Road intersection sorting method and system

Abstract

The invention discloses a road intersection sorting method and system, and relates to the technical field of traffic. The method comprises the following steps: acquiring traffic track data; constructing an undirected weighted traffic network by taking the intersection as a point, taking the road section as an edge and taking the road section and the length as an edge weight based on the traffic trajectory data; extracting network features and traffic features in the undirected weighted traffic network, and inputting the network features and the traffic features to the trained intersection importance evaluation model to obtain an operation efficiency value of each intersection; according to the operation efficiency values, importance ranking is carried out on all the intersections; identifying a plurality of key intersections based on a sorting result, and calculating a congestion diffusion index of the traffic network by simulating sequential removal of the key intersections; and further evaluating the influence degree of each key intersection on the congestion propagation speed, range and recovery capability in the network.

Description

Technical Field

[0001] This application relates to the field of transportation technology, and in particular to a method and system for sequencing road intersections. Background Technology

[0002] Vehicle trajectory data, due to its ability to reflect the dynamic behavior of individual vehicles and reveal the potential correlations between vehicles, road segments, intersections, and even multiple intersections, has become the most widely used data type in the transportation field. Trajectory data has been extensively applied in various traffic scenarios, including vehicle route planning, queue length estimation, congestion prediction, traffic state classification, and intersection efficiency assessment. Therefore, accurately identifying these important intersections is of great significance for alleviating traffic pressure, improving urban traffic resilience, and predicting congestion.

[0003] Existing technologies for key node identification mainly include methods based on topology, traffic demand, and real-world travel trajectories. Different methods emphasize different aspects of node importance and evaluation criteria, leading to varying identification results. One paper proposes a method to address the problem of existing key node identification algorithms neglecting the relationships between nodes and their neighbors, identifying key nodes through weighted clustering coefficients and the influence of neighboring nodes on the network topology. Another paper comprehensively considers network structure indicators such as node degree and demand indicators such as hub passenger volume and operational intensity, using the K-shell hierarchical algorithm to evaluate the importance of key nodes. Yet another paper uses GPS data combined with network topology and traffic congestion index, employing the DWNodeRank algorithm to identify key nodes in the road network. A third paper constructs a primary and secondary indicator evaluation system, using a Gaussian mixture clustering model for hierarchical quantification to evaluate the operational efficiency of intersections.

[0004] However, existing technologies often rely on a single indicator of network topology or multiple indicators of traffic flow characteristics to evaluate the importance of nodes, failing to fully consider the impact of signal control on intersection efficiency and the interaction between nodes and their neighbors, thus leading to inaccurate identification of critical intersections. Summary of the Invention

[0005] Therefore, it is necessary to provide a method and system for sequencing road intersections to address the aforementioned technical problems.

[0006] The present invention adopts the following technical solution: This invention provides a method for sorting road intersections, including: Obtain traffic trajectory data; Based on traffic trajectory data, traffic operation characteristics are calculated; intersections in the traffic trajectory data are used as network nodes and road segments as connecting edges, and the weighting value of road segments is determined according to the road segment length and the number of lanes; an undirected weighted traffic network is constructed based on the network nodes, connecting edges, and the weighting value of road segments; and the network structure characteristics of intersections are extracted from the undirected weighted traffic network. Based on the network structure characteristics and traffic operation characteristics, calculate the operating efficiency value of each intersection; Based on the aforementioned operational efficiency values, the intersections are ranked by importance to obtain the ranking results of the intersection importance.

[0007] Preferably, it further includes: Based on the importance ranking of intersections, several key intersections are identified; a specified number of key intersections are removed sequentially from the undirected weighted traffic network graph, and the traffic congestion index of the undirected weighted traffic network after each removal of key intersections is calculated. Based on the traffic congestion index of the undirected weighted traffic network after each removal of a key intersection, the relationship between key intersections and the speed, scope, and recovery capacity of traffic congestion propagation is determined.

