In order to explore the service level of waterways in restricted bridge area waters, and to improve the navigation efficiency and economic benefits of the shipping industry, a cellular automata dynamics model was selected to model and simulate the waterways in restricted bridge area waters and their adjacent waterways. The waterway was divided into restricted bridge area waterway and unrestricted bridge area waterway, and the cellular automata evolution rules were set respectively. The initial state presetting and evolution rule setting of the complex system could be completed efficiently by using cellular automata. The restricted bridge area waterway of the Hu-Su-Tong Yangtze River Public-Railway Bridge was taken as a case study to carry out the simulation and analysis by Python programming language, and the saturation degree and the service level of the waterway in this water area were obtained. The research results show that the saturation of the waterway shows a clear characteristic of changing over time. The experimental model is more accurate in simulating the characteristics of the traffic flow in the real environment; and the service level of waterway in the restricted bridge area can be effectively evaluated under specific conditions according to the proposed simulation model.

In order to determine the optimal plan for improving the strata and pipeline reinforcement of large-diameter shield tunnels, digital seismometer micro motion detection technology was used to comprehensively analyze the distribution and density of the strata in the reinforcement area. At the same time, unit excavation energy consumption technical indicators were introduced to evaluate the construction efficiency of each reinforcement plan. Numerical simulation was conducted to study the disturbance of pore stress in the strata during shield tunneling under different reinforcement plans, and the control effect of different reinforcement methods on seepage disturbance. The research results show that: the reinforcement of construction piles has the worst control effect on pipeline settlement due to the unreinforced shallow soil, followed by sleeve valve pipes. Combination grouting reinforcement can significantly improve the control effect on pipeline settlement. Through micro-motion detection, it can be seen that sleeve valve pipe grouting has a better filling effect on shallow clay layers, while MJS construction method piles can effectively compensate for the shortcomings of sleeve valve pipe grouting reinforcement, which has good adaptability to the cohesive soil layer between crushed stones and the lower silt rock layer, can effectively strengthen the compactness of the formation and reduce the permeability coefficient. By analyzing the changes in shield tunneling construction efficiency before and after reinforcement, it can be concluded that the uniformity of the strata in the sleeve valve pipe reinforcement section is poor, with large fluctuations in unit energy consumption and actual loss values of the strata. The compaction and uniformity of the strata after MJS grouting reinforcement are relatively high, and the working face in the reinforcement area is also relatively stable, which is conducive to controlling surface settlement and deviation from the tunnel axis.

In recent years, due to river sand mining and flood disaster, the local scouring at the bridge site often develops very violently, which is very easy to cause scour damage to the bridge pile foundation in the disaster area. The rapid assessment method of bearing capacity of bridge pile foundations under the coupling effect of scouring and vehicle loads was developed by iterative finite element calculation. Firstly, based on the constitutive curve of concrete, an improved p-y curve that could take into account the scour pit shape and stress history was used to simulate the interaction between the pile foundation and the soil. Subsequently, based on the load displacement curve at the top of the pier, the allowable stress method and the ultimate bending moment method were respectively adopted to calculate the bearing capacity of the bridge pile foundation under the combined effect of scouring and vehicle loads. Finally, the finite element software ABAQUS was used to establish the solid model of single column pile and soil model of bridge substructure to verify the accuracy of the proposed method. The results show that: when the scouring depth is between 0%~33% pile length, the ultimate load value of the bridge pile foundation decreases rapidly. After the scouring depth reaches 33% of the pile length, the downward trend of the ultimate load value slows down and tends to be smooth gradually.

Regarding the issue of missed, erroneous, and false detections in the detection of multiple defects in insulators, an IND-YOLO insulator defect-detection algorithm based on improved YOLOv8 algorithm was proposed. By combining the deformable convolution Dcnv2 and the C2f structure in the network, the algorithm parameters could be reduced, and more attention could be paid to the variable characteristics of insulator defects algorithm. The coord attention attention mechanism was used to improve the feature extraction ability of the region of interest. Considering the insufficient ability of the model to detect small target defects, IND-YOLO algorithm added shallow output. The Siou loss function was used to accelerate the convergence speed of the model. The experiment results show that the average detection accuracy Map@0.5 of the proposed model reaches 0.943, which is 5.6% higher than that of benchmark model. The F1score reaches 0.92, which is 4.6% higher than that of YOLOv8, and the detection speed can reach 62 frames per second. The proposed model has a wide application prospect in insulator multi-defect detection.

