model. The modal analysis of the finite element model was used to compare the first-order frequencies, second-order frequencies, and the mid-span deflection displacements calculated by the coupling model with the mea-sured data, to verify the reliability of the model. Considering temperature, vehicle speed, bridge surface irregularity and other working conditions, the dynamic response analysis of the concrete curved bridge under vehicle load was carried out based on the finite element analysis software ANSYS, and the fatigue hazard points of the model were determined, and the equivalent stress data under different working conditions were obtained. The stress data were statistically processed using the rainflow counting method, and the initial crack length, crack type, critical crack length, and calculation method for the crack propagation life of the concrete curved bridge based on the fracture mechanics theory were determined. The study solved the problems of the vehicle-bridge coupled vibration response analysis method for fatigue analysis, the fatigue remaining service life assessment method for curved bridges, and the influence of different working conditions on the fatigue remaining service life. The results show that the crack propagation presents an exponential growth characteristic of initial slow and later accelerated; the vehicle driving position and temperature gradient have a significant impact on the fatigue life, while the change in bridge surface irregularity level has a relatively smaller impact.

Abstract:The analytical prediction method for ground settlement deformation of underground tunnels considering the surface building (structure) load-induced ground settlement due to tunnel excavation takes the structural load as an equivalent distributed load. It establishes analytical models for the deformation field of surrounding rock under the action of half-space structural load and the buried tunnel under the condition of no load in the half-space. Based on the calculation results of the two models, the ground settlement deformation under the condition of structure-underpass tunnels is predicted. For the deformation field of the surrounding rock under the action of the half-space structural load, the integral of the uniformly distributed force applied to the surface is used to obtain the stress distribution formula of the half-space, and then the displacement field distribution caused by the building load in the half-space is calculated. For the buried tunnel under the condition of no load in the half-space, the Airy stress function is used to obtain the displacement field expression of the tunnel excavation in the semi-infinite space based on the simplified form of the shallow-buried tunnel problem. On the basis of the two solutions, the analytical solution for the ground settlement deformation caused by the underpass tunnel under the influence of surface building loads is obtained. Through this analytical prediction, the ground settlement characteristics under the combined action of structural load and tunnel excavation are analyzed, and the settlement curve is drawn to analyze the surface settlement pattern. The calculated settlement deformation and the measured settlement data are compared, and the result shows that the relative error between the calculated value and the measured value is 3.9%, which proves that the calculation model can well describe the coupling effect of building load and tunnel excavation, and verifies the accuracy and applicability of the calculation model.

Abstract:To investigate the variation in shear mechanical properties of fractured surrounding rock at different depths from the tunnel sidewall, combining field tests and numerical simulations. First, a surface fracture network model of the surrounding rock was established using the window method. Then, drilling cameras were used to obtain images of crack distribution at different depths from the tunnel sidewall, and the crack network parameters at different depth positions were optimized. The surrounding rock was divided into 5 regions from the surface to the interior, and the diameter and bulk density variation patterns of the dominant fracture network from the surface to the interior of the fractured surrounding rock were obtained; Subsequently, discrete element numerical calculation software was used to establish models of fractured rock masses in different regions, and direct shear tests were conducted to obtain shear mechanical parameters. The results showed that both cohesion and internal friction angle of the rock mass generally increase with depth. Compared with the 6 to 8 m region, the cohesion in the 8 to 10 m region continued to increase, while the internal friction began to stabilize and slightly decreased with depth. And verify the distribution law of shear mechanical performance parameters of surrounding rock through dual hole acoustic wave testing method; Finally, based on the study of shear mechanical para-meters of rock mass at different depths in direct shear tests, numerical simulation methods were used to construct tunnel models with different numbers of surrounding rock areas. The surrounding rock was divided into regions representing different shear mechanical parameters according to depth, and the deformation of the tunnel under different mechanical parameters was compared and analyzed with the on-site monitoring results. The results showed that the settlement value of the arch crown and the horizontal convergence value of the arch foot of tunnel model, which divided more regions, were more in line with the actual situation on site.