[0008] Preferably, an undirected weighted traffic network is constructed based on the weighted values ​​of network nodes, connecting edges, and road segments, specifically including: For any two adjacent network nodes, if there is a connecting edge between the two nodes, then the two network nodes are connected through the connecting edge and its corresponding weight value; if there is no connecting edge between the two nodes, then the two nodes are not connected. By connecting all network nodes with their corresponding weights and edges, an undirected weighted traffic network is obtained. The expression for the undirected weighted traffic network is: ; In the formula, G represents an undirected weighted traffic network. For the set of intersections, , Represents a set of road segments. Indicates connecting edges The weighted value, .

[0009] Preferably, the network structure features include the degree, strength, and weighted combination index of the nodes; The degree is the number of edges associated with an intersection node, used to characterize the number of direct connections between an intersection and other nodes in the network. The formula for calculating the degree is: ; In the formula, Indicates an intersection The degree, Indicates an intersection and Is there a connection indicator function? The intensity is a weighted form of degrees, representing the degree of association between any intersection node and other intersection nodes. It is used to characterize the information propagation and control capabilities of an intersection when congestion occurs. The formula for calculating the intensity is: ; In the formula, Represents a node The strength, Indicates an intersection To the intersection The length of the road segment between them Indicates an intersection To the intersection The number of lanes between them.

[0010] The formula for calculating the weighted combination index is as follows: ; In the formula, , Weights representing degree and intensity, , This indicates a weighted composite indicator.

[0011] Preferably, traffic operation characteristics include peak hour traffic volume, 95th percentile queue length, parking rate, maximum queue time index, average speed, and average delay.

[0012] Preferably, the operating efficiency value of each intersection is calculated using a super-efficiency DEA model, and the calculation formula is as follows: ; ; ; ; ; In the formula, This represents the operating efficiency value of the intersection. The smaller the value, the higher the operating efficiency of the intersection; This represents the total number of network structure features and traffic operation features, i.e., the number of input variables. , This represents a parameter used to adjust the input and output deviation. This represents input slack variables, indicating adjustments to the input variables and reflecting the underutilized portion of the input. These are output slack variables, representing adjustments to the output variables and reflecting the portion of the output that did not reach the ideal value. This represents the decision variable, indicating the weight of other intersections relative to the intersection being evaluated. Indicates the first The first intersection One input variable, Indicates the first being evaluated The first intersection One input variable, This represents the output decision variable, used to measure the output. Indicates the first The first intersection One output variable, Indicates other resource constraint variables, Indicates the first being evaluated Specific resource variables for each intersection.

[0013] Preferably, the formula for calculating the traffic congestion index is: ; In the formula, Represents network nodes i The congestion index; Indicates the section of road leading to the entrance. The traffic congestion index; Indicates the section of road leading to the entrance. Length, Indicates the section of road leading to the entrance. The number of lanes; This indicates the number of road segments that the network node enters.

[0014] This invention provides a road intersection sequencing system, comprising: A traffic trajectory data preprocessing module is used to acquire traffic trajectory data; calculate traffic operation characteristics based on the traffic trajectory data; use intersections of the traffic trajectory data as network nodes and road segments as connecting edges, and determine the weighting value of road segments based on the road segment length and the number of lanes; construct an undirected weighted traffic network based on the network nodes, connecting edges, and the weighting value of road segments; and extract the network structure features of intersections from the undirected weighted traffic network. The intersection importance assessment module calculates the operational efficiency value of each intersection based on the network structure characteristics and traffic operation characteristics; and ranks the intersections by importance based on the operational efficiency values ​​to obtain the intersection importance ranking result. The congestion propagation simulation module is used to identify multiple key intersections based on the importance ranking of intersections; sequentially remove a specified number of key intersections from the undirected weighted traffic network graph; calculate the traffic congestion index of the undirected weighted traffic network after each removal of key intersections; and determine the relationship between key intersections and the speed, range, and recovery capacity of traffic congestion propagation based on the traffic congestion index of the undirected weighted traffic network after each removal of key intersections.

