48 results about "Trip chain" patented technology

The invention discloses a subway passenger flow simulation deduction method considering an individual trip chain and an activity mode, and relates to the technical field of subway operation management, and the method comprises the following specific steps: constructing a topological framework, calculating an external factor attention weight, constructing an individual trip behavior library, carrying out dynamic topological adjustment, and executing passenger flow simulation and deduction. By constructing the space-time heterogeneous topology model and dynamically adjusting the weight values of the nodes and the edges, the precision and the real-time performance of subway passenger flow simulation are improved, physical parameters and historical operation data of urban subway lines are collected, an external factor input layer and a calibration layer are additionally arranged, and the accuracy of subway passenger flow simulation is improved. The spatial-temporal heterogeneous topology model can reflect the influence of external factors on the subway passenger flow in real time, the station retention weight and the line passing weight are dynamically corrected by calculating the attention weight of the external factors and combining data in the individual travel behavior library, and the prediction of the subway passenger flow change is realized. And a scientific basis is provided for subway operation managers.

The invention discloses a truck trip chain identification method based on trajectory data, which comprises the following steps: S1, acquiring GPS trajectory data of a truck, and preprocessing the GPS trajectory data, including data cleaning, noise removal and trajectory segmentation; s2, by setting a speed threshold value and a time threshold value, the stop point of the truck is recognized; s3, spatial matching is carried out on the stay point and road network data, and the accurate position of the stay point is determined; s4, segmenting the track of the truck based on the time sequence and the space distribution of the stop points to form a plurality of travel segments; s5, analyzing the driving characteristics and the stop characteristics of each travel section, and removing stop points related to non-freight; and S6, identifying key staying points related to loading and unloading activities, and connecting the key staying points according to a time sequence to form a truck travel chain. According to the method, the actual running path and the logistics movable chain of the truck can be clearly restored, and high-precision technical support is provided for freight path analysis and industrial chain space structure recognition.

The invention provides an electric vehicle charging load prediction method based on travel characteristics, which adopts a Monte Carlo method to simulate travel and charging behaviors of electric vehicles, and comprises the following steps: based on a travel chain model, extracting the initial travel time, the initial travel place, the initial charge state and the travel purpose of each time period of each electric vehicle; according to the travel purpose, extracting the functional area which each electric vehicle needs to reach in each time period; according to the functional area needing to be reached, path planning is carried out on the driving path of each electric vehicle; calculating the driving energy consumption on the driving path, and updating the charge state of the electric vehicle; the charging probability of the electric vehicle is judged according to the charge state; and finally, according to the charging probability of each electric vehicle in each area in different time periods, obtaining the charging load space-time distribution of the electric vehicles. By comprehensively considering various travel characteristics, the reliability of the prediction method is enhanced, and the problem of insufficient prediction precision is solved.

The application provides a low-carbon travel carbon emission reduction amount verification method and system based on travel chain big data, which comprises the following steps: step 1) obtaining sample data of travel data; step 2) when the sample data is walking data, going to step 3); when the sample data is bicycle data, going to step 4); when the sample data is electric bicycle data, going to step 5); step 3) verifying the sample data by using a walking travel distance verification method and removing duplicates to obtain a final effective travel distance; calculating the carbon emission of walking according to the final effective travel distance; step 4) verifying the sample data by using a bicycle travel distance verification method and removing duplicates to obtain a final effective travel distance; calculating the carbon emission of the bicycle according to the final effective travel distance; step 5) verifying the sample data by using an electric bicycle travel distance verification method and removing duplicates to obtain a final effective travel distance; and calculating the carbon emission of the electric bicycle.

This invention relates to the field of electric vehicle trip chain simulation technology, specifically to a method and system for simulating the trip chain of electric heavy-duty trucks considering charging strategy optimization. The steps are as follows: Real-time updates of target station distance and battery charge; upon arrival at a loading / unloading station and when the loading status meets loading / unloading conditions, updating the loading / unloading status and executing an event; after the event, updating to a driving status; upon arrival at a charging station and when the battery charge meets charging conditions, if there is an available charging pile or the overall satisfaction level is greater than a threshold, updating the charging status and executing an event; after the event, updating to a driving status; if the overall satisfaction level is less than the threshold, directly updating to a driving status; updating the target station for the vehicle in the driving status and repeating the above steps until the entire trip chain simulation is completed. This invention has significant reference value for electric heavy-duty truck capacity demanders in terms of operational scheduling plans and station configuration, and can promote improvements in fleet operating efficiency and economy.