The main disease of pile foundation bridges is the decrease in the height of pile-soil interaction caused by foundation scour, which reduces the bearing capacity of the pile foundation. In traditional scour numerical model, it is assumed that the sediment in the upstream riverbed is not initiated and the water flow remains clear without sediment, ignoring the influence of suspended mass on scour depth in actual turbid water environments. In order to predict the scour evolution of the bridge submerged foundation in turbid water environments, the series model extension method was employed on the basis of the computational fluid dynamics method, a numerical calculation model for multi-phase live-bed scour was established with the combination of the theory of sediment initiation and transport. Compared with Melville classical scour test, the accuracy of the numerical model was verified. Taking a typical pile foundation bridge as the research object, the changes in local scour depth and the final equilibrium scour depth of the bridge foundation were predicted and compared with empirical formulas and field scanning results. The research results show that the empirical formula used in the specification is conservative in calculating the scour depth. The proposed multi-phase simulation method of the live-bed scour evolution can provide a more accurate prediction of the scour depth, which is closer to the field scanning results.

In order to investigate the manifestation of the scale effect on the shear strength of coarse grains and to explore the method of eliminating the scale effect, large-scale consolidation and drainage (CD) triaxial tests were carried out on sand-cobble gravels and rock piles. Due to the size limitation of the apparatus, the equivalent substitution method and mixing method were used to scale the sand-cobble gravel and the piled stone respectively, so that the maximum particle size dM of the scaled samples reached 10, 20, 40, and 60 mm, respectively. The test results show that dM has a significant impact on the apparent cohesion c and effective internal friction angle φ′ of the c shear strength parameters of coarse-grained materials, and c and φ′ decrease gradually with the increase of dM. Moreover, the relationship curves between dM and c as well as φ′ can all be expressed as logarithmic functions. According to the relationship between dM and c as well as φ′, the relation between shear strength and dM is obtained, which can be used to predict the shear strength of the original graded coarse grains in a relatively simple way. On the other hand, an estimation formula for predicting the shear strength of the original graded coarse grains has been derived based on previous research results, and the reliability of this formula has been verified by experimental data. And it has been mutually proven with the relation between shear strength and dM that both formulas can accurately predict the shear strength of the original grade, thus proposing a method to eliminate the influence of scale effect on shear strength.

In order to accurately evaluate the environmental benefits of warm mix regeneration integration technology, the boundary range of environmental benefits during road construction was determined based on the life cycle method. The carbon emission calculation model was established. Through on-site investigation and data analysis, the energy consumption and greenhouse gas emissions of warm mix regeneration technology and hot mix technology were calculated respectively by the proposed model. The energy saving and emission reduction benefits of warm mix regeneration fusion technology were compared. The result show that during the construction process of warm mix recycled pavement, the energy consumption of raw material production is 324.47 MJ, and its greenhouse gas emissions are 19.65 kg CO2e; the energy consumption of mixture mixing is 217.28 MJ, and its greenhouse gas emissions are 18.15 kg CO2e. Compared with hot mix technology, warm mix regeneration fusion technology reduces energy consumption and greenhouse gas emissions during the mixing stage of the mixture. When the RAP content is 40%, the energy consumption of warm mix regeneration pavement during the construction period is reduced by 33.33%, and CO2 emissions are reduced by 20.11%.

According to the filler characteristics of filling-stone subgrade of highway in mountainous area, the microscopic parameters of field fillers were determined by using a combination method of laboratory test and numerical simulation. Discrete element software was used to simulate the vibration compaction, and the impact of various rolling parameters on subgrade compaction was investigated. The rapid detection method suitable for the compaction quality of filling-stone subgrade in mountainous areas was explored by combing with the water filling method test results. The results show that in the vibration compaction process of filling-stone subgrade, the force chain system between particles is a dynamic process of continual breaking and re-establishment, and the entire subgrade structure is formed by the contact between large particle size aggregate particles to form a strong chain. It is recommended to use a compaction parameter combination with a vibration frequency of 28 Hz, excitation force of 400 kN, and amplitude of 1.0 mm, which has better compaction efficiency and quality. It is found that there is a negative correlation between the compactness and penetration of heavy dynamic penetration test of large particle size filling-stone subgrade in mountainous areas, and the heavy dynamic penetration test can be used as a rapid detection technology for the compactness detection of large particle size filling-stone subgrade. Sensors embedded in the filling-stone subgrade provide long-term monitoring and perception of its service performance. The results show that the subgrade settlement rate is around 0.5 mm per day, with no uneven settlement. The soil pressure peaks around 70 days after construction and remains stable, and the maximum horizontal displacement of the slope is 7 mm. During the monitoring period, all monitoring indicators dont reach the set warning values, which verifies the effectiveness and reliability of the proposed rolling parameter combination and compaction quality control technology of heavy dynamic penetration test suitable for filling-stone subgrade of mountainous expressway.