[0015] The present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned road intersection sorting method.

[0016] The present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned road intersection sorting method.

[0017] The above-mentioned at least one technical solution adopted in this invention can achieve the following beneficial effects: The road intersection sorting method and system provided by this invention can effectively improve the accuracy of intersection identification and show significant advantages in congestion spread control, signal optimization, and dynamic adaptability. Specifically, it uses intersections in traffic trajectory data as network nodes and road segments as connecting edges; it determines the weighting value of connecting edges based on the length of the road segment and the number of lanes in the road segment; it constructs an undirected weighted traffic network based on network nodes, connecting edges, and the weighting value of connecting edges, and extracts multiple traffic operation features and network topology features from the undirected weighted traffic network. The operation features characterize the real-time load of nodes, and the topology features characterize the structural importance of nodes. The combination of the two can completely capture the two-dimensional information of traffic state and network structure. By calculating the intersection operation efficiency through the traffic operation features and network topology features extracted from the undirected weighted traffic network, it can effectively avoid the limitations of traditional methods that rely on subjective weights and single indicators, thereby accurately identifying the key intersections that have the greatest impact on traffic flow.

[0018] Furthermore, this invention quantifies the role of intersections in the congestion propagation process, verifies the impact of removing highly important intersections on network stability, and proves that the sorting method provided by this invention is superior to traditional methods in reducing congestion propagation speed, reducing the number of infected nodes, and accelerating network recovery. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 A flowchart illustrating a road intersection sorting method provided by the present invention; Figure 2 This invention provides an urban road trajectory distribution map for a road intersection sorting method. Figure 3 This invention provides a vehicle trajectory distribution range map for a road intersection sorting method. Figure 4 A diagram illustrating the intersection study range for a road intersection sequencing method provided by the present invention; Figure 5 A heat map of intersection congestion index provided by the present invention for a road intersection sorting method; Figure 6 A speed comparison chart before and after removing one important intersection in a road intersection sorting method provided by the present invention; Figure 7 A speed comparison diagram before and after removing two important intersections is provided in the present invention for a road intersection sorting method; Figure 8 The present invention provides a method and system for identifying key intersections, which includes a speed comparison chart before and after removing three important intersections. Figure 9 A comparison chart of infection numbers before and after removing one important intersection, provided by the present invention, for a road intersection sorting method; Figure 10 A comparison chart of infection numbers before and after removing two important intersections in a road intersection sorting method provided by this invention; Figure 11 A comparison chart of infection numbers before and after removing three important intersections in a road intersection sorting method provided by this invention; Figure 12 A comparison chart of the number of infections among different sorting algorithms for a road intersection sorting method provided by this invention; Figure 13 A comparison of infection numbers before and after considering network structure in a road intersection sorting method provided by this invention; Figure 14 A schematic diagram of a road intersection sequencing system provided by the present invention; Figure 15 A diagram of a computer device for implementing a road intersection sorting system provided by the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in the specification without creative effort are within the scope of protection of this application.

[0022] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0023] Figure 1 This is a schematic diagram of a road intersection sorting method and system flow according to the present invention, which specifically includes the following steps: S101: Obtain traffic trajectory data.

[0024] Specifically, the data used in this invention mainly includes vehicle GPS data and actual road network data. Vehicle GPS data was collected from June 9th to June 15th, 2020, with a sampling interval of 1 second for trajectory points. Data records include upload date, upload time, data source identifier, vehicle ID, longitude, latitude, speed, and direction angle. To ensure the accuracy of the analysis results, all data underwent rigorous data cleaning and preprocessing before use, including length filtering of trajectory segments, deduplication, speed anomaly filtering, road matching and removal of deviations, trajectory smoothing, and coordinate transformation to ensure data accuracy and continuity. Through these steps, invalid or abnormal data is filtered out, making the trajectory data more accurately reflect the behavior of vehicles in the road network.

