Urban traffic dispatching method and platform driven by multi-dimensional data

Abstract

The present invention discloses a multi-dimensional data-driven urban traffic dispatching method and platform, which relates to the technical field of traffic dispatching. The method includes: obtaining urban traffic facility distribution information, determining a set of key traffic areas; deploying fixed source detection equipment, obtaining a set of multi-dimensional traffic operation data streams in key areas and a set of multi-dimensional real-time data streams of road traffic; obtaining a set of traffic operation related indicators, performing preprocessing and indicator feature extraction; performing congestion prediction analysis, obtaining urban traffic congestion parameters, performing dispatch analysis on urban traffic congestion parameters, and determining a set of traffic dispatching strategy solutions; performing dispatching simulation and effect comparison and optimization, and performing traffic dispatching optimization control. The present invention solves the technical problems of insufficient traffic dispatching and low operating efficiency in the prior art, and achieves the technical effect of accurately grasping traffic conditions and improving traffic dispatching and operating efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic dispatching, and in particular to a multi-dimensional data driven urban traffic dispatching method and platform. Background Art

[0002] With the continuous acceleration of urbanization and the continuous increase in the number of cars, urban traffic congestion is becoming increasingly serious. Traditional urban traffic management systems have many limitations in dealing with complex and changeable traffic conditions. In terms of traffic data collection, in the past, they often relied on a single type of detection equipment, such as relying solely on fixed source detection equipment. This resulted in the collected data being not comprehensive enough and unable to accurately reflect the dynamic changes in the entire urban traffic. In addition, data from different sources lacked effective integration, and the data format and standards were not unified, making it difficult to conduct comprehensive analysis. In terms of traffic dispatch decision-making, in the past, dispatch strategies were mainly formulated based on experience, lacking scientific quantitative analysis and evaluation, and after the dispatch strategy was implemented, it was difficult to accurately evaluate and optimize its effects.

[0003] The existing technology has technical problems of insufficient traffic scheduling and low operating efficiency. Summary of the invention

[0004] The present application provides a multi-dimensional data-driven urban traffic scheduling method and platform, which is used to solve the technical problems of insufficient traffic scheduling and low operating efficiency in the prior art.

[0005] In view of the above problems, the present application provides a multi-dimensional data-driven urban traffic scheduling method and platform.

[0006] In a first aspect of the present application, a multidimensional data-driven urban traffic dispatching method is provided, the method comprising:

[0007] The urban traffic facility distribution information is obtained through the urban traffic management system, and the urban traffic facility distribution information is three-dimensionally modeled to generate an urban traffic facility spatial model. The key area of ​​the urban traffic facility spatial model is identified to determine a set of key traffic areas; fixed source detection equipment is deployed on the set of key traffic areas, and a set of multi-dimensional traffic operation data streams in the key areas is collected and obtained through the fixed source detection equipment, and a mobile source detection equipment is used to record and obtain a multi-dimensional real-time data stream of road traffic; a set of traffic operation related indicators is obtained, and based on the traffic operation related indicator set, the multi-dimensional traffic operation data stream set in the key areas and the multi-dimensional real-time data stream of road traffic are pre-processed and the indicator feature is extracted to obtain a set of urban traffic operation indicator features; congestion prediction analysis is performed based on the set of urban traffic operation indicator features to obtain urban traffic congestion parameters, and scheduling analysis is performed on the urban traffic congestion parameters to determine a set of traffic scheduling strategy solutions; scheduling simulation and effect comparison and optimization are performed based on the set of traffic scheduling strategy solutions to obtain a target traffic scheduling strategy solution, and traffic scheduling optimization control is performed through the target traffic scheduling strategy solution.

[0008] The second aspect of the present application provides a multi-dimensional data-driven urban traffic dispatching platform, the platform comprising:

[0009] A module for determining a set of key traffic areas, the module is used to obtain the distribution information of urban traffic facilities through the urban traffic management system, perform three-dimensional modeling on the distribution information of urban traffic facilities, generate a spatial model of urban traffic facilities, identify key areas of the spatial model of urban traffic facilities, and determine a set of key traffic areas; a multi-dimensional data stream acquisition module, the module is used to deploy fixed source detection equipment on the set of key traffic areas, collect and obtain a set of multi-dimensional traffic operation data streams in key areas through the fixed source detection equipment, and simultaneously use mobile source detection equipment to record and obtain multi-dimensional real-time data streams of road traffic; a module for acquiring a set of urban traffic operation indicator characteristics, the module is used to acquire traffic operation related indicators A set of traffic operation-related indicator sets is used to preprocess the key area multidimensional traffic operation data flow set and the road traffic multidimensional real-time data flow, and extract indicator features to obtain a city traffic operation indicator feature set; a traffic scheduling strategy solution set determination module is used to perform congestion prediction analysis based on the city traffic operation indicator feature set, obtain urban traffic congestion parameters, perform scheduling analysis on the urban traffic congestion parameters, and determine a traffic scheduling strategy solution set; a traffic scheduling optimization control module is used to perform scheduling simulation and effect comparison and optimization based on the traffic scheduling strategy solution set, obtain a target traffic scheduling strategy solution, and perform traffic scheduling optimization control through the target traffic scheduling strategy solution.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The urban traffic management system is used to obtain the distribution information of urban traffic facilities, and the three-dimensional modeling of the distribution information of urban traffic facilities is performed to generate the spatial model of urban traffic facilities, identify key areas, and determine the set of key traffic areas; fixed source detection equipment is deployed on the set of key traffic areas to collect and obtain the set of multi-dimensional traffic operation data streams in key areas, and at the same time obtain the multi-dimensional real-time data stream of road traffic; a set of traffic operation related indicators is obtained, and pre-processing and indicator feature extraction are performed to obtain a set of urban traffic operation indicator features; congestion prediction analysis is performed based on the set of urban traffic operation indicator features to obtain urban traffic congestion parameters, and scheduling analysis is performed on the urban traffic congestion parameters to determine a set of traffic scheduling strategy solutions; scheduling simulation and effect comparison and optimization are performed based on the set of traffic scheduling strategy solutions to obtain the target traffic scheduling strategy solution, and traffic scheduling optimization and control are performed through the target traffic scheduling strategy solution. The technical effect of accurately grasping the traffic conditions and improving the efficiency of traffic scheduling and operation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic flow chart of a multi-dimensional data-driven urban traffic scheduling method provided in an embodiment of the present application.

[0014] Figure 2 A schematic diagram of the structure of a multi-dimensional data-driven urban traffic dispatching platform provided in an embodiment of the present application.

[0015] Explanation of the reference numerals: a module for determining a set of key traffic areas 10 , a module for acquiring a multi-dimensional data stream 20 , a module for acquiring a set of characteristics of urban traffic operation indicators 30 , a module for determining a set of traffic scheduling strategy solutions 40 , and a module for optimizing and controlling traffic scheduling 50 . DETAILED DESCRIPTION

[0016] This application provides a multi-dimensional data-driven urban traffic scheduling method and platform to solve the technical problems of insufficient traffic scheduling and low operating efficiency in the prior art.

