To investigate the effects of different fiber types, temperature and frequency on the resilient modulus of asphalt materials, the frequency spectrum and temperature spectrum of dynamic resilient modulus was established by combining with Sigmoidal functions based on the indirect tensile tests. The influence of temperature and frequency on the dynamic resilient modulus of lignin, polyester, and polypropylene fiber-reinforced asphalt mixtures was studied deeply. The results demonstrate that all three fiber-modified asphalt mixtures exhibit significant frequency and temperature sensitivity. At 10 Hz, the resilient modulus of lignin-, polyester-, and polypropylene-modified mixtures decreases by 22 520, 19 659 and 21 108 MPa, respectively, with increasing temperature. At 20 ℃, the resilient modulus of the above three mixtures increases by 6 139, 6 373 and 5 946 MPa, respectively, with ascending frequency. The variation rule of dynamic resilient modulus exhibits typical viscoelastic behavior of asphalt mixtures. Frequency spectra indicate that at any frequency, the resilience modulus of polyester fiber asphalt mixture is lower than that of lignin and polypropylene fiber asphalt mixtures, which indicates that polyester fiber asphalt mixture has good energy dissipation capacity, while lignin and polypropylene fiber asphalt mixture have better deformation resistance. Temperature spectra indicate that polyester fiber exhibits a lower resilience modulus within the low temperature range and demonstrates superior resistance to low temperature cracking. Whereas lignin and polypropylene fiber exhibit a higher resilience modulus within the high temperature range and demonstrate their superior rutting resistance. The established temperature spectrum model exhibits good accuracy and can accurately predict the dynamic resilience modulus of asphalt mixtures across a wide temperature range.

To clear the effect of the physicochemical properties of fillers on the rheological properties of asphalt mastic, dynamic shear rheological (DSR) tests, multiple stress creep recovery (MSCR) tests and bending beam rheological (BBR) tests were employed to study the effects of high and low temperature performances of SBS-modified asphalt mastic, which was based on the differences in physicochemical properties of limestone fillers from three different sources. Utilizing the thickness of asphalt adsorption film as a link, the influence mechanism of the physicochemical property differences of fillers on the rheological properties of asphalt mastic was analyzed. Subsequently, a grey correlation analysis model and two-factor variance analysis were employed to analyze the correlation and coupling effects of influencing factors. The research results indicate that as the volume fraction of filler increases, the thickness of the asphalt adsorption film is increased, the complex modules and rutting factors of asphalt mastic are increased, and high-temperature rheological properties and resistance to deformation are enhanced. Meanwhile, the creep stiffness of asphalt mastic increases, while the creep rate decreases, resulting in a reduction of low-temperature stress relaxation capacity and a deterioration of low-temperature crack resistance of the asphalt mastic. Furthermore, the greater disparities in physicochemical indicators of fillers such as specific surface area, alkaline component content, average particle size, the more significant effect on the rheological properties of asphalt mastic. Specifically, the larger specific surface area of filler, the higher alkaline component content and the smaller particle size, the better high-temperature performance but the worse low-temperature performance for asphalt mastic. The correlation between the physicochemical property indicators of fillers and the rheological properties of SBS modified asphalt mastic is ranked from high to low as follows: filler volume fraction > specific surface area > average particle size> CaO content > SiO2 content > MgO content.

To address the issues of low target recognition accuracy and poor detection real-time performance in the task of railway foreign object intrusion limit detection, an orbital foreign object intrusion limit detection model combining bidirectional feature fusion with focusing loss was proposed. The lightweight S-GhostNet was used as the backbone network to extract features, aiming to reduce computational complexity and enhance processing speed. A bidirectional feature fusion structure (Sim-DFPN) was designed. The non-parametric attention mechanism was introduced to enable the model to focus more on the feature information of the target region, which effectively suppressed interference from invalid background noise and enhanced the flow of information between different levels. Furthermore, in order to further improve detection performance, the intersection over union (IoU) loss was restructured. A linearly spaced method was used to design an adaptive focusing edge loss function, Focaler EIoU, and experiments were conducted on the railway foreign object intrusion limit dataset. The research results show that the proposed network achieves a detection accuracy of 90% on this dataset, with a model size of only 55 MB and a frame rate of 79 frames per second.