[0025] S102: Calculate traffic operation characteristics based on traffic trajectory data; use intersections from the traffic trajectory data as network nodes and road segments as connecting edges, and determine the weighting value of road segments based on road segment length and number of lanes; construct an undirected weighted traffic network based on network nodes, connecting edges, and the weighting values ​​of road segments; and extract the network structure features of intersections from the undirected weighted traffic network. See the schematic diagram below. Figure 2 .

[0026] Optionally, for any two adjacent network nodes, if there is a connecting edge between the two nodes, then the two network nodes are connected through the connecting edge and its corresponding weight value; if there is no connecting edge between the two nodes, then the two nodes are not connected; after connecting all network nodes through connecting edges and their corresponding weight values, an undirected weighted traffic network is obtained; the expression of the undirected weighted traffic network is: ; In the formula, G represents an undirected weighted traffic network. For the set of intersections, , Represents a set of road segments. Representing an edge The weight, .

[0027] Optionally, the formula for calculating the weighted composite index is: ; In the formula, , Weights representing degree and intensity, , Eigenvalues ​​representing the structural features of the topography Represents a node The degree, Represents a node The intensity.

[0028] Optionally, the degree is the number of edges associated with the intersection node, used to characterize the number of direct connections between the intersection and other nodes in the network.

[0029] Optionally, the intensity is a weighted form of degrees, used to characterize the information dissemination and control capabilities of an intersection when congestion occurs.

[0030] Specifically, an undirected weighted complex network is established, where intersections are abstracted as points, road segments are abstracted as edges, and the number of lanes and length of road segments are used as weights; an undirected weighted complex network G is constructed. Network adjacency matrix Indicates; of which: Represents the set of intersections. Represents a set of road segments. Representing an edge The weight of the node With nodes There are edges connecting them. = ,otherwise =0.

[0031] Specifically, the study area comprised 15 traffic light-controlled intersections. This area experienced frequent traffic congestion during the study period, making it a typical example of urban traffic congestion research. The analysis of each intersection covered all approach road segments within 150 meters of the intersection center to ensure a comprehensive analysis of traffic flow around the intersection.

[0032] S103: Calculate the operating efficiency value of each intersection based on the network structure characteristics and traffic operation characteristics; rank the intersections by importance based on the operating efficiency values ​​to obtain the importance ranking result of the intersections.

[0033] Specifically, intersections are used as nodes in the network, and road segments are used as connecting edges to form an undirected weighted network. Edge weights are assigned, taking into account factors such as road segment length and number of lanes, to characterize the traffic connection strength between intersections. Network structural features, including degree and strength, are extracted to describe the network connectivity characteristics of intersections; traffic operation features are selected from key indicators such as peak hour traffic volume, parking rate, maximum queue time index, and average speed to comprehensively measure the traffic efficiency of intersections.

[0034] The formula for calculating the operating efficiency value of each intersection is as follows: ; ; ; ; ; In the formula, This represents an operational efficiency value, used to evaluate the relative efficiency of an intersection. The smaller the value, the higher the efficiency of the intersection. Indicates the number of input variables. , This represents a parameter used to adjust the input and output deviation. This represents input slack variables, indicating adjustments to the input variables and reflecting the underutilized portion of the input. These are output slack variables, representing adjustments to the output variables and reflecting the portion of the output that did not reach the ideal value. This represents the decision variable, indicating the weight of other intersections relative to the intersection being evaluated. Indicates the first The first intersection One input variable, Indicates the first being evaluated The first intersection One input variable, This represents the output decision variable, used to measure the output. Indicates the first The first intersection One output variable, Indicates other resource constraint variables, Indicates the first being evaluated Specific resource variables for each intersection.

[0035] Specifically, degree This refers to the number of edges associated with an intersection node, reflecting the number of direct connections between the intersection and other nodes in the network. Intersections with higher degrees have closer connections to other nodes and may affect more road segments during congestion. The formula is as follows:

[0036] ; in, Indicates an intersection The degree, Indicates an intersection and Is there a connection indicator function?