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Embodiment 1, as Figure 1 As shown, the present application provides a multi-dimensional data-driven urban traffic scheduling method, the method comprising:

[0019] S100: Acquire urban traffic facility distribution information through an urban traffic management system, perform three-dimensional modeling on the urban traffic facility distribution information to generate an urban traffic facility space model, identify key areas of the urban traffic facility space model, and determine a set of key traffic areas.

[0020] Specifically, the urban traffic management system stores a wealth of geographic information and traffic facility data. First, the system is used to obtain information on the distribution of urban traffic facilities, which covers various attributes of roads such as location, length, width, number of lanes, and the distribution of facilities such as traffic lights, traffic signs, and surveillance cameras. Then, the three-dimensional modeling technology is used to transform the acquired information into an intuitive spatial model of urban traffic facilities. With the help of three-dimensional modeling software, roads, bridges, traffic lights and other facilities are accurately presented based on the actual geographical location and spatial relationship. Such a model can observe traffic facilities from different perspectives to better discover potential problems and spaces that can be optimized. Then, the established model is deeply analyzed, and by considering factors such as historical traffic data, road connectivity and importance, key areas such as large traffic flow, easy congestion, and frequent accidents are identified, and finally the set of key traffic areas is determined.

[0021] S200: Deploy fixed source detection equipment on the set of key traffic areas, collect and obtain a set of multi-dimensional traffic operation data streams in the key areas through the fixed source detection equipment, and simultaneously use mobile source detection equipment to record and obtain multi-dimensional real-time data streams of road traffic.

[0022] Specifically, the processing of key traffic area collection is crucial. First, the fixed source detection equipment is deployed in these key areas. These fixed source data detection equipment have specific installation locations. They are firmly fixed at key points on the road. For example, the ring coil detector is buried under the road surface. When a vehicle passes by, it will cause the change of the coil magnetic field, so that the vehicle passing can be accurately detected and the number and speed of the vehicle can be recorded. Microwave detectors use microwave technology to sense the existence and movement of vehicles by emitting microwave signals and receiving reflected signals. It can monitor data stably without being affected by weather and light conditions. There are also cameras, which are distributed in various key locations. Through image recognition technology, not only the number and speed of vehicles can be obtained, but also more detailed information such as the type of vehicle, license plate number, etc. can be identified. These fixed source detection equipment work together to continuously monitor and record passing vehicles, integrate the various data collected, and form a multi-dimensional traffic operation data stream collection in key areas, covering data in multiple dimensions such as vehicle flow, vehicle speed, and vehicle model.

[0023] At the same time, mobile source detection equipment also plays an important role. Take the vehicle-mounted GPS device as an example. It continuously updates the location information and driving speed data as the vehicle moves. When the vehicle-mounted GPS devices on many vehicles work at the same time, they can reflect the real-time status of road traffic from multiple mobile perspectives. The high-definition camera and infrared thermal imager carried by the drone collect data from a unique aerial perspective. The high-definition camera can clearly capture the overall traffic picture of the road. The infrared thermal imager can obtain traffic information by detecting the thermal radiation of objects under special circumstances, such as at night or in bad weather conditions. These mobile source detection devices record and transmit data, and finally form a multi-dimensional real-time data stream of road traffic, which complements the data collected by the fixed source detection equipment and provides strong support for a comprehensive and accurate understanding of road traffic conditions.

[0024] S300: Obtain a set of traffic operation related indicators, and preprocess and extract indicator features from the key area multidimensional traffic operation data stream set and the road traffic multidimensional real-time data stream based on the traffic operation related indicator set to obtain a city traffic operation indicator feature set.

[0025] Specifically, we first obtain a set of traffic operation related indicators including traffic volume, vehicle speed, congestion index, delay time, etc. These indicators are like a "barometer" of traffic conditions, reflecting the actual situation of traffic operation from different angles. Then, based on this set of indicators, we process the multidimensional traffic operation data stream set of key areas and the multidimensional real-time data stream of road traffic. On the one hand, we perform data preprocessing. When facing abnormal values ​​in the data set, such as unreasonable data generated by equipment failure or special events, we need to carefully identify and use reasonable methods to process them, such as deleting abnormal values ​​or replacing them with average values. For data missing problems, we use linear interpolation or historical data average values ​​to fill in the missing data according to the characteristics and relevance of the data. At the same time, we need to unify various data formats from different devices and systems into a standard format. On the other hand, we extract indicator features. Through time series analysis, we calculate the statistics of data such as traffic volume over a period of time and analyze the changing trend of vehicle speed. We extract spatial features in combination with the spatial structure of the road network, and study the relationship between different road sections and the impact of key areas. We also perform correlation analysis to explore the internal connection between various indicators, and finally obtain a set of urban traffic operation indicator features that can comprehensively and accurately reflect the status of urban traffic operation.

[0026] S400: Perform congestion prediction analysis based on the urban traffic operation indicator feature set to obtain urban traffic congestion parameters, perform scheduling analysis on the urban traffic congestion parameters, and determine a set of traffic scheduling strategy solutions.

[0027] Specifically, the neural network model is used to take the characteristic set of urban traffic operation indicators as the input of the neural network. These characteristic data, such as traffic volume, vehicle speed, congestion index, delay time, etc., are input into the first layer of the neural network after preprocessing and normalization. The neurons in this layer will perform preliminary feature extraction and conversion on these data. Then, the neural network performs deeper information processing and pattern recognition through the hidden layer. In the hidden layer, neurons gradually learn the potential relationship between traffic indicator characteristics and traffic congestion through complex connections and weight adjustments. For example, when the traffic volume continues to increase over a period of time, the vehicle speed gradually decreases, and the congestion index increases, the neural network will judge the probability and degree of possible congestion based on these changing trends and their interactions. After multiple training and adjustments, the neural network can more accurately perform congestion prediction analysis based on the input traffic operation indicator characteristics, thereby obtaining urban traffic congestion parameters including congestion time, location, scope, duration and congestion degree.

[0028] Next, these congestion parameters are used for scheduling analysis, and the congestion parameters are input into another neural network module again. The model will evaluate various possible traffic scheduling measures based on different congestion situations and corresponding historical scheduling strategy effect data. For example, for congestion in a specific area, the model will analyze the potential impact of different scheduling strategies such as adjusting traffic light time, setting temporary traffic control, and optimizing bus routes on alleviating congestion. Through the simulation and evaluation of various strategies, a set of multiple feasible traffic scheduling strategy solutions is finally determined, providing a basis for the subsequent selection of the best scheduling solution.

[0029] S500: Based on the set of traffic scheduling strategies, scheduling simulation and effect comparison are performed to obtain a target traffic scheduling strategy, and traffic scheduling optimization management and control is performed through the target traffic scheduling strategy.

