Based on data preprocessing, digital analysis of the asphalt pavement construction process was conducted, and important construction processes and meteorological parameters affecting construction temperature were screened using the random forest (RF) algorithm. Based on the selected important parameters, the construction temperature prediction model was established by multi-layer perception (MLP). Based on feedback control theory, the control principle of the PID controller was analyzed, and a comprehensive and multi-stage temperature feedback control model was constructed in conjunction with the construction temperature prediction model. In order to solve the problem that the super-parameter cannot be self-tuning, the genetic algorithm (GA) is used to optimize the feedback control model to make the model adaptive. The feedback control model’s construction process decision was compared and analyzed with the actual construction process parameters, and the robustness of the model’s feedback control results was evaluated to effectively adjust the construction process parameters and achieve precise control of asphalt mixture construction temperature. The research results showed that the construction temperature prediction model could accurately predict the construction temperature. The comprehensive feedback control model could feedback control all parameters, while the multi-stage feedback control model could maintain the determined parameters and feedback control other parameters. In addition, the GA-PID feedback control model based on genetic algorithms had the performance of adaptive tuning of hyperparameters. Through analysis of the feedback control results, it was found that the construction process parameters obtained from the GA-PID feedback control model were evenly distributed within the effective range of actual process parameters, maintaining good consistency with the actual construction process parameters. The GA-PID control system had good robustness for different prediction models and different temperature control, and the proposed construction process decision was consistent with actual situations.

The main bridge of the Daoqingzhou river crossing adopts a double-layer variable-height prestressed steel truss composite girder scheme. The web members are triangular trusses. The main span is 276 m, the standard truss height is 9.5 m, and the truss height near the fulcrum is increased from 9.5 m to 23.0 m. This bridge is a dual-purpose bridge for road and rail. Its biggest structure features: larger span, lower truss height, and heavier load. In order to share part of the main truss force, the upper road deck of the main bridge of Daoqingzhou cross-river passage adopts steel dense beams + concrete slabs which are combined with the beam system. The concrete slab and the steel beam are connected by shear nails. The prestress of the concrete slab is anchored on the steel beam; The lower track bridge deck adopts the integral bridge deck system of longitudinal and transverse beams + orthotropic slabs. The partial weighted section adopts a box structure, and the interior is filled with iron sand concrete for weight. In the position of truss height change, the advantages and disadvantages of two kinds of bridge deck systems, flat chord and no flat chord, are compared. At the same time, in order to meet the architectural appearance requirements of the triangular truss structure, the lower rail bridge deck adopts an innovative form of bridge deck structure at the variable height position, and it is the point-supported integral bridge deck system. The side longitudinal beams and box shaped crossbeams are set on the bridge deck system. A stable plane frame formed by side longitudinal beams and box shaped crossbeams, And through vertical supports, the rail deck system is supported on the corbels extended by the inclined web rods, and the structure is clearly stressed. In the aspect of force, a detailed structrucal calculation for the upper deck and the lower deck of the standard section is made. The calculation results show that the structural stiffness and stress are within the requirements of the specification. The more prominent torsion problem of the crossbeam is analyzed in-depth, and transverse bridge frame model and local finite element model are built for structural safety verification. In addition, for the position of Corbel force transmission, fine structural finite element calculation is carried out. Then the stress state of each component plate in this part is analyzed to ensure the safety of the structure.

In order to realize the accurate recognition of weather images in road scenes, a road weather recognition method based on road segmentation is proposed, and a sort of road segmentation fusion network (RSFN), with the overall road weather image features and road features, is established by designing a method for extracting road area characteristics combining with the semantic segmentation model. First, the original images are preprocessed through the road segmentation network to obtain the binary images, and the road area information is obtained by using the convolutional feature mask (CFM). Subsequently, a convolutional neural network, which is composed of overall network branches and road network branches, is established and used to extract the overall image area features and focus on extracting the road weather features respectively. In view of the extracted irregular road characteristics, the overall image features and the road local features are fused with CFM. Finally, the weather recognition of key road areas is carried out through a fully connected layer, and the recognition of 5 types of weather (cloudy, sunny, foggy, rainy, and snowy) is realized considering the overall weather recognition. A road multi-class weather dataset (RMWD) is established by collecting real surveillance videos of expressway in multiple urban areas with different road sections and weather conditions, and compared with different network models on testing results. The result shows that (1) under the situation of decrease in parameters and computations, the RSFN weather recognition algorithm has accuracy rate and recall rate of 85.40% and 80.30% that improved by at least 3.97% and 3.86% respectively; (2) the key areas of extracting features from network models are placed in roads by using the road weather recognition method based on road segmentation, and the effective extraction of road weather features is realized by using RSFN algorithm, which can be applied to real-time and accurate weather recognition in road scenes effectively.

In addition to helping passengers go out to reach the destination, the conventional bus travel should pay more attention to the travel experience, that is, the sense of happiness during the journey. Based on this, taking Guiyang City as an example, the paper constructs SEM-Logit model to explore the influencing factors of public transport happiness of residents in mountainous cities. The main conclusions are as follows: (1) When residents in mountain cities travel by bus, their happiness is more likely to be affected by bus service and operation service; (2) In the sense of access to bus resources of passengers in mountain cities, the perception of balance and comfort and convenience significantly affects the positive emotions of residents on bus travel; (3) Compared with the traffic community divided by 1000m isometric quadrilateral and road, the built environment index extracted from the street community divided by administrative unit has a higher degree of correlation with residents’ happiness of bus travel. (4) In terms of community built environment, population density and per capita length of road network were positively correlated with residents’ happiness of public transport trips, while distance to the city center and per capita length of bus routes were negatively correlated with them. (5) The degree of association between satisfaction, sense of gain, built environment and travel happiness is significantly different due to age heterogeneity.