[0037] Definition: Strength. This is a weighted form of the degree calculation, taking into account the strength of the connection between the intersection and other nodes, such as the number of lanes and traffic volume. It further describes the information propagation and control capabilities of the intersection during congestion, as shown in the formula below:

[0038] ; in, Indicates an intersection To the intersection The length of the road segment between them Indicates an intersection To the intersection The number of lanes between them.

[0039] Furthermore, to comprehensively evaluate the network characteristics of intersections, this paper adopts a weighted average method, combining the degree and intensity indices to obtain a weighted combined index, as shown in the following formula: ; in, , Weights representing degree and intensity, , This represents the weighted combined index after merging. Represents a node The degree, Represents a node The intensity.

[0040] The operational efficiency of an intersection should be evaluated from the perspective of traffic demand, taking into account the influence of various factors such as time, space, and signal control. By extracting traffic flow parameters from multiple dimensions, the operational status of the intersection can be comprehensively evaluated. This invention selects five key indicators: peak hour traffic volume, 95th percentile queue length, parking rate, maximum queue time index, average speed, and average delay, to evaluate the operational efficiency of the intersection and analyze the changing trends of traffic operation status through these indicators. Definition: Peak hour traffic volume. Defined as the total number of vehicles passing the stop line at an intersection during peak traffic hours. The calculation method is as follows:

[0041] ; in, This indicates peak hour traffic volume. Indicates the first The number of vehicles that pass through the intersection stop line within a certain time period. This indicates the number of time periods within a peak hour.

[0042] Definition: Parking rate. Defined as the proportion of vehicles forced to stop due to red lights or other factors among the total number of vehicles passing through a certain road segment or intersection, such as... Figure 3 The image shows the vehicle trajectory distribution range. Stopped vehicles refer to vehicles whose speed drops to 0 during operation. The number of stops for the same vehicle during the same process can be counted repeatedly. The calculation method is as follows:

[0043] ; in, Indicates parking rate, Indicates the number of parking spaces. The parking rate, which represents the number of vehicles passing through, reflects the traffic flow at an intersection. A higher parking rate indicates that the intersection imposes more restrictions on traffic flow.

[0044] Definition: Average delay. Parking delay. Defined as the average additional time lost by a parked vehicle when forced to stop at an intersection or road segment due to red lights or other obstructions. The calculation method is as follows:

[0045] ; in, Indicates average delay. Indicates the time of free circulation. Indicates the number of vehicles that do not stop. This represents the total number of vehicles passing through. Stopping delays reflect the time lost by vehicles during traffic flow due to traffic control measures, and are an important indicator for evaluating the effectiveness of signal control.

[0046] Definition: Queue Time Index. Queuing time is defined as the ratio of queuing time to the signal control cycle time, where queuing time refers to the time from the first time a vehicle stops queuing until it passes through the intersection. The calculation method is as follows:

[0047] ; in, This represents the maximum queuing time index. Indicates the time when a vehicle first stops at a signalized intersection. Indicates the time at which vehicles pass the stop line at a signal-controlled intersection. This represents the unit cycle time, assuming that the average cycle time before a vehicle reaches the stop line, i.e., the signal timing duration, is constant. This paper uses the maximum queuing time index of each approach lane at the intersection as the overall queuing time index of the intersection, calculated using the following formula:

[0048] ; in: The maximum value of the queuing time index for each approach lane within a certain time interval is called the intersection queuing time index. Indicates the first time interval within a certain time interval. Queue time index at the entrance lane; Indicates the first time interval within a certain time interval. The moment when a vehicle first stops at the entrance lane; Indicates the first time interval within a certain time interval. The moment when vehicles entering the intersection pass the stop line at the intersection; Indicates the number of lanes at the intersection entrance.

[0049] Furthermore, to address the issue that traditional models determine intersection effectiveness by checking if the efficiency value is 1, but fail to differentiate which unit is optimal when multiple units have an efficiency value of 1, the ranking result is not the final output. Instead, it serves as input for subsequent evaluation of each intersection's ability to control congestion propagation within the network, providing support for network stability simulation and propagation situation analysis.