[0030] Specifically, we first work on the set of traffic dispatching strategies. On the one hand, we conduct dispatching simulation. By building a simulation environment, we input each traffic dispatching strategy in the set into it one by one. In this simulation environment, it will simulate the real traffic scene and the impact of various strategies on the traffic system after implementation, such as the vehicle's driving trajectory, the traffic conditions at the intersection, and the changes in road congestion. On the other hand, we compare and optimize the effects, determine evaluation indicators such as average travel time, vehicle delay time, and road capacity, and then extract these indicator data corresponding to each solution from the simulation results, and compare and analyze the effects of different solutions. Through comprehensive evaluation and screening, we determine the optimal target traffic dispatching strategy from many solutions. Finally, we apply this target solution to the actual traffic dispatching optimization and control. Through a series of measures such as adjusting traffic signals, optimizing line settings, and guiding vehicle diversion, we can make the operation of the traffic system more efficient, alleviate traffic congestion, and improve the operation quality of the entire urban traffic.

[0031] In a possible manner, step S100 further includes:

[0032] Step S110: Acquire information on factors affecting urban traffic, where the information on factors affecting urban traffic includes traffic roads, sections prone to congestion, regional density distribution, and key facilities.

[0033] Step S120: Performing regional grid division on the urban transportation facility space model according to a preset grid division size to obtain a transportation facility space grid model.

[0034] Step S130: performing association analysis on each grid area in the traffic facility space grid model based on the urban traffic influencing factor information to obtain a grid area traffic factor parameter set.

[0035] Step S140: Based on the grid area traffic factor parameter set, the traffic facility space grid model is screened and marked as a key area to determine the traffic key area set.

[0036] Specifically, obtaining information on factors affecting urban traffic covers many key elements. In terms of traffic roads, it includes road types, such as main roads, secondary roads or branch roads, as well as the number of lanes and the location distribution of intersections. These factors play a fundamental role in the capacity and direction of traffic flow. Information on congested sections directly reflects the bottleneck areas in traffic operations. Regional density distribution is also crucial, among which population density and job density distribution will greatly affect the generation and distribution of traffic demand. For example, densely populated areas usually have stronger traffic demand. In addition, areas where key facilities such as schools, hospitals, and commercial centers are located have also become key traffic areas because they attract a large number of people and vehicles.

[0037] The preset grid division size is determined according to the regional size of the urban transportation facility spatial model and its distribution density. The determination of this size is very critical. It needs to fully consider the characteristics of the spatial model to ensure the rationality and effectiveness of the division. According to the determined preset grid division size, the urban transportation facility spatial model is regionally grid-divided to obtain the transportation facility spatial grid model. Through this grid division method, the originally complex traffic area can be divided into relatively regular and easy-to-analyze small grid units.

[0038] Based on the information on urban traffic influencing factors obtained above, correlation analysis is performed on each grid area in the traffic facility space grid model. For example, a comprehensive analysis is performed based on factors such as the road conditions around a grid area, population density, whether it is close to congested roads, and whether it contains key facilities, so as to obtain a set of traffic factor parameters for the grid area, which may include vehicle flow, passenger flow, traffic demand intensity, etc.

[0039] Based on the obtained grid area traffic factor parameter set, the key areas of the traffic facility space grid model are screened and marked. For example, if the traffic factor parameters of a grid area meet certain conditions, such as the traffic volume exceeds a certain threshold, is located near a congested road section, or is close to key facilities with high traffic demand, it will be marked as a key area. After such a screening and marking process, the key traffic area set is finally determined, providing a clear target area for subsequent targeted traffic analysis and scheduling.

[0040] In a possible manner, step S140 further includes:

[0041] Step S141: constructing a traffic network diagram by taking each grid area in the traffic facility space grid model as a node and the traffic flow direction between each grid area as an edge.

[0042] Step S142: Based on the grid area traffic factor parameter set, the centrality and betweenness of the traffic network diagram are calculated in sequence to obtain a grid area node centrality set and a grid area node betweenness set.

[0043] Step S143: performing weighted calculation on the grid area node centrality set and the grid area node betweenness set to obtain a grid area node criticality coefficient set.

[0044] Step S144: Based on the grid area node criticality coefficient set, the transportation facility space grid model is screened and marked for critical areas to determine the transportation critical area set.

[0045] Specifically, in order to more intuitively analyze the relationship between the grid areas in the spatial grid model of transportation facilities, each grid area is regarded as a node, and the traffic flow direction between the grid areas is regarded as an edge, so that a traffic network diagram is constructed. This traffic network diagram is like a map of urban traffic, which clearly shows the traffic connection between the grid areas.

[0046] We start to conduct an in-depth quantitative analysis of the traffic network graph. We start with the calculation of centrality. Based on the set of grid area traffic factor parameters, for each node corresponding to a grid area, we evaluate the degree of direct connection between it and other nodes in the network. If a node has direct edges connected to many other nodes, it means that it has a high degree of direct connection in the traffic network. At the same time, we also need to consider the importance of the congestion impact of the node in the area. For example, if a grid area node itself is prone to congestion, and its congestion will have a greater impact on other nodes connected to it, then its importance in congestion propagation is relatively high, and the corresponding centrality will also be high. Then we calculate the betweenness. For each node in the traffic network graph, we analyze its role in the connection process of other nodes. If the traffic flow between two nodes that are not directly connected needs to pass through a specific node to achieve connection, then the intermediary role of this node is very obvious. For example, some grid area nodes located at the intersection of major traffic routes, many traffic flows converge here from different directions and then disperse to other directions. Such nodes have a high degree of connection in the entire traffic network, and their betweenness is also relatively large. By performing such centrality and betweenness calculations on all nodes in the traffic network diagram, we can finally obtain the grid area node centrality set and the grid area node betweenness set. These two sets record the centrality and betweenness values ​​corresponding to each grid area node, respectively, providing a quantitative basis and foundation for further analysis and screening of key areas.

[0047] First, the grid area node centrality set and grid area node betweenness set are used as basic data. Centrality reflects the degree of direct connection between a node and other nodes and its influence in the propagation of congestion, while betweenness reflects the degree of passage of a node in the process of connecting other nodes and the intermediary role it plays. Then, determine the appropriate weight distribution. The distribution of weights needs to comprehensively consider the characteristics and actual needs of the transportation network. For example, if it is believed that centrality is relatively more important in measuring the criticality of a node, a slightly higher weight can be given to centrality; conversely, if the role of betweenness in the transportation network is more prominent, the weight of betweenness can be appropriately increased. Next, a weighted calculation is performed. For each grid area node, its centrality value and betweenness value are multiplied by the corresponding weight according to the determined weight, and then the two are added. Through this calculation method, the contribution of the two factors of centrality and betweenness to the criticality of the node is comprehensively considered. After performing such weighted calculations on all grid area nodes in turn, a set of grid area node criticality coefficients is obtained. Each coefficient in this set represents the comprehensive criticality of the corresponding grid area node. The higher the coefficient value, the more important and critical the node is in the transportation network, and the greater the impact on the operation and congestion of the entire transportation network. Such a set of criticality coefficients provides a strong quantitative basis for the subsequent accurate screening of key areas, which helps to carry out more scientific transportation planning and management.