The application provides a cross-border passenger travel chain identification method and system, which comprises the following steps: screening the base stations of the exit port / airport based on the base station position information, and preliminarily screening the potential cross-border passengers; obtaining the full-chain travel of the potential cross-border passengers in the country through the residence area pre-identification algorithm based on the mobile phone signaling of the passengers in the country; constructing a residence mathematical model, screening the residence point with the highest credibility in the exit port / airport as the cross-border entry and exit point through the residence mathematical model, and determining the position of the cross-border entry and exit point; distinguishing the entry and exit behaviors of the cross-border passengers according to the residence time of the cross-border passengers in the port / airport and the mobile phone signaling disappearance characteristics; and combining the full-chain travel of the cross-border passengers in the country, the position of the cross-border entry and exit point and the entry and exit behaviors to construct the cross-border passenger travel chain. Through the scheme, the accuracy of the cross-border passenger travel investigation data can be improved, and the coverage time and coverage range of the investigation data can be improved.

The invention discloses a charging network guiding method and system based on trip chain simulation, and the method comprises the steps: firstly, building an electric vehicle trip probability model based on a trip chain theory, generating a virtual scene containing a vehicle driving track and charge state evolution through employing a Monte Carlo simulation method, and when a charging triggering condition is satisfied, carrying out the simulation of an electric vehicle trip probability model; recording the demand information to form a charging load basic data set for charging network planning; then, acquiring short-time load prediction data of a charging station; and finally, based on the prediction data, generating a guidance index related to the charging congestion degree, issuing the index to a user terminal for charging navigation, and reporting the prediction data to a power grid system at the same time. The problems that charging network planning lacks data support and information islands exist in operation are solved, planning basis is provided through scientific simulation, information linkage among the charging stations, the users and the power grid is established, and the cooperative operation efficiency of the whole energy traffic system is effectively improved.

The invention discloses an electric vehicle space-time charging adjustable flexibility quantification method and device, and the method comprises the steps: building a charging space-time flexibility model based on a trip chain, and covering slow charging time and fast charging space flexibility modeling; a double-criterion random dynamic user balance model is constructed, and fast and slow charging loads in a balance state are obtained; and finally, quantitatively evaluating the slow charging time flexibility and the fast charging space flexibility potential by using the flexibility model. According to the method, traffic, price and user perception differences can be comprehensively considered, the charging load space-time schedulable characteristics are comprehensively reflected, and powerful support is provided for traffic-power system cooperative scheduling.

The application provides a travel service recommendation method and device, an electronic device and a storage medium. The method comprises the following steps: performing regional clustering based on station position information of a traffic station to obtain a plurality of candidate travel region pairs, and constructing a candidate travel chain; obtaining first historical travel data in a preset historical period, and extracting corresponding second historical travel data for a preset operation time interval; for each candidate travel region pair, determining a maximum number of transfers, and obtaining a target historical passenger flow in a continuous historical time interval; combining a preset screening condition to screen a target travel region pair; extracting third historical travel data, generating a travel inappropriateness evaluation index of each operation time interval based on the first and third historical travel data for each target travel region pair, and recommending a travel service to a travel object according to the travel inappropriateness evaluation index. The application can provide personalized traffic services for travel objects with the same travel demand and provide a good travel experience.

The invention relates to the technical field of intelligent traffic, and particularly discloses a traffic flow prediction system based on trip chain dynamic evolution. The system comprises a trip chain real-time sensing and reconstruction module, a trip decision behavior dynamic deduction module, a group trip demand evolution calculation module and a macroscopic traffic flow dynamic prediction module which are connected in sequence. Through real-time sensing and reconstruction of an individual trip chain, individual behaviors are dynamically deduced, group demands are simulated and aggregated based on an intelligent agent, and a macroscopic model is driven to carry out prediction. The system constructs a complete technical closed loop from microscopic individual behavior real-time perception to macroscopic road network flow dynamic prediction. According to the system, through cooperation of the travel decision behavior dynamic deduction module and the group travel demand evolution calculation module, fine simulation of a group travel behavior emergence effect under an emergency situation is realized.