To analyze the damage state of highway slope under the influence of strong earthquake, a field survey of earthquake damaged slopes in Yangbi earthquake affected area was carried out by UAV scanning. The damage state of typical earthquake damaged slope, that is K70C slope, was quantitatively evaluated by the combination of electrical measurement method and damage theory, and the stability of K70C slope was analyzed by using grid reconstruction technology and finite element simulation. The analysis results show that the instability mode of the slopes along Dayangyun Highway under the influence of Yangbi earthquake is mainly traction type. Based on the difference of the damaged soil state, the K70C slope can be divided into three parts: gravel soil, fractured soil and slate, and the material damage factor threshold is 0.44. According to the strength reduction method, the safety factor of K70C slope after seismic disturbance is 1.105, and the maximum plastic strain is 0.26 in this state and the maximum clearance displacement is 0.83m. There are two plastic deformation zones in K70C slope under the influence of earthquake, which are respectively located at the slope top at the depth from 10 to 20 m and at the back edge of the second-grade slope to the top of the fifth-grade slope. After rainfall infiltration in the rainy season, two landslides occur again in K70C, and the slip characteristics are basically consistent with the results of finite element analysis.

Aiming at the problem of insufficient adaptability and detection accuracy in lane detection in complex scenes, a lane detection and tracking algorithm based on instance segmentation was proposed. The proposed algorithm employed an Encoder-Decoder network model for instance segmentation and incorporated a convolutional block attention module (CBAM) to enhance the accuracy of lane line segmentation. The segmented images that could effectively isolate road characteristics from environmental interference were obtained. Finally, combining the variable perspective transformation matrix to transform the road image, polynomial fitting was used to generate the lane line parameter equation. The research results show that the proposed algorithm achieves a detection accuracy of 96.60% on the TuSimple dataset and an FS1 score of 79.8% on the CULane dataset, which not only significantly improves the segmentation speed of lane lines but also has good robustness and detection accuracy in complex traffic environments.

The formation and propagation mechanisms of traffic oscillations are significantly influenced by the heterogeneity in driving behavior. By integrating the dual parameters of driver reaction ability and driving behavior asymmetry, the classic car following model was improved, and an asymmetric car-following model was constructed to reveal the micro-evolutionary patterns of traffic oscillations. Stability analysis indicates that traffic flow instability phenomena occur within specific ranges of speed and headway parameters. Simulation experiments further identify three kinds of triggering conditions for oscillations: traffic flow parameters entering the theoretical instability zone, initial headways of aggressive drivers being unable to match their speed requirements within the stable range as well as mismatch between speed and headway parameters for ordinary or conservative drivers within the stable range. Quantitative research based on the standard deviation of speed shows that the oscillation propagation speed of conservative driver fleets at low speeds is faster than that of aggressive drivers and the oscillation caused by the disturbance of the headway shows a concave growth. While the growth pattern caused by speed disturbance occurs differentiated growth patterns among driver types—exponential growth for aggressive and ordinary drivers, and linear growth for conservative drivers. When initial speed and headway achieve the parameter matching state, the propagation of oscillations is significantly suppressed or even blocked. This study establishes a quantitative mapping relationship between driving behavior characteristic parameters and the evolutionary patterns of oscillations, confirming that parameter matching can reduce the risk of oscillation propagation, which provides theoretical support for the design of cooperative control strategies for intelligent connected vehicles, especially in areas such as adaptive cruise control system parameter optimization and enhancement of the stability of mixed traffic flow.