[0048] Based on the critical coefficient set of grid area nodes, the key areas of the spatial grid model of transportation facilities are screened and marked. If the critical coefficient of a grid area node reaches a certain threshold or meets specific conditions, it will be marked as a key area. Through such a screening and marking process, the set of key transportation areas is finally determined. These identified key areas will become the focus of subsequent transportation planning and management, so that targeted measures can be taken to optimize traffic conditions and improve traffic efficiency.

[0049] In a possible manner, step S300 further includes:

[0050] Step S310: Obtain a data preprocessing procedure, which includes time series arrangement, data cleaning and data normalization.

[0051] Step S320: preprocessing the key area multidimensional traffic operation data stream set and the road traffic multidimensional real-time data stream according to the data preprocessing procedure to obtain a standard traffic operation multidimensional data stream set.

[0052] Step S330: Determine a traffic data source dimension set according to the fixed source detection device and the mobile source detection device, and obtain a dimensional data feature extraction program set based on the traffic data source dimension set.

[0053] Step S340: Based on the traffic operation related indicator set, multi-indicator feature extraction is performed on the standard traffic operation multidimensional data stream set according to the dimensional data feature extraction program set to obtain the urban traffic operation indicator feature set.

[0054] Specifically, obtain the data preprocessing program. The data preprocessing program here includes several key parts. Time series arrangement is used to organize the time sequence of data to ensure the orderliness of data in the time dimension; data cleaning is to remove noise, outliers and erroneous data in the data to ensure the accuracy and reliability of the data; data normalization is to unify the data to the same scale and range to avoid the adverse effects of data of different dimensions on subsequent analysis.

[0055] According to the above data preprocessing procedures, the key area multi-dimensional traffic operation data flow set and the road traffic multi-dimensional real-time data flow are processed, and the time sequence of the data is made clear through time sequence arrangement. After data cleaning, those data that do not meet the requirements are eliminated, and then the data is standardized by data normalization. After these processes, a standard traffic operation multi-dimensional data flow set is obtained, which provides a high-quality data foundation for subsequent analysis.

[0056] Fixed source detection equipment and mobile source detection equipment are the key basis for determining the set of traffic data source dimensions. Fixed source detection equipment usually generates specific types of data, such as loop coil detectors, microwave detectors, etc., which provide information such as traffic flow and vehicle speed. Taking microwave detectors as an example, they obtain data by emitting microwave signals and receiving reflected signals. These data have certain characteristics and rules, while mobile source detection equipment, such as vehicle-mounted GPS devices and cameras carried by drones, also generate different types of data. Vehicle-mounted GPS devices mainly provide vehicle location and speed information, while cameras carried by drones can obtain image data. According to the data types generated by these different devices, the set of traffic data source dimensions can be determined. For example, the infrared data dimension uses infrared rays to detect the presence and movement of vehicles or pedestrians. These data reflect the thermal radiation characteristics of objects; the image data dimension, the images obtained by the camera contain rich visual information, such as the shape, color, and position of the vehicle; the microwave data dimension, such as the data obtained by the microwave detector mentioned above, reflects the interaction between microwave signals and vehicles. Based on the determined set of traffic data source dimensions, because data of different dimensions have different characteristics and forms of expression, corresponding feature extraction methods are required. For the infrared data dimension, it is necessary to extract the characteristics of the thermal radiation characteristics of the object, temperature distribution, etc.; for the image data dimension, image recognition technology and target detection algorithm are used to extract the number, type, driving direction and other characteristics of the vehicle; for the microwave data dimension, it is necessary to extract relevant traffic features such as vehicle speed and distance based on the characteristics and change rules of microwave signals. Therefore, according to the characteristics and needs of different dimensions, a corresponding set of dimension data feature extraction programs is obtained. Each program is optimized for data of a specific dimension so that useful feature information can be accurately and efficiently extracted to provide strong support for subsequent traffic analysis and decision-making.

[0057] The set of traffic operation related indicators plays a key guiding role. This set covers indicators such as traffic flow, speed, occupancy, emergencies, etc. Each indicator reflects the different characteristics and conditions of traffic operation. First, facing the standard traffic operation multidimensional data stream set, which contains rich traffic data that has been preprocessed, we start working according to the set of dimensional data feature extraction programs. Because data of different dimensions have different characteristics and structures, the corresponding feature extraction programs will perform special processing for these differences. For the indicator of traffic flow, statistics and analysis are performed on data from different sources. For example, from the data obtained from fixed source detection equipment, the traffic flow is calculated based on the number of vehicles passing through a certain detection point in a specific time period; from the image data of mobile source detection equipment, the traffic flow is estimated through image recognition and counting technology. In terms of vehicle speed, multi-source data is combined, such as the vehicle speed data directly measured by fixed source detection equipment, or the vehicle speed is inferred by calculation from the location and time information of mobile source detection equipment. For occupancy rate, by analyzing the proportion of space occupied by vehicles on the road, relevant information can be extracted from data of different dimensions, such as judging the occupancy of the road from the sensing data of fixed source detection equipment, or identifying the area covered by vehicles from image data to estimate the occupancy rate. For emergencies, such as traffic accidents or road construction, comprehensive analysis of abnormal scenes in image data, sudden changes in fixed source detection equipment data, and abnormal fluctuations in other related data can be used to timely discover and extract the feature information of these emergencies. Through the feature extraction procedures of these different dimensional data, fine extraction work is carried out in accordance with the way associated with each traffic operation related indicator, and finally the extracted feature information is integrated to form a feature set of urban traffic operation indicators. This set comprehensively and accurately reflects the actual operation status and characteristics of urban traffic, and provides a solid foundation for subsequent traffic analysis, prediction and decision-making.

[0058] In a possible manner, step S400 further includes:

[0059] Step S410: Acquire a city traffic congestion database, extract indicator features from the city traffic congestion database according to the traffic operation related indicator set, and obtain a sample data set of city traffic operation indicator features.

[0060] Step S420: Perform traffic congestion clustering training based on the urban traffic operation index characteristic sample data set to construct a traffic operation congestion multidimensional evaluation model, wherein the traffic operation congestion multidimensional evaluation model includes a congestion index evaluation model, a congestion area evaluation model, and a congestion duration evaluation model.

[0061] Step S430: performing congestion prediction analysis on the urban traffic operation index feature set based on the traffic operation congestion multidimensional evaluation model, and outputting congestion index evaluation information, congestion area evaluation information and congestion duration information.

[0062] Step S440: Obtain the urban traffic congestion parameters according to the congestion index evaluation information, the congestion area evaluation information and the congestion duration information.