The application discloses a kind of motor vehicle trip chain intelligent generation method based on attention mechanism, including obtaining the trip observation data of individual vehicle using AVI detector, constructs the association system of trip observation and behavior attribute, and generates embedding vector representation by semantic and syntax embedding method;Combined with the characteristics of fragmented data, the multi-head attention mechanism of spatio-temporal association is fused, the coding and decoding structure is improved, and the trip chain is generated;Adopt the multitask pre-training strategy of autoregressive generation and contrast learning;Through trip group clustering and hierarchical fine-tuning strategy, implement individual heterogeneity adaptation using lightweight adapter, generate candidate trip chain in instruction-reply form, filter high-confidence results according to generation probability and spatio-temporal consistency, provide accurate and diversified support for motor vehicle trip chain reconstruction;The application can efficiently and accurately infer and generate complete motor vehicle trip chain under fragmented observation conditions, and consider the generalization ability and individual difference adaptability of the model.

The invention provides a cross-border passenger trip chain identification method and system, and the method comprises the steps: screening out a base station of a port / airport based on the position information of a base station, and carrying out the preliminary screening of potential cross-border passengers; the method comprises the following steps of: acquiring full-chain travel of potential cross-border travelers at home through a residence area pre-identification algorithm based on mobile phone signaling when the travelers reside at home; constructing a residence mathematical model, screening a residence point with the maximum credibility in the port / airport through the residence mathematical model as a cross-border access point, and determining the position of the cross-border access point; distinguishing entry and exit behaviors of the cross-border passengers according to the residence time of the cross-border passengers at the ports / airports and the mobile phone signaling disappearance characteristics; and merging full-chain travel, cross-border entry and exit point positions and entry and exit behaviors of the cross-border passengers in the home, and constructing a cross-border crowd travel chain. Through the scheme, the accuracy of the cross-border passenger travel survey data can be improved, and the coverage time and coverage range of the survey data are improved.

The invention discloses a bus OD derivation and road network flow matching method based on multi-source data and electronic equipment, and the method comprises the steps: obtaining and preprocessing bus multi-source data in a target area, and building a station-line-vehicle three-dimensional mapping relation with direction information; the bus multi-source data comprises bus GPS (Global Positioning System) data, station GIS (Geographic Information System) data and IC (Integrated Circuit) card swiping data; determining passenger types and historical travel laws according to the IC card swiping data, analyzing travel chain features according to the historical travel laws, and then deducing getting-off stations and time of passengers in combination with POI types of stations; according to the deduced get-off station of the passenger and the corresponding route, determining a station sequence interval of the travel path of the passenger on the bus route; then, according to the corresponding relation between the line station and the road section number in the road network, the passenger travel path is mapped to the actual bus road network, the total passenger flow of each road section is counted, and flow statistical data with the road section as the unit is generated.

The invention discloses an electric vehicle charging scheduling method and electronic equipment, relates to the technical field of new energy, and can accurately understand a travel intention and a charging demand of a user by acquiring information such as a starting point, a terminal point, a departure moment and an initial charge state of a target electric vehicle. And the charging scheduling scheme and the actual travel plan of the user are ensured to be closely fit. The trip chain is constructed based on the starting point and the ending point, and the optimal path is generated, thereby ensuring the reasonability and high efficiency of path selection. And the optimal path and the initial charge state are combined to accurately predict the time, the place and the required charge state of the charging service of the electric vehicle, so that the charging station can be planned in advance, and the time cost for searching the charging pile is reduced. Charging can be reasonably arranged in an idle window of a user travel by calculating a charging schedulable time period. And a moss growth optimization algorithm is adopted to solve the charging power scheduling model, so that the feasibility of a charging scheduling scheme and the stability of the whole system are ensured.

The application discloses an intra-city traveler portrait system based on mobile signaling data, comprising a data preprocessing module, a travel chain extraction module, a travel mode recognition module, a daily travel mode recognition module, a travel path flow recognition module and a traveler portrait module. The system loads the dwell point data processed by the mobile signaling data supplier and combines the administrative division data for preprocessing, extracts the travel chain, recognizes the travel mode by using the Gaode map API and the log Gaussian mixture model, recognizes the daily travel mode by combining the convolution self-encoder and the K-means algorithm, counts the travel path flow, and finally summarizes the labels of the travel characteristics to construct the traveler portrait. The system optimizes the data processing flow, improves the accuracy of the travel mode recognition, enhances the interpretability of the travel mode recognition, and associates the macro traffic flow with the individual travel path, thereby providing a precise and efficient analysis tool for traffic demand management.