Abstract:To address the problem that existing convolutional neural networks have difficulty in capturing long-range dependencies in human faces, a dual-branch residual network based on large convolution kernels is proposed and applied to face age estimation. Firstly, to overcome the limitation of small receptive fields of traditional small convolution kernels, large convolution kernels are adopted to improve the residual module of the deep learning model (ResNet), thereby expanding the effective receptive field of the network and more efficiently capturing the global information and long-range dependencies in face images. Secondly, considering the crucial role of facial fine features in age estimation, a detail downsampling module is introduced, which can minimize the loss of detailed information at the initial stage of the network. In the design of the network structure, the original residual module of ResNet and the improved large convolution residual module are innovatively connected in parallel to form a dual-branch residual network, and an attention module is utilized to achieve feature fusion between the two branches. To further enhance the deep mining of face age features, two large convolution residual mo-dules are concatenated after the dual-branch residual network, and through progressive feature abstraction, the model's ability to model complex age patterns is strengthened. Finally, in response to the challenges brought by the label ordinality characteristics in age estimation, the constructed dual-branch residual network is combined with ordinal regression methods. By converting age values into ordered label sequences for modeling, the model's ability to distinguish age changes is effectively improved. Experimental results show that the proposed method reduces the mean absolute error (MAE) by up to 0.46 on the UTK-FACE dataset and by up to 0.09 on the FG-NET dataset.

Abstract:To explore the mechanism by which the green space ratio in rural parks affects the psychological recovery of users, this study constructed three types of green space ratio scenarios ranging from 0% to 10%, 10% to 50%, and 50% to 80% using virtual reality (VR) technology. Physiological data were collected through electroencephalogram (EEG) and eye-tracking (ET) techniques, and a comprehensive evaluation was conducted using the Perceived Recovery Scale (PRS). Among them, the alpha waves (8 to 13 Hz) of EEG are closely related to relaxation and psychological recovery, and an increase in amplitude indicates a more relaxed state of the participants. The study found that the medium green space ratio (30% to 50%) scenario performed best in terms of security, fixation duration, and activation of EEG alpha waves (frontal F4 electrode), significantly outperforming the low green space ratio (0% to 10%) and high green space ratio (50% to 80%) scenarios. It can more effectively promote emotional regulation and thus reveals the non-linear relationship between green space ratio and psychological recovery. The low green space ratio scenario has a low security score, a shortened fixation duration to 200 ms, a decreased a wave amplitude, and the poorest perceptual recovery benefit. It is recommended to introduce dynamic landscapes to enhance the sense of depth. The high green space ratio scenario has good privacy but a single visual stimulus, a dispersed fixation duration (200~250 ms), and a lower recovery benefit. It should be moderately open to enhance transparency. The research conclusion indicates that the green space ratio of rural parks should be controlled within 10% to 50%, and combined with multi-level plants and dynamic landscape configuration, to achieve the best psychological recovery benefit, providing a scientific basis for the therapeutic design of rural parks.

Abstract:To reveal the intrinsic relationship between visual saliency calculation and the visual saliency experience and visual quality of the main space, based on a profound understanding of the visual perception process of space through visual computing, quantitative analysis was conducted on the visual saliency elements that affect the visual quality of the underlying space. The method of combining visual saliency calculation with eye-tracking was adopted to collect corresponding data for analyzing the visual quality of the building's underlying space. Firstly, the basic principle of visual saliency calculation was elaborated. Secondly, virtual reality models of different schemes for the open space of the underlying public building at different stair positions were established, eye-tracking experiments were carried out, and the ITTI visual saliency calculation model was used to generate visual calculation saliency maps. Finally, combined with subjective questionnaire surveys, the correlation between the visual saliency of the building's underlying space and the visual quality evaluation was analyzed based on the data. The results show that the real-time visual experience eye-tracking hotspots map and the visual saliency map generated by visual saliency calculation have a high similarity, and the visual saliency values are positively correlated with the subjective visual quality evaluation. The stairs located in the middle of the underlying space and the spaces with complete geometric shapes have visual saliency and become important factors affecting the visual quality of the building's underlying space. Finally, it was verified that computer visual saliency calculation conforms to the visual perception laws of building spaces and can be used as an effective tool for evaluating the visual quality of building spaces, providing a quantitative basis for human-oriented architectural design.