[0063] Specifically, the first thing to obtain is the urban traffic congestion database, which is an extremely important resource. It contains the historical traffic congestion data of the city in the past period of time. These data record various congestion conditions that have occurred in urban traffic at different times, different locations and under different conditions. The set of traffic operation related indicators covers key factors such as traffic flow, vehicle speed, occupancy rate, etc., which are closely related to traffic congestion conditions. When operating the database according to the set of traffic operation related indicators, the program will filter and extract. For the traffic flow indicator, the number of vehicles passing through a specific time period and a specific location will be found from the database; for the speed indicator, the vehicle speed data of the corresponding road section and time point will be extracted; the occupancy rate indicator involves the acquisition of information on the proportion of roads occupied by vehicles. Through such sophisticated operations, characteristic data that conform to the set of traffic operation-related indicators are extracted from the urban traffic operation congestion database. These data constitute the urban traffic operation indicator characteristic sample data set. This data set is a refinement and summary of historical traffic congestion data. It retains the characteristic information most relevant to traffic congestion, and provides representative and targeted samples for subsequent model training and analysis, making traffic congestion analysis and prediction based on these data more accurate and reliable, and laying a solid foundation for a deep understanding of the laws and characteristics of urban traffic congestion.

[0064] The work is carried out with the sample data set of urban traffic operation index characteristics as the cornerstone. For the traffic congestion clustering training of the neural network model, the sample data set is first input into the neural network. The neural network is composed of a large number of neuron nodes, which interact with each other through complex connections and weights. During the training process, the sample features in the data set will be propagated and processed in the network. In terms of the congestion index evaluation model, the neural network will learn the complex mapping relationship between various traffic operation indicators and congestion index in the sample data. For example, indicators such as traffic flow, vehicle speed, and road occupancy rate enter the network as information in the input layer, and are processed layer by layer and weighted by neurons in the hidden layer. There can be multiple hidden layers, and each hidden layer is constantly extracting higher-level and more abstract features. Finally, a value representing the congestion index is obtained at the output layer. By continuously adjusting the weights and biases of the network, the congestion index output by the model is as close as possible to the congestion index in the actual sample, thereby improving the accuracy of the model.

[0065] When constructing the congested area assessment model, the neural network also uses the location information, traffic flow distribution and other relevant indicators in the sample data. The network will learn and classify according to the characteristic patterns of different areas. For example, if the traffic volume in a certain area is concentrated and the speed is low, and the surrounding road connections are complex, after analysis and training by the network, it can accurately determine whether the area belongs to a congested area. At the same time, through repeated learning of samples from different areas, the model can gradually improve the accuracy of judging congestion in different types of areas.

[0066] In the congestion duration assessment model, the neural network analyzes the changes in sample data in the time series. Considering the dynamic changes in traffic conditions over time, the network learns the characteristics of traffic indicators at the beginning of congestion, the trend of indicator changes during congestion, and the characteristic transformation at the end of congestion. By learning from these time series data, the model can predict how long congestion may last based on the current traffic status and similar patterns in historical data.

[0067] After repeated training and optimization, a multi-dimensional assessment model for traffic congestion was finally constructed, which includes a congestion index assessment model, a congestion area assessment model, and a congestion duration assessment model. This model can comprehensively utilize various traffic operation indicators and historical data to conduct a comprehensive and accurate assessment of traffic congestion conditions from multiple dimensions, providing a scientific and reliable basis for subsequent traffic management and scheduling.

[0068] Based on the constructed multi-dimensional assessment model for traffic congestion, a congestion prediction analysis is performed on the newly acquired set of urban traffic operation indicator features. Each sub-model in the model plays a role respectively. The congestion index assessment model outputs congestion index assessment information, which can intuitively reflect the current level of traffic congestion; the congestion area assessment model outputs congestion area assessment information, accurately locating the location and scope of congestion; the congestion duration assessment model outputs congestion duration information, informing the length of time that traffic congestion may last.

[0069] The three aspects of information, namely, congestion index assessment information, congestion area assessment information and congestion duration information, are integrated to form the parameters of urban traffic congestion. Congestion index assessment information is a key quantitative indicator that intuitively reflects the degree of traffic congestion. If the congestion index is high, it means that the traffic congestion is serious, the vehicles on the road are moving slowly, and the traffic flow is close to or exceeds the road's capacity. By analyzing the congestion index assessment information, we can clearly understand the intensity level of overall traffic congestion. Congestion area assessment information accurately locates the specific location of traffic congestion, and it clarifies which sections and areas are experiencing traffic congestion. For example, whether the main roads in the city are congested, or whether there is congestion at a certain intersection or on the roads around a specific commercial area, this information can be obtained through the congestion area assessment information. Congestion duration information provides key data on the duration of traffic congestion. It provides the time span from the beginning to the possible end of traffic congestion. Understanding the duration of congestion helps predict the development trend of traffic congestion and assess its impact on urban traffic and residents' travel. For example, if congestion in a certain area lasts for a long time, a longer-term traffic management strategy or infrastructure improvement is needed. By combining these three aspects of information, comprehensive urban traffic congestion parameters can be obtained. These parameters provide an accurate description of traffic congestion conditions, which can better understand the nature and characteristics of traffic congestion, and thus formulate more effective traffic management measures, planning decisions and travel arrangements to alleviate traffic congestion and improve the operating efficiency of urban traffic.

[0070] In a possible manner, step S400 further includes:

[0071] Step S450: construct a traffic dispatching strategy library, perform matching analysis based on the urban traffic congestion parameters and the traffic dispatching strategy library, and obtain a traffic congestion dispatching strategy set.

[0072] Step S460: Determine a set of scheduling parameter optimization algorithms according to the traffic congestion scheduling strategy set.

[0073] Step S470: Based on the scheduling parameter optimization algorithm set, the traffic congestion scheduling strategy set is respectively subjected to scheduling optimization analysis to determine the traffic scheduling strategy solution set.

[0074] Specifically, first of all, the strategies in the strategy library include signal light adjustment strategies. For example, according to the changes in traffic flow in different periods, the cycle duration and phase sequence of the signal lights can be reasonably adjusted to optimize the traffic efficiency of the intersection. There are also road control strategies, such as setting up temporary traffic control measures in specific congested areas or construction areas, such as one-way streets and no-go zones, to guide vehicle diversion and alleviate local traffic pressure. In addition, it also includes bus line optimization strategies, which adjust the direction, station settings or departure frequency of bus lines to better meet the travel needs of citizens and reduce the impact of buses on road traffic. After having such a rich strategy library, the next step is to match and analyze based on urban traffic congestion parameters. Urban traffic congestion parameters comprehensively reflect the status of traffic congestion, including the degree of congestion (mild, moderate or severe congestion), the area of ​​congestion (in the city center, main roads or secondary roads, etc.) and the duration of congestion (short-term peak-hour congestion or long-term continuous congestion). These congestion parameters are compared and analyzed one by one with each strategy in the strategy library. For the signal light adjustment strategy, if the congestion in a certain area is caused by the unbalanced traffic flow at the intersection, it is possible to consider matching the signal light timing in the area with the corresponding signal light adjustment strategy to see whether the traffic flow distribution can be improved by adjusting the signal light. For the road control strategy, if a certain section of road is seriously congested due to construction or sudden accidents, find a road control strategy suitable for the area to see whether temporary traffic control can be set to guide vehicles to avoid the congested area. For the bus line optimization strategy, if the congestion in a certain area is related to the concentrated travel of buses, evaluate whether the traffic pressure can be alleviated by adjusting the bus line or departure frequency. After such a comprehensive and detailed matching analysis, those strategies that are compatible with the current traffic congestion parameters are selected and combined together to obtain a set of traffic congestion scheduling strategies. This set contains all traffic scheduling strategies that meet the current traffic congestion conditions, providing a rich selection and a solid foundation for further optimization and implementation.