The invention specifically discloses a scenic spot vehicle behavior identification method based on a trip chain state machine and unsupervised clustering, and relates to the technical field of intelligent traffic. The method comprises the following steps: firstly, judging vehicle access states according to passing data of scenic spot checkpoints, grouping and sequencing, and constructing three types of trip chains; dividing the vehicles into large vehicles, medium vehicles and small vehicles, aggregating and analyzing travel chain data of each vehicle type, and constructing vehicle behavior feature vectors; obtaining a weighted vehicle trip chain feature vector after feature optimization, standardization and adaptive weighting; after the abnormal trip chain is detected and isolated, an initial clustering result of vehicle behaviors is obtained, and the optimal clustering number is determined; obtaining a stable vehicle behavior clustering result through stability evaluation parameter adjustment; and finally, outputting a final recognition result according to the distribution difference of the vehicle type clustering centers. The method improves the recognition accuracy and adaptability, provides data support for fine management of scenic spot traffic, and can also be applied to traffic scenes with clear access boundaries.

The invention discloses a user trip chain synthesis and labeling method based on an NFC-SIM card, and relates to the technical field of data processing, and the method comprises the steps: responding to a selection operation of at least one target card swiping record in a labeling task set, and in each track point of signaling track data corresponding to the target card swiping record, labeling the target card swiping record in the labeling task set; determining a candidate alignment point which is in space-time alignment with the target card swiping record; determining a travel OD segment based on the candidate alignment points and the signaling trajectory data; in response to a travel mode label carried by the travel mode labeling operation, associating the travel mode label to the travel OD segment; and in the multiple travel OD segments of the same user, determining time-space continuous target OD segments, synthesizing the target OD segments into a composite travel chain, generating a combined tag based on the travel mode tag of each target OD segment, and associating the composite travel chain. According to the invention, the accuracy of user trip chain labeling is improved.

The application discloses a user travel chain synthesis and labeling method based on an NFC-SIM card, relates to the technical field of data processing, and comprises the following steps: in response to a selection operation on at least one target card swiping record in a labeling task set, determining candidate alignment points that are aligned in space and time with the target card swiping record from each track point of signaling track data corresponding to the target card swiping record; determining a travel OD section based on the candidate alignment points and the signaling track data; in response to a travel mode label carried by a travel mode labeling operation, associating a travel mode label with the travel OD section; in multiple travel OD sections of the same user, determining target OD sections that are continuous in space and time, synthesizing the target OD sections into a composite travel chain, and generating a combined label based on the travel mode labels of the target OD sections and associating the composite travel chain. The application improves the accuracy of user travel chain labeling.

The present application relates to a kind of electric vehicle charging power calculation method, the method includes the following steps: using the database data of Internet of Vehicles to the space-time characteristic parameter in trip chain is statistically analyzed and is synthesized and establishes mathematical model to simulate user travel regularity;Different probability distribution functions are constructed to simulate the single charging behavior characteristics of electric vehicle user in real time;Determine the power size of charging and the duration length of charging;The charging power of single electric vehicle in daily is estimated and calculated;Obtain the space-time distribution curve of electric vehicle charging power in single day.The present application considers the charging behavior habit of user, the state of electric vehicle battery and the anxiety psychology of user to potential trip, establishes more real user charging decision model, can more practically and accurately estimate the space-time distribution of electric vehicle charging power in single day, has strong practicability.

The invention provides a travel service recommendation method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the region clustering based on the site position information of a traffic site, obtaining a plurality of candidate travel region pairs, and constructing a candidate travel chain; obtaining first historical travel data in a preset historical period, and extracting corresponding second historical travel data for a preset operation time interval; for each candidate travel area pair, determining a maximum transfer frequency, obtaining a target historical passenger flow in a continuous historical time interval, and screening a target travel area pair in combination with a preset screening condition; and extracting third historical travel data, aiming at each target travel area pair, generating a travel inadequacy evaluation index of each operation time interval based on the first and third historical travel data, and carrying out travel service recommendation on the travel object according to the travel inadequacy evaluation index. According to the invention, personalized traffic services can be provided for travel objects with the same travel requirements, and good travel experience is provided.