In order to study the safe lane changing behavior in highway weaving areas and reveal the influence of potential heterogeneous interaction characteristics in the merging decision-making behavior of vehicles on entrance ramps on the merging behavior, the grouped mixed parameter logit model that took into account the mixed parameter correlation and heterogeneity in means was proposed. The proposed model was calibrated by the variables extracted from the NGSIM dataset, and the influence of each variable on the merging decision was quantitatively analyzed by combining with marginal effect. The research shows that the grouped mixed parameter logit model, which combines mixed parameter correlation and mean heterogeneity, exhibits significantly better data fitting performance than the conventional grouped mixed parameter logit model, the grouped mixed parameter logit model, which only deals with mixed parameter correlation, and the grouped mixed parameter logit model, which only explores the phenomenon of mean heterogeneity. Two significant mixing parameters captured by the proposed model and the mean value of the mixed parameter of the longitudinal distance difference between the front and rear vehicles in the target lane increase with the increase of the speed difference between the converging vehicle and the vehicle in the front of the target lane, and decrease with the increase of the lateral distance difference between the converging vehicle and the vehicle in the front of the target lane. There is a significant positive correlation between the above two mixed parameters, whose interaction will increase the probability of the merging vehicle driver changing lanes.

Research on the security incident identification method for highway reconstruction and expansion zones is relatively weak, lack of complete and efficient detection methods. A safety incident identification model based on an unsupervised double-layer stacking framework was proposed for the unique requirements of the guided improvement points in highway reconstruction and expansion zone. The proposed model aimed to enhance the accuracy rate of automatic identification of traffic safety incidents. By integrating Bootstrap resampling technology, multiple weak learners, and the XGBoost meta-learner, the challenges posed by changes in road alignment and traffic flow status in reconstruction and expansion zones were effectively addressed. The proposed model demonstrated high precision and accuracy in identifying abnormal states during security incidents. The significant contribution of each component of the double-layer stacking framework and the advantages of the overall model were verified through ablation experiments. Compared with traditional models (SVM, RF, KNN), it is indicated that the proposed model has improved identification accuracy by over 30% and precision by over 7%. This study not only provides an efficient technical solution for traffic safety management in highway reconstruction and expansion areas but also paves new paths for future research and application of safety incident identification in intelligent transportation systems.

In order to further investigate the ultimate load-bearing behavior of hull plates, the ultimate strength of hull plates under the combined action of longitudinal cyclic loading and lateral pressure was studied. The non-linear finite element numerical simulation was carried out by ABAQUS software and the effect of different lateral pressures on the ultimate strength of hull plates under longitudinal cyclic loading was analyzed. Meanwhile, different plate flexibility coefficients, the welding residual stress in the process of welding and the Bauschinger effect of steel under reverse loading were also considered. The research results show that the welding residual stress significantly affects the ultimate strength and plastic yield distribution of the hull plate at the first ultimate state under the combined action of lateral pressure and longitudinal cyclic loading. After five cycles of compression-tension, the residual ultimate strength of the hull plates considering only the initial deflection defect decreases by 23.77%~39.58%, while that considering intact initial imperfections decreases by 21.82% ~36.17%. It is indicated that the influence of welding residual stress on the ultimate strength of hull plates is weakened by lateral pressure. The smaller the plate flexibility coefficient is, the less impact the elevation of lateral pressure has on the ultimate strength of hull plates at the first compression.

In order to evaluate the aerodynamic characteristics and structural strength of the protective cover of the pantograph detection device of EMU, the flow field simulation was conducted on the protective cover installed on EMU roof, the aerodynamic loads at different speeds and motion directions were obtained. The pressure and aerodynamic resistance of the protective cover under various speeds and accelerations were tested using a dynamic model experimental platform, which verified the accuracy of the flow field simulation. Finite element analysis was conducted on the structural strength of the protective cover by taking aerodynamic and vibration loads as load inputs for structural simulation. The results show that the pressure on the windward side of the protective cover is the highest, and the pressure in the central area is greater than that in the edge area. The airflow on the leeward side separates, causing backflow and lower pressure. When the inflow velocity is positive or negative, the resistance coefficients of the protective cover with the lower base mounting flange are 0.35 and 0.32, respectively, and the resistance coefficients of the protective cover itself are 0.14 and 0.12, respectively. Under the action of aerodynamic loads, inertial loads and impact loads, the safety factor is greater than the allowable factor, and the structural strength of the protective cover meets the requirements for use.