[0075] After having a set of traffic congestion scheduling strategies, it is necessary to determine a suitable set of scheduling parameter optimization algorithms to further improve the effectiveness of these strategies. For the neural network algorithm, it has unique advantages in this process. The neural network algorithm has strong learning and adaptive capabilities. First of all, for different traffic congestion scheduling strategies, the neural network algorithm can conduct in-depth analysis and understanding of the various parameters involved. For example, for the signal light timing strategy, the neural network can take historical traffic data and current traffic congestion conditions as input, and establish a complex relationship model between traffic flow, congestion degree and signal light timing parameters by learning a large amount of data. During the training process, the neural network continuously adjusts its weights and biases to minimize the prediction error and find the optimal combination of signal light timing parameters. For road control strategies, the neural network can optimize the parameters of control measures (such as the scope of the control area, the length of control time, etc.) according to factors such as traffic flow patterns, road structures and surrounding environments in different regions. It can learn the changing trends of traffic conditions under different parameter combinations, so as to determine the most effective road control parameter settings. In this way, according to the different strategies in the traffic congestion scheduling strategy set, the neural network algorithm is selected and applied to construct an algorithm set that can effectively optimize the parameters of various scheduling strategies, providing powerful tools and methods support for subsequent traffic scheduling optimization analysis, so that the traffic scheduling strategy can more accurately adapt to the actual traffic conditions and improve the efficiency and effectiveness of traffic management.

[0076] The set of scheduling parameter optimization algorithms plays a key role. For each strategy in the set of traffic congestion scheduling strategies, the corresponding appropriate optimization algorithm is used to carry out scheduling optimization analysis. For example, for the signal light timing strategy, the algorithm continuously tries different combinations of parameters such as signal light cycles and phase differences, and finds the optimal parameter combination among many possibilities based on evaluation indicators such as intersection traffic volume, queue length, and traffic efficiency; for the road control strategy, the algorithm simulates and evaluates in different combinations of control area ranges, time lengths, methods, and other parameters to determine the best road control plan; for the bus line optimization strategy, the route direction, station settings, departure frequency and other parameters are adjusted and evaluated by comprehensively considering factors such as passenger demand, operating costs, and road congestion. Finally, through repeated optimization of each scheduling strategy, a set of traffic scheduling strategy solutions is determined, each of which is the best choice under specific traffic conditions and constraints, providing a reliable basis for the implementation of traffic scheduling.

[0077] In a possible manner, step S500 further includes:

[0078] Step S510: Use digital twin technology to simulate the urban traffic facility space model and the urban traffic operation congestion database to build a traffic operation scheduling simulation twin.

[0079] Step S520: Use the traffic operation scheduling simulation twin to perform scheduling simulation on the traffic scheduling strategy solution set respectively to obtain a multi-strategy solution simulation scheduling parameter set.

[0080] Step S530: Evaluate and compare the multi-strategy simulation scheduling parameter set through a scheduling effect evaluation function to find the best solution and determine the target traffic scheduling strategy.

[0081] Specifically, digital twin technology has become a key tool that combines the spatial model of urban traffic facilities and the urban traffic congestion database for simulation. The spatial model of urban traffic facilities provides information on the physical structure and layout of urban traffic, including the location and form of roads, intersections, key facilities, etc. The urban traffic congestion database contains historical traffic congestion data, reflecting the status of traffic operation under different circumstances. Through digital twin technology, a traffic operation and scheduling simulation twin that is highly similar to the real traffic system is built in a virtual environment. This twin can simulate the operating laws and dynamic changes of the real traffic system, providing a reliable virtual test platform for subsequent evaluation and optimization of scheduling strategies.

[0082] The traffic operation scheduling simulation twin is used to simulate the scheduling of each plan in the traffic scheduling strategy set. The traffic scheduling strategy set is a series of scheduling strategies obtained through screening and optimization in the previous steps. When each plan is input into the simulation twin, it will simulate the operation of the traffic system under the strategy according to the parameters and rules in the plan, and generate corresponding scheduling simulation parameters. These parameters may include changes in traffic flow, increase in vehicle speed, reduction in congestion area, shortening of congestion duration, etc. By simulating each plan, a set of simulation scheduling parameters for multiple strategy plans is obtained, which provides a data basis for subsequent evaluation and comparison.

[0083] The dispatch effect evaluation function is introduced to evaluate and compare the simulation dispatch parameter sets of multiple strategy solutions. The dispatch effect evaluation function quantitatively evaluates the simulation dispatch parameters of each solution according to specific evaluation indicators and weights. The evaluation indicators include the degree of improvement in traffic flow, the reduction in average travel time, and the increase in road capacity. The evaluation function calculates and compares the parameters of different solutions to find the solution with the best effect. After repeated evaluation and screening, the target traffic dispatch strategy solution is finally determined. This solution is considered to be the most effective dispatch strategy to improve traffic conditions and alleviate congestion after comprehensive consideration of various factors and indicators, providing clear guidance for the actual implementation of traffic dispatch.

[0084] Embodiment 2, based on the same inventive concept as the multi-dimensional data driven urban traffic dispatching method in the above embodiment, Figure 2 As shown, the present application provides a multi-dimensional data-driven urban traffic dispatching platform, and the platform and method embodiments in the present application embodiments are based on the same inventive concept. The platform includes:

[0085] The key traffic area set determination module 10 is used to obtain the urban traffic facility distribution information through the urban traffic management system, perform three-dimensional modeling on the urban traffic facility distribution information, generate an urban traffic facility space model, identify key areas of the urban traffic facility space model, and determine the key traffic area set.

[0086] The multi-dimensional data stream acquisition module 20 is used to deploy fixed source detection equipment on the key traffic area set, collect and acquire the key area multi-dimensional traffic operation data stream set through the fixed source detection equipment, and simultaneously use mobile source detection equipment to record and acquire the multi-dimensional real-time data stream of road traffic.

[0087] The urban traffic operation indicator feature set acquisition module 30 is used to obtain a traffic operation related indicator set, and based on the traffic operation related indicator set, the key area multidimensional traffic operation data stream set and the road traffic multidimensional real-time data stream are preprocessed and indicator features are extracted to obtain the urban traffic operation indicator feature set.

[0088] The traffic scheduling strategy solution set determination module 40 performs congestion prediction analysis based on the urban traffic operation indicator feature set, obtains urban traffic congestion parameters, performs scheduling analysis on the urban traffic congestion parameters, and determines a traffic scheduling strategy solution set.

[0089] The traffic scheduling optimization control module 50 performs scheduling simulation and effect comparison based on the traffic scheduling strategy solution set, obtains the target traffic scheduling strategy solution, and performs traffic scheduling optimization control through the target traffic scheduling strategy solution.