The application provides a travel chain processing method and device, equipment and storage medium, and relates to the technical field of big data. The travel chain processing method comprises the following steps: determining a feature constraint condition of a target travel chain according to a first historical travel chain and a second historical travel chain which are close in time sequence; the feature constraint condition comprises a constraint condition that at least one target travel feature in the target travel chain needs to meet; determining a target travel feature set meeting the feature constraint condition from a historical travel feature set; each historical travel feature in the historical travel feature set is determined based on a historical travel chain set corresponding to the historical travel feature set, and the historical travel chain set comprises historical travel chains with similar travel features; and generating the target travel chain by using a plurality of historical travel features in the target travel feature set. According to the application, missing features of various types can be completed, and the continuity, integrity and accuracy of the travel chain completion can be improved.

The invention relates to the field of big data and traffic planning, and particularly discloses a holiday travel passenger flow calculation method based on a mobile phone signaling position stay point, which performs real-time parallel cleaning and multi-dimensional fusion on incremental mobile phone signaling, traffic operation and scene attribute data through a stream-oriented computing architecture. And an effective space-time increment calculation mechanism of a road network topology constraint is introduced, a signaling drift point is mapped to a real road network through probabilistic matching and Viterbi path optimization, and an effective path distance conforming to a physical rule is used for replacing a spherical linear distance, so that false staying is accurately stripped. Meanwhile, according to the scheme, multi-source data are fused in real time on the basis of a streaming computing architecture, POI semantics are used for conducting dynamic verification and complementation on a trip chain, and extracted high-dimensional spatial-temporal features are input into a GRU + GAT hybrid model. Through full-link optimization from data underlying logic repair to upper-layer model space-time perception, dynamic measurement and calculation of global passenger flow are realized.

The application provides a low-carbon travel carbon emission reduction amount verification method and system based on travel chain big data, which comprises the following steps: step 1) obtaining sample data of travel data; step 2) when the sample data is walking data, going to step 3); when the sample data is bicycle data, going to step 4); when the sample data is electric bicycle data, going to step 5); step 3) verifying the sample data by using a walking travel distance verification method and removing duplicates to obtain a final effective travel distance; calculating the carbon emission of walking according to the final effective travel distance; step 4) verifying the sample data by using a bicycle travel distance verification method and removing duplicates to obtain a final effective travel distance; calculating the carbon emission of the bicycle according to the final effective travel distance; step 5) verifying the sample data by using an electric bicycle travel distance verification method and removing duplicates to obtain a final effective travel distance; and calculating the carbon emission of the electric bicycle.

The application discloses a traffic flow prediction method based on predefined space-time joint graph decoding and belongs to the technical field of public transport data analysis and prediction. The method comprises the following steps: collecting rail transit AFC gate passage data and train operation timetable, constructing a space-time enhanced OD matrix through abnormal value filtering and trip chain completion, and constructing a time dimension heterogeneous graph connection by using a dynamic time warping algorithm; designing a space-time physical coding module, introducing a time attention mechanism and a space attention mechanism, replacing a feature point multiplication form with feature splicing, reserving effective information obtained after passing through the two attention layers, and balancing the space-time feature fusion proportion in combination with a gate aggregation structure; constructing a space-time physical decoding module, modeling the space-time dependence of target flow on a sequence under the condition that the target flow is unknown, and reflecting comprehensive and dynamic space-time correlation by using a convolution predefined and adaptive space-time joint graph.

The application discloses a bus passenger alighting station inference method based on a travel chain and an integrated learning, and comprises the following steps: collecting bus multi-source data; performing data cleaning and preprocessing on the multi-source data; fusing the multi-source data and judging a passenger boarding station according to a station arrival time window; performing preliminary inference on the alighting station based on a deterministic algorithm; establishing a deterministic algorithm to infer the passenger alighting station according to a travel chain assumption and passenger travel history; performing supplementary inference on the alighting station based on the integrated learning; introducing built environment and bus line external information for a journey that cannot be inferred by the deterministic algorithm; and establishing a multi-classification model based on an integrated learning two-layer stacking framework to further infer the passenger alighting station. The application realizes non-aggregated, station-level single-ticket bus passenger alighting station inference based on bus multi-source big data, can obtain accurate bus passenger flow space-time distribution characteristics, and provides a reference for bus operation organization optimization.