The addition of virgin asphalt to restore the performance of waste asphalt in factory mixed hot recycled asphalt mixtures (HRAM) often results in a decrease in their resistance to water damage. The evolution law of void characteristics in hot recycled asphalt mixtures under freeze-thaw cycles was analyzed by X-ray CT scanning experiments and digital image processing methods, in order to reveal the mechanism of freeze-thaw damage in hot recycled asphalt mixtures. Research shows that after 20 freeze-thaw cycles, the void area of HRAM increases to 4.7 times that of unfrozen asphalt mixture. The freeze-thaw cycle exacerbates void damage, causing micro-volume voids to expand until adjacent voids are connected. In the first 5 freeze-thaw cycles, the damage to micro-volume voids is primarily driven by expansion, and the proportion of voids with V<0.1 mm3 is increased. As the number of freeze-thaw cycles continues to increase, the expansion of adjacent void volumes is dominated by connectivity, and the proportion of voids with V<0.1 mm3 is decreased, while the rate of decrease in the proportion of void amounts with 0.1 mm3≤V<5.0 mm3 slows down. The effect of voids connectivity leads to a decrease in the proportion of voids amounts with 0.1 mm3≤V＜5.0 mm3, while the proportion of voids amounts with 5.0 mm3≤V＜10 mm3 increases. The freeze-thaw cycle causes both the average equivalent diameter and expected diameter of voids in the mixture to increase with varying degrees, and the difference between the three-dimensional fractal dimension and the spherical fractal dimension becomes larger. The morphology of the void space area develops in a more irregular direction.

To precisely identify traffic conflicts involving motor vehicles at urban intersections, a novel automated video detection method based on improved YOLOv8 object detection and DeepSORT trajectory tracking algorithm was proposed. By adding a small-target detection layer, incorporating attention mechanism and optimizing the loss function, the detection performance for small-scale and occluded vehicles was enhanced. An extended Kalman filter (EKF) was employed to handle non-linear motion trajectories, and cubic spline interpolation was utilized to fill in completely missing trajectories, thereby improving trajectory accuracy. Conflicts were quantified based on the time-to-collision (TTC) metric. Research results demonstrate that compared to the YOLOv8 and YOLOv5 algorithms, the improved algorithm has achieved a training accuracy improvement of 6.66% and 8.94%, respectively, and a recall rate improvement of 6.61% and 13.30%, respectively. In terms of tracking performance, compared to YOLOv8+DeepSORT and YOLOv5+DeepSORT, the improved algorithm has achieved an improvement of 4.58% and 7.10% in tracking accuracy, and an increase of 3.82% and 9.49% in tracking success rate, respectively. Based on the conflict detection results of the ROC curve, the improved algorithm achieves an AUC value of 0.854, outperforming other methods.

Aiming at the existing problems of the semantic segmentation network model of urban road, such as fuzzy segmentation boundary and insufficient semantic segmentation accuracy of target image, the semantic segmentation network for roads under extended pyramid and axial cross attention method was proposed. Firstly, the backbone network used an improved reparametrized VGG network (RepVGG+), which was faster in computation, memory-efficient and more flexible. Secondly, to enhance the ability of the model to express global information, the extended feature pyramid network (EFPN) was proposed and the multi-branch extended convolution acceleration module (ECAM) was designed to improve semantic segmentation effect. At the same time, to enhance the network model's attention to segmentation boundaries, a multi-scale axial-cross attention (MSACA) module was designed, and the spatial channel block (SCBlock) was utilized to replace ordinary convolution, thereby removing spatial information redundancy. The research results show that the proposed model can improve the issue of blurry segmentation boundaries and enhance the segmentation accuracy of the target image. The mIoU value on Cityscapes dataset reaches 81.3%, which is 4.8% higher than that of the basic model (DeepLabv3plus), reaching a good level of semantic segmentation on this dataset so far.

The accurate prediction of future traffic flow is the key of solving the urban traffic congestion problem. Aiming at the problems of insufficient time-periodic feature extraction and incomplete spatial feature mining in traffic flow prediction, a traffic flow prediction model based on multi-spatiotemporal features and graph attention network (MSTFGAN) was proposed. Firstly, three kinds of periodic data with different time periods were constructed according to the hourly, daily and weekly cycles as the input of the model, and then the temporal features of the three kinds of periodic data were extracted by using temporal convolutional network (TCN) with a gating mechanism. Secondly, an adaptive neighbor matrix was constructed to learn the correlation between the traffic sensor nodes by applying the trainable parameter, thereby revealing the potential spatial characteristics in the traffic network. Then, a graph attention network (GAT) was used to assign different weights to each road network node dynamically to capture the dynamic spatial correlation between the traffic sensor nodes; furthermore, a gated fusion module was designed to adaptively fuse implicit spatial features and dynamic spatial features. Finally, temporal and spatial features were fused based on an attention mechanism for accurate traffic flow prediction. Experimental validation was performed on two real-world highway datasets, that is, PeMS04 and PeMS08. The research results show that compared with the best-performing baseline model DSTAGNN, MSTFGAN reduces the mean absolute error (EMA) by 3.802% , the root mean square error (ERMS) by 3.780%, and the mean absolute percentage error (EMAP) by 3.356%, which indicates that MSTFGAN can effectively improve the accuracy of the traffic flow prediction and accurately predict the traffic flow trend.