[0090] Furthermore, the critical traffic area set determination module 10 further includes:

[0091] The urban traffic influencing factor information acquisition unit is used to acquire urban traffic influencing factor information, and the urban traffic influencing factor information includes traffic roads, congested sections, regional density distribution and key facilities.

[0092] The traffic facility space grid model acquisition unit is used to perform regional grid division on the urban traffic facility space model according to a preset grid division size to obtain the traffic facility space grid model.

[0093] A grid area traffic factor parameter set acquisition unit is configured to perform association analysis on each grid area in the traffic facility space grid model based on the urban traffic influencing factor information to acquire a grid area traffic factor parameter set.

[0094] A critical traffic area set determination unit is configured to perform critical area screening and marking on the traffic facility space grid model based on the grid area traffic factor parameter set to determine the critical traffic area set.

[0095] Furthermore, the critical traffic area set determination unit further includes:

[0096] A traffic network diagram construction unit is used to construct a traffic network diagram by taking each grid area in the traffic facility space grid model as a node and the traffic flow direction between each grid area as an edge.

[0097] A grid area node set acquisition unit is provided, wherein the grid area node set acquisition unit sequentially calculates the centrality and betweenness of the traffic network diagram based on the grid area traffic factor parameter set to obtain a grid area node centrality set and a grid area node betweenness set.

[0098] A grid area node critical coefficient set acquisition unit is used to perform weighted calculation on the grid area node centrality set and the grid area node betweenness set to obtain a grid area node critical coefficient set.

[0099] A key area screening and marking unit is used to perform key area screening and marking on the traffic facility space grid model based on the grid area node criticality coefficient set to determine the traffic key area set.

[0100] Furthermore, the urban traffic operation index feature set acquisition module 30 further includes:

[0101] A data preprocessing program acquisition unit is used to acquire a data preprocessing program, wherein the data preprocessing program includes time series arrangement, data cleaning and data normalization.

[0102] A standard traffic operation multidimensional data stream set acquisition unit is used to preprocess the key area multidimensional traffic operation data stream set and the road traffic multidimensional real-time data stream according to the data preprocessing procedure to obtain a standard traffic operation multidimensional data stream set.

[0103] A dimensional data feature extraction program set acquisition unit is used to determine the traffic data source dimension set according to the fixed source detection device and the mobile source detection device, and obtain the dimensional data feature extraction program set based on the traffic data source dimension set.

[0104] A city traffic operation indicator feature set acquisition unit, wherein the city traffic operation indicator feature set acquisition unit performs multi-indicator feature extraction on the standard traffic operation multidimensional data stream set based on the traffic operation related indicator set and according to the dimensional data feature extraction program set to obtain the city traffic operation indicator feature set.

[0105] Furthermore, the traffic scheduling strategy solution set determination module 40 also includes:

[0106] The urban traffic operation index characteristic sample data set acquisition unit is used to acquire an urban traffic operation congestion database, extract index characteristics from the urban traffic operation congestion database according to the traffic operation related index set, and obtain an urban traffic operation index characteristic sample data set.

[0107] A traffic congestion multidimensional assessment model construction unit is provided, wherein the traffic congestion multidimensional assessment model construction unit performs traffic congestion clustering training based on the urban traffic operation index characteristic sample data set to construct a traffic congestion multidimensional assessment model, wherein the traffic congestion multidimensional assessment model includes a congestion index assessment model, a congestion area assessment model and a congestion duration assessment model.

[0108] A congestion prediction and analysis unit performs congestion prediction and analysis on the urban traffic operation index feature set based on the traffic operation congestion multi-dimensional evaluation model, and outputs congestion index evaluation information, congestion area evaluation information and congestion duration information.

[0109] The urban traffic congestion parameter acquisition unit is used to obtain the urban traffic congestion parameter according to the congestion index evaluation information, the congestion area evaluation information and the congestion duration information.

[0110] Furthermore, the traffic scheduling strategy solution set determination module 40 also includes:

[0111] A traffic congestion scheduling strategy set acquisition unit is used to construct a traffic scheduling strategy library, and obtain a traffic congestion scheduling strategy set based on matching analysis between the urban traffic congestion parameters and the traffic scheduling strategy library.

[0112] A scheduling parameter optimization algorithm set determination unit is used to determine a scheduling parameter optimization algorithm set according to the traffic congestion scheduling strategy set.

[0113] A scheduling optimization analysis unit is used to perform scheduling optimization analysis on the traffic congestion scheduling strategy set based on the scheduling parameter optimization algorithm set to determine the traffic scheduling strategy solution set.

[0114] Furthermore, the traffic dispatch optimization control module 50 also includes:

[0115] A traffic operation scheduling simulation twin building unit, wherein the traffic operation scheduling simulation twin building unit is used to simulate the urban traffic facility space model and the urban traffic operation congestion database by using digital twin technology to build a traffic operation scheduling simulation twin.

[0116] A multi-strategy scheme simulation scheduling parameter set acquisition unit is used to use the traffic operation scheduling simulation twin to perform scheduling simulation on the traffic scheduling strategy scheme set respectively to obtain a multi-strategy scheme simulation scheduling parameter set.

[0117] The target traffic scheduling strategy scheme determining unit is used to evaluate, compare and optimize the multi-strategy scheme simulation scheduling parameter set through a scheduling effect evaluation function to determine the target traffic scheduling strategy scheme.

[0118] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0119] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0120] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A multi-dimensional data-driven urban traffic dispatching method, characterized in that: The method comprises: Acquire urban traffic facility distribution information through the urban traffic management system, perform three-dimensional modeling on the urban traffic facility distribution information, generate an urban traffic facility space model, identify key areas of the urban traffic facility space model, and determine a set of traffic key areas; Deploy fixed source detection equipment on the key traffic area set, collect and obtain the key area multi-dimensional traffic operation data flow set through the fixed source detection equipment, and use mobile source detection equipment to record and obtain the multi-dimensional real-time data flow of road traffic; Acquire a set of traffic operation related indicators, and preprocess and extract indicator features from the key area multidimensional traffic operation data stream set and the road traffic multidimensional real-time data stream based on the set of traffic operation related indicators to obtain a set of urban traffic operation indicator features; Perform congestion prediction analysis based on the urban traffic operation indicator feature set to obtain urban traffic congestion parameters, perform scheduling analysis on the urban traffic congestion parameters, and determine a set of traffic scheduling strategy solutions; Based on the set of traffic scheduling strategies, scheduling simulation and effect comparison are performed to obtain the target traffic scheduling strategy, and traffic scheduling optimization control is performed through the target traffic scheduling strategy; The step of determining a set of key traffic areas includes: Acquire information on factors affecting urban traffic, including traffic roads, sections prone to congestion, regional density distribution, and key facilities; Performing regional grid division on the urban transportation facility space model according to a preset grid division size to obtain a transportation facility space grid model; Based on the information of the influencing factors of urban traffic, correlation analysis is performed on each grid area in the traffic facility space grid model to obtain a grid area traffic factor parameter set; Based on the grid area traffic factor parameter set, the traffic facility space grid model is screened and marked for key areas to determine the traffic key area set; The determining of the set of critical traffic areas includes: Taking each grid area in the traffic facility spatial grid model as a node and the traffic flow direction between each grid area as an edge, a traffic network diagram is constructed; Based on the grid area traffic factor parameter set, the centrality and betweenness of the traffic network diagram are calculated in sequence to obtain a grid area node centrality set and a grid area node betweenness set, including: based on the grid area traffic factor parameter set, for each node corresponding to the grid area, evaluating the degree of direct connection between it and other nodes in the network to obtain the grid area node centrality, for each node in the traffic network diagram, analyzing the role it plays in the connection process of other nodes to obtain the grid area node betweenness; Performing weighted calculation on the grid area node centrality set and the grid area node betweenness set to obtain a grid area node criticality coefficient set; The transportation facility space grid model is screened and marked for key areas based on the grid area node criticality coefficient set to determine the transportation key area set.