[0050] S104: Based on the operating efficiency value, sort the intersections by importance to obtain the importance ranking result of the intersections.

[0051] S105: Based on the importance ranking of the intersections, identify multiple key intersections; sequentially remove a specified number of key intersections from the undirected weighted traffic network graph, and calculate the traffic congestion index of the undirected weighted traffic network after each removal of key intersections.

[0052] The higher the operating efficiency value, the more critical the crossover is.

[0053] Optionally, the formula for calculating the traffic congestion index is: ; In the formula, Represents a node i The congestion index; Indicates the section of road leading to the entrance. The traffic congestion index; Indicates the section of road leading to the entrance. Length, Indicates the section of road leading to the entrance. The number of lanes; This indicates the number of road segments leading to the node.

[0054] Specifically, when calculating the node congestion index, the congestion situation of the road segments connected to each node must first be considered. To quantify the congestion index of a node, the weighted average of the road segment congestion indices is defined as the node congestion index, where the weights are determined by the length of the node's entrance lanes and the number of lanes.

[0055] S106: Based on the traffic congestion index of the undirected weighted traffic network after each removal of a key intersection, determine the relationship between key intersections and the speed, scope, and recovery capacity of traffic congestion propagation.

[0056] Specifically, this invention selects trajectory data from each intersection within the research scope from 17:00 to 20:00 for analysis, such as... Figure 4 The diagram shows the study area of ​​the intersection. The congestion index is calculated every 10 minutes, and the congestion index is classified according to Table 1. The congestion index calculation results for each intersection during the evening rush hour are shown below. Figure 5 As shown.

[0057] from Figure 5 It can be seen that peak-hour congestion is mainly concentrated at intersections numbered 8, 10, 11, 12, and 15. The congestion index of most intersections starts to be high around 5 PM, indicating that these are the peak congestion periods, which fluctuate in the following period. There are significant differences in congestion conditions among different intersections. Some intersections (such as intersections numbered 1, 2, and 3) have high congestion indices at multiple time periods, showing a continuous slow-moving trend. The congestion indices of intersections numbered 3, 6, and 7 are relatively low, and they are in a smooth state for most of the time. The congestion indices of intersections 10, 11, and 12 are mostly between 2 and 4 between 5 PM and 6 PM, indicating a congested state, and the index even approaches 2.6 at some time periods. These intersections have greater traffic pressure during peak hours.

[0058] Table 1. Congestion Level Classification Standards To identify key intersections in the transportation network, this study employs a super-efficiency model. Maximum queuing time, network characteristics, parking rate, average delay, and 95% queue length are used as input variables, while peak hour traffic flow and average speed are used as output variables. These key indicators allow the model to effectively measure the resource utilization efficiency and importance of each intersection to the transportation network.

[0059] As shown in Table 3, intersections C3, C4, C5, and C14, with efficiency values ​​exceeding 1, exhibit higher efficiency. This indicates that these intersections have more rational resource allocation and higher traffic flow under current conditions, potentially having a more significant impact on the overall traffic network operation. These high-efficiency intersections may also be more important because they serve as models in terms of network location, traffic distribution, and management strategies. Improving resource allocation at other low-efficiency intersections C1 and C7 can make the overall network operate more efficiently. The slack variables reveal the priorities for improvement, providing not only directions for efficiency enhancement but also the importance priorities of the intersections.

[0060] Intersection C1 has a large positive value for maximum queuing time, indicating that queuing time at this intersection affects its efficiency, suggesting that this intersection may be one of the bottlenecks in traffic congestion. Intersections C6 and C7 also have large slack values ​​for delay and queuing time, suggesting that the low efficiency of these intersections may negatively impact the overall network traffic efficiency. Therefore, these intersections should be prioritized in the optimization scheme. Based on the data of the slack variables, resources can be prioritized to allocate to these intersections that have a greater impact on the overall network smoothness, thereby improving the stability and flow of the entire network.

[0061] Table 2 Model Output Results Table 3. Intersection efficiency identification results As can be seen from the intersection efficiency values ​​in Table 3, intersections C9, C10, and C13 are the top three most important intersections. In the infectious disease model, by removing 1, 2, and 3 key intersections respectively, the changing trends of infection speed and the number of infected nodes were observed; Figure 6 The results show that in the initial time step after removing one key intersection, the infection rate fluctuates slightly, with the largest peak occurring in the first time step, but the amplitude is small. Compared with the baseline network, the infection transmission rate changes more smoothly after removing one intersection. The relatively small fluctuation in infection rate indicates that the removal of a single intersection has a relatively limited impact on the network structure; the infection rate before and after removing two important intersections is as follows: Figure 7 As shown, in the initial time steps after removing the two intersections, the fluctuation range of the infection rate increases, especially in the first 10 time steps, particularly the 1st and 10th time steps. The fluctuation range of the infection rate is larger than that of the baseline network, indicating that the network's stability in controlling infection spread has decreased, and the infection spread rate is more unstable; Figure 8Analysis shows that in the initial stage after removing the three intersections, the fluctuation range of the infection rate increased significantly, especially reaching its maximum peak in the first time step. After removing the three key intersections, the fluctuation of the infection transmission rate was the most dramatic.

[0062] Compared to the baseline network, removing three intersections resulted in the widest and longest fluctuation range in infection rate, indicating that the absence of key intersections severely weakened network stability, leading to more widespread and uncontrollable congestion spread. Sequentially removing important intersections revealed that removing one key intersection resulted in a slight but insignificant fluctuation in infection rate, suggesting that the absence of a single intersection had a relatively small impact on overall congestion spread. Removing two intersections caused more significant fluctuations in infection rate, indicating that the absence of multiple intersections began to significantly weaken the control over congestion spread. Removing three intersections had the greatest impact on infection rate, intensifying spread and significantly increasing the difficulty of control, demonstrating the crucial role of multiple important intersections in controlling congestion spread within the network. The fluctuation range of infection rate gradually increased after removing one, two, and three key intersections, indicating a gradual decrease in network stability. These results demonstrate that key intersections play a vital role in controlling the spread of infection and enhancing network stability; the more intersections removed, the greater the scope and intensity of infection spread, and the more significantly the network's mobility and stability are affected. This verifies the importance of critical intersections identified based on the DEA model in traffic management. Appropriate intersection control and optimization strategies can help effectively suppress the spread of congestion and improve the overall stability of the network.

[0063] Figures 9 to 11 This shows the trend of the number of infected nodes after removing 1 to 3 key intersections. The horizontal axis represents time steps, and the vertical axis represents the number of infected nodes. After removing one key intersection, the number of infected nodes in the baseline network rises rapidly in the first 10 time steps, peaking at about 8 nodes, while the number of infected nodes after removing the node is slightly lower, peaking at about 6 nodes.

[0064] This indicates that removing one critical intersection has a limited effect on suppressing infection spread; the intensity of infection spread decreases slightly, but not significantly. In subsequent time steps, the number of infected nodes in the baseline network gradually decreases, while the network recovers slightly faster after removing nodes, indicating that removing a small number of nodes has a positive effect on the network recovery process. After removing two critical intersections, the peak number of infected nodes in the baseline network is approximately 10, while the peak after removing nodes significantly decreases to approximately 7 nodes, indicating that removing two critical intersections effectively suppresses the intensity and scope of infection spread. After more than 20 time steps, the number of infected nodes in the network after removing nodes quickly returns to zero, while the recovery of the baseline network is relatively slow, indicating that removing two critical intersections has a more significant effect on suppressing infection spread. After removing three critical intersections, the peak number of infected nodes in the baseline network is approximately 8, while the peak after removing nodes further decreases to approximately 5 nodes. Compared to the cases of removing one