In order to study the casualties caused by the use of equipment by passengers in subway stations, the equipment failure and the resulting casualty process were studied. Firstly, the use of equipment by passengers and the failure were discussed. Then, the process of casualties was studied. Finally, concepts and methods such as casualty patterns and importance degree were put forward. Research shows that equipments failure can pose safety hazards to passengers and even cause casualties and mass accidents. Failures of equipment such as automatic ticket checking machine, escalator and safety door can evolve into casualty accident. The evolution network diagram of casualty process was drawn. The concepts and mathematical models of path evolution pattern, path connectivity probability, causal event importance and transfer probability importance were proposed. The case analysis shows that the failure of safety door and escalator is more likely to cause casualties, and the specific evolution conditions are crucial to the evolution of casualty accidents.

Exploring the influencing mechanism of risky driving behavior of young truck drivers is of great significance for improving the safety of road transportation. A causal diagram model was constructed based on the theory of planned behavior (TPB). Data from 312 professional truck drivers were collected through a questionnaire survey, and two causal inference methods, namely structural equation model (SEM) and Bayesian network (BN), were integrated to analyze the psychological influencing mechanisms. The results show that: there are significant differences in the frequency of risky driving behaviors of young truck drivers of different genders, driving hours, violation frequencies and crash experience. The extended TPB model demonstrates strong explanatory power. Behavioral intention, perceived behavioral control, and risk perception have a significant impact on risky driving behavior, with behavioral intention having the greatest effect and risk perception having a negative impact. Additionally, 17.1% of young truck drivers exhibit high levels of risky driving behaviors. When perceived behavioral control and behavioral intention shift from a low state to a high state, the probability of high-level risky driving behavior increases by 16.1% and 32.1%, respectively.

Through real-ship tests, key observations were conducted on the characteristic segments such as the straight section in Yangtze-to-Huaihe communication segment, the continuously curved section in the lake area and bridge clustering area in the Yangtze-to-Huaihe water diversion project. The change characteristics of the rudder angle, the pitching and rolling and the amount of subsidence and their influencing factors under the conditions of different characteristic segments were analyzed. The research results show that under the condition of artificial canal waterways, the distribution of rudder angles is relatively concentrated, and the range is smaller. The pitching angle is not affected by the direction of the waterway or bridges and culverts, and the longitudinal attitude is relatively stable. The rolling angle is significantly influenced by the turning radius of the bends, and the range of rolling angles of the continuously curved section in the lake area is the maximum (0.78°). The subsidence is relatively small in the experimental sections with different waterway characteristics, with a maximum subsidence of 0.193m, which accounts for only 6.03% of the draft depth.

In order to investigate the treatment effect after the deployment of retaining works of the debris flow ditch at Sanyanyu in Zhouqu, Gansu Province, Massflow software was mainly used to simulate and analyze the characteristics such as the mud depth, flow rate and range after setting the retaining dam + drainage channel at the Sanyanyu Ditch under different rainfall conditions (P=1%, P=2%, P=0.5%). The influence of the rainfall and the supporting barrier factors on the debris flow ditch was studied. Finally, the hazard zoning of the Sanyanyu debris ditch after treatment was carried out. The research results show that the peak values of mud depth and flow rate are positively correlated with the rainfall intensity in the case of no retaining works. When retaining works are set, the increase of mud depth in the upstream is significant, and the mud depth downstream of the retaining works decreases significantly, with the decrease of mud depth in the accumulation area up to 4 m. After the treatment, the high-risk area is reduced from 59.6% to 27.2%, and the treatment effect of retaining dam and drainage channel is remarkable, thereby the degree of geologic hazard is greatly reduced. The research results can lay the preliminary foundation and scientific basis for the reinforcement, optimized design, monitoring, and other follow-up work of the debris flow prevention and control project in this area.