The invention belongs to the technical field of cold-chain highway freight, and discloses a cold-chain freight vehicle classification method based on travel task complexity. The method comprises the following steps: collecting cold chain truck track point data based on a Beidou satellite positioning system; preprocessing the collected cold chain truck track point data; an OD pair is constructed based on the preprocessed cold chain truck track point data; constructing a cold-chain freight vehicle trip chain based on the constructed OD pair; extracting trip chain features based on the constructed cold chain freight vehicle trip chain; obtaining a trip characteristic portrait of the cold chain freight vehicle based on the extracted trip chain characteristics; and constructing a vehicle multi-task scheduling optimization model based on the obtained travel characteristic portrait of the cold-chain freight vehicle. The invention aims to provide support for fine management and operation optimization of cold-chain logistics.

The invention relates to the technical field of new energy vehicle charging infrastructure planning and traffic big data application, and provides a high-speed service area charging station optimization site selection and constant volume method based on electric vehicle travel chain characteristics. A vehicle travel chain is reconstructed by using multi-source data such as a highway ETC gantry, and potential charging requirements of each time period and each service area are obtained by combining initial SOC random simulation and a segmented energy consumption model. A double-layer optimization model of upper-layer site selection (minimization of construction and driving energy consumption cost)-lower-layer constant volume (minimization of waiting time and pile construction cost) is established, constraints such as coverage, service capability and power grid access are introduced, and a particle swarm optimization algorithm is adopted for solving. Results show that the method can give consideration to service coverage and efficiency, reduce the total cost and queuing time, and improve the high-speed charging service level.

This invention discloses a method for analyzing the travel demand of public transport commuters based on spatiotemporal features, comprising: acquiring historical travel data of public transport passengers and constructing a personal travel behavior dataset; filtering commuter passenger groups based on travel frequency and analyzing spatiotemporal patterns of travel based on passenger travel time, entropy values ​​of travel stops, and travel chain similarity; extracting travel chain dimension features, time dimension features, and spatial dimension features related to the prediction target based on the spatiotemporal patterns of travel; constructing a feature matrix based on the extracted features; training the features in the feature matrix using a distributed gradient enhancement model; and predicting and outputting the travel demand of commuters based on the trained travel demand prediction model. This invention achieves efficient and accurate prediction of commuter passenger travel demand by integrating multidimensional travel features of commuters with machine learning algorithms.

The application discloses a kind of air and rail intermodal travel chain identification method based on ticket data.The method constructs passenger continuous travel sequence by fusing aviation and railway heterogeneous ticket data, uses travel mode identification difference value to screen out iron-air or air-iron candidate pair;Introduce field station-city mapping space constraint to remove pseudo intermodal of different place transfer;For the same city candidate pair, the minimum transfer time is instantiated according to hub space distance and connection characteristics in three scenarios, and the minimum value of the absolute physical upper limit of the travel chain and the statistical threshold of the pure transfer group fitted by the Gaussian mixture model is determined to determine the maximum transfer time, form a differentiated dynamic effective transfer time threshold interval, retain the candidate pair whose transfer time falls within the interval, and output air and rail intermodal travel chain.The present application makes up for the lack of space constraints, poor adaptability of fixed threshold and other shortcomings of existing methods, has the advantages of high identification accuracy, low misjudgment rate, lightweight algorithm and easy engineering expansion.

This invention provides a multi-task individual basic travel chain prediction method and system based on computational graphs, specifically comprising: Step 1: Data acquisition and construction of a data sample set, followed by preprocessing of the acquired data; Step 2: Graph structure modeling of spatial relationships between traffic zones, generating a weighted adjacency matrix at the region level, and using a multimodal graph representation learning model to generate graph embedding vectors and node embedding vectors for traffic zones; Step 3: Construction of a multi-task neural network model comprising a sequentially connected input layer, a shared layer, and multiple task output layers, with a context-aware module introduced in each task output layer to dynamically adjust sample-level weights and bias parameters; Step 4: Joint optimization of the learning model and the multi-task neural network model. This invention contributes to promoting the deep integration and practical application of artificial intelligence in the field of transportation engineering.