2. The multidimensional data-driven urban traffic dispatching method according to claim 1, characterized in that: The obtained urban traffic operation index feature set includes: Obtaining a data preprocessing procedure, wherein the data preprocessing procedure includes time series arrangement, data cleaning, and data normalization; Preprocessing the key area multidimensional traffic operation data stream set and the road traffic multidimensional real-time data stream according to the data preprocessing program to obtain a standard traffic operation multidimensional data stream set; Determine a set of traffic data source dimensions according to the fixed source detection device and the mobile source detection device, and obtain a set of dimensional data feature extraction programs based on the set of traffic data source dimensions; Based on the traffic operation related indicator set, multi-indicator feature extraction is performed on the standard traffic operation multidimensional data stream set according to the dimensional data feature extraction program set to obtain the urban traffic operation indicator feature set.

3. The multidimensional data-driven urban traffic dispatching method according to claim 1, characterized in that: The obtaining of the urban traffic congestion parameters comprises: Acquire a city traffic congestion database, extract indicator features from the city traffic congestion database according to the traffic operation related indicator set, and obtain a sample data set of city traffic operation indicator features; Based on the sample data set of characteristic urban traffic operation indexes, traffic congestion clustering training is performed to construct a multidimensional evaluation model for traffic operation congestion, wherein the multidimensional evaluation model for traffic operation congestion includes a congestion index evaluation model, a congestion area evaluation model, and a congestion duration evaluation model; Based on the multi-dimensional evaluation model for traffic congestion, the urban traffic operation index feature set is subjected to congestion prediction analysis, and congestion index evaluation information, congestion area evaluation information, and congestion duration information are output; The urban traffic congestion parameter is obtained according to the congestion index evaluation information, the congestion area evaluation information and the congestion duration information.

4. The multidimensional data-driven urban traffic dispatching method according to claim 1, characterized in that: The method of determining a set of traffic dispatching strategy solutions includes: Constructing a traffic dispatch strategy library, performing matching analysis based on the urban traffic congestion parameters and the traffic dispatch strategy library, and obtaining a traffic congestion dispatch strategy set; Determining a set of scheduling parameter optimization algorithms according to the set of traffic congestion scheduling strategies; Based on the scheduling parameter optimization algorithm set, the traffic congestion scheduling strategy set is respectively subjected to scheduling optimization analysis to determine the traffic scheduling strategy solution set.

5. The multi-dimensional data driven urban traffic dispatching method according to claim 3, characterized in that: The method of obtaining the target traffic dispatching strategy includes: Using digital twin technology to simulate the urban traffic facility space model and the urban traffic operation congestion database, and build a traffic operation scheduling simulation twin; The traffic operation scheduling simulation twin is used to perform scheduling simulation on the traffic scheduling strategy scheme set respectively to obtain a multi-strategy scheme simulation scheduling parameter set; The multi-strategy scheme simulation scheduling parameter set is evaluated, compared and optimized through a scheduling effect evaluation function to determine the target traffic scheduling strategy scheme.

6. A multi-dimensional data-driven urban traffic dispatching platform, characterized by: The platform includes: A key traffic area set determination module, the key traffic area set determination module is used to obtain urban traffic facility distribution information through an urban traffic management system, perform three-dimensional modeling on the urban traffic facility distribution information, generate an urban traffic facility space model, identify key areas on the urban traffic facility space model, and determine a key traffic area set; A multi-dimensional data stream acquisition module, wherein the multi-dimensional data stream acquisition module is used to deploy fixed source detection equipment on the set of key traffic areas, collect and acquire a set of multi-dimensional traffic operation data streams in the key areas through the fixed source detection equipment, and simultaneously use mobile source detection equipment to record and acquire multi-dimensional real-time data streams of road traffic; An urban traffic operation indicator feature set acquisition module, the urban traffic operation indicator feature set acquisition module is used to acquire a traffic operation related indicator set, and pre-process the key area multi-dimensional traffic operation data stream set and the road traffic multi-dimensional real-time data stream based on the traffic operation related indicator set and extract indicator features to obtain an urban traffic operation indicator feature set; A traffic dispatching strategy solution set determination module, wherein the traffic dispatching strategy solution set determination module performs congestion prediction analysis based on the urban traffic operation indicator feature set, obtains urban traffic congestion parameters, performs dispatch analysis on the urban traffic congestion parameters, and determines a traffic dispatching strategy solution set; A traffic dispatch optimization control module, which performs dispatch simulation and effect comparison based on the traffic dispatch strategy set, obtains a target traffic dispatch strategy, and performs traffic dispatch optimization control through the target traffic dispatch strategy; The key traffic area set determination module also includes: An urban traffic influencing factor information acquisition unit, the urban traffic influencing factor information acquisition unit is used to acquire urban traffic influencing factor information, the urban traffic influencing factor information includes traffic roads, congested sections, regional density distribution and key facilities; A traffic facility space grid model acquisition unit, the traffic facility space grid model acquisition unit is used to perform regional grid division on the urban traffic facility space model with a preset grid division size to acquire the traffic facility space grid model; A grid area traffic factor parameter set acquisition unit, which performs association analysis on each grid area in the traffic facility space grid model based on the urban traffic influencing factor information to acquire a grid area traffic factor parameter set; A critical traffic area set determination unit, which performs critical area screening and marking on the traffic facility space grid model based on the grid area traffic factor parameter set to determine the critical traffic area set; The critical traffic area set determination unit further includes: A traffic network diagram construction unit, the traffic network diagram construction unit is used to construct a traffic network diagram by taking each grid area in the traffic facility space grid model as a node and the traffic flow direction between each grid area as an edge; A grid area node set acquisition unit, wherein the grid area node set acquisition unit sequentially calculates the centrality and betweenness of the traffic network diagram based on the grid area traffic factor parameter set to obtain a grid area node centrality set and a grid area node betweenness set; A grid area node critical coefficient set acquisition unit, the grid area node critical coefficient set acquisition unit is used to perform weighted calculation on the grid area node centrality set and the grid area node betweenness set to obtain a grid area node critical coefficient set; A key area screening and marking unit is used to perform key area screening and marking on the traffic facility space grid model based on the grid area node criticality coefficient set to determine the traffic key area set.

Images