In order to study the flow-induced vibration characteristics of water-saving ship lock valves under rapid opening and closing, a three-dimensional mathematical model of the water transfer gallery and valve was established, taking a large domestic canal water-saving ship lock as an example. The standard k-ε turbulence model, dynamic mesh technology and fluid-structure coupling technology were used to study the natural frequency and modal characteristics of the valve, and the influence of different acting water heads on the parameters around the valve such as the dimensionless velocity magnitude U*, dimensionless pressure P* and valve structure deformation under rapid opening and closing was discussed. The results show that in the dry mode, when the two sides and the top of the valve slot are fixed, the fundamental frequency of the structure is 103.82 Hz, and the natural frequency increases with the increase of modal order, with the frequency difference between two adjacent higher-order modes gradually decreasing. In the rapid opening stage of the valve, the flow velocity and pressure show obvious non-steady-state characteristics, especially under high water head conditions. The flow velocity in the downstream area of the valve increases rapidly, and the high-speed area is concentrated at the bottom edge of the valve and the bottom of the downstream gallery. When the head difference is 30 m, U* can reach a maximum of 1.40 and P* can reach 1.25. The valve deformation position is concentrated in the middle of the valve beam and the bottom edge of the valve. After Fourier transform analysis, it is found that the head difference is positively correlated with the main frequency and amplitude of the valve vibration, and the deformation curve of the valve is approximately in the shape of an “8”. The research results can provide references for optimizing the operating conditions and the structural safety design of water transfer valves in water-saving ship locks.

To investigate the spatial variation rule of the solar radiation temperature field during the lifting construction process of arch ribs with oversize segment, based on the principles of heat transfer and spatial occlusion theory, the air heat exchange effect in the enclosed space of the steel box structure was taken into account. Subsequently, the temperature effect of the arch rib lifting section during the lifting construction process that took into variations of wind speed account was analyzed, and a finite element calculation flowchart for numerical simulation of the spatial temperature field of the arch rib lifting section was established. Subsequently, taking the Haihe River Bridge of the West Central Ring Expressway in Tianjin as an engineering example, a 24-hour prediction of the non-uniform temperature field was performed, and its temperature effects were analyzed. Finally, a detailed analysis will be conducted on the temperature effect of the arch rib lifting section during the lifting construction process, taking into account variations of wind speed. The results indicate that during the lifting operation from 1:00 to 6:00, the arch rib lifting segment experiences minimal bending and twisting deformation at the closure section, and the longitudinal displacement of the centroid of the section remains relatively stable, making this period an ideal time for construction.

Aiming at the problem of calculating pre-offsetting of leaning busbar saddle in spatial cable suspension bridge, an improved calculation method was proposed. By analyzing the relationship between mechanics and geometry at the cable saddle, a set of twelve nonlinear equations were constructed, which were solved by Newton-Raphson algorithm, and the saddle pre-offsetting was obtained. At the same time, the principle of the algorithm, the iterative process and the method of selecting the initial value were introduced, and the accuracy and reliability of the proposed method were proved by the numerical examples. Compared to previous methods, the proposed calculation method can meet the accuracy requirements for the construction of spatial suspension bridge, with simpler calculation and faster iteration, which provides an effective theoretical basis for the construction and optimization of spatial suspension bridge.

Steel truss-concrete composite arch is a new type of arch structure composed of concrete top and bottom slabs and steel truss webs, which can greatly reduce the dead weight of concrete arch bridge and the construction difficulty. Combined with the development of UHPC material, the UHPC was proposed to replace the conventional concrete, forming a steel truss-UHPC composite arch. Based on the Wumengshan Bridge, the trial design of this new composite arch bridge was carried out. The results show that: when basically maintaining the same cross-section dimensions as those in the original arch ring, steel cross braces are used to replace the UHPC top and bottom slabs of the middle box. The stresses in the UHPC top and bottom slabs, the embedded steel chords and the steel truss webs are all less than their strength design values in the construction stage and the bridge completion stage. The stability coefficients for the maximal cantilever, bare arch and the completed bridge are all greater than 4.0. The trial design of the arch ring meets all requirements of the specifications in terms of strength, stability and deflection calculation. Compared with the steel truss-concrete composite arch ring, the steel consumption amount of the steel truss-UHPC composite arch is reduced by 16%, the self-weight is reduced by 49.9%, and the maximum lifting weight of the arch ring segment is reduced by 39%. The axial force reduction at critical section is approximately from 18.6% to 22.0%, and the bending moments of the arch foot and the crown section are reduced by about 20% and 60%, respectively. The compressive and flexural stiffness of the section decrease, the stability is reduced, and the difference in dynamic characteristics is relatively small.