To alleviate the sharp decline in traffic efficiency caused by reduced number of lanes or the presence of obstacles and reduce the risk of accidents due to frequent lane changes or acceleration and deceleration in expressway operation area, taking the Wufengshan north-south 4-lane river crossing channel in Jiangsu Province as the research environment, the related research on typical operation area of expressway is conducted, a comprehensive indicator system for efficiency, safety and fuel consumption in expressway operation area is constructed. Based on Vissim simulation software, considering different traffic volumes, assembly rates and warning distance conditions, the road scenario and warning model are created. The optimal warning distance for vehicles under different conditions in expressway operation area is studied by using Bayesian neural network for prediction. The result shows that the average relative errors of this method for predicting the comprehensive, safety and energy consumption indicators of expressway operation area are 0.4%, 0.4% and 0.2% respectively. Through the analysis of the optimal warning distance for comprehensive efficiency, safety and energy consumption indicators, it is concluded that (1) higher assembly rates lead to better result in all the 3 indicators, and the optimal warning distance becomes shorter; (2) under free flow condition, the optimal warning distance is 300-800 m and it increases with the increase of traffic volume; (3) under saturated traffic flow condition, the optimal warning distance is 800-1 200 m, but the increase in traffic volume at this time has little influence on the warning distance; (4) after traffic volume is oversaturated, the optimal warning distance is 1 200-1 800 m, and it continues to increase as the traffic volume increases. The research result can provide some reference for the safe driving of vehicles in expressway operation area and the issuance of road warning by traffic management departments, and also provide a method for predicting the optimal distance of vehicle warning in expressway operation areas.

In order to improve the accuracy of urban road network traffic flow prediction with missing values, the generator and discriminator of the generative adversarial network are reconstructed, the loss function is improved, and the traffic generative adversarial imputation network (TGAIN) is proposed for the completion of the missing data of traffic flow. Based on empirical mode decomposition (EMD), graph convolutional networks (GCN) and gated recurrent unit (GRU), EMD-GCN-GRU model is designed for urban road network traffic flow prediction. First, the traffic flow data is processed by empirical mode decomposition and each component of the same level is reconstructed as the input of the subsequent prediction model. Then, the graph convolutional networks are used to learn the road network topology to capture the spatial characteristic of traffic flow, and the gated recurrent unit is employed to capture the temporal characteristic of traffic flow. For the road network traffic flow data with missing values, TGAIN is used to complete the data, and then EMD-GCN-GRU is used to predict the traffic flow. The Shenzhen average vehicle speed data set is used to construct a variety of typical traffic flow data with different missing patterns and different missing rates to simulate the actual missing situation. The effectiveness of the method is verified on the ModelArts development platform. The results show that compared with the commonly used matrix factorization imputation method, the TGAIN model has higher completion accuracy in the random missing mode of the dataset and has better completion performance when the non-random missing rate is lower than 50%. Compared with other seven prediction algorithms, the proposed prediction method has higher prediction accuracy. Combining the data imputation method TGAIN with the traffic flow prediction method EMD-GCN-GRU for urban road network traffic flow prediction with missing values can significantly reduce the negative impact of missing data and data noise on traffic flow prediction and capture the spatial and temporal correlation of network traffic flow, which improves the accuracy of urban road network traffic flow prediction.

The transportation industry is a concentrated place of energy consumption and carbon emissions. The development of low-carbon transportation has become a new industrial form of developing a low-carbon society. The research is based on the construction of the "dual carbon" goals, and constructed a fuzzy comprehensive evaluation system with the first-level index layer of transportation infrastructure energy efficiency and development and operation, which explore the effect of low-carbon construction after the implementation of urban low-carbon transportation. First, the study used the objective function method to obtain the combined weights of the indicators combined with the AHP-CV method; secondly, according to the functional objectives of low-carbon transportation construction, it establishes five secondary indicators of subordination and development, including infrastructure management, transportation and tools, transportation system construction, transportation environmental protection policies, low-carbon transportation environment. Finally, it selected Harbin, which a typical low-carbon city in Northeast China, to dynamically evaluate the development level of low-carbon transportation from 2017 to 2020. The results show that: In the selected years at the later stage of low-carbon implementation, the development of low-carbon transportation showed a weak upward trend. And the final degree of membership is "medium, good", which is consistent with the actual deployment of low-carbon transportation in Harbin in the past three years, and reflects the rationality of the early subject selection. In addition, the impact of the 2020 epidemic and the positive drive of the "dual-carbon" response have a significant impact on low-carbon indicators. According to the results of the grading index membership degree, it is recommended that the development should focus on continuing to promote the application of low-carbon transportation, enhancing citizens’ low-carbon awareness and rationalizing the construction of a low-carbon urban transportation system. Finally, the results of the model is consistent with the status, which can provide a reference for the construction of low-carbon transportation areas under the background of the "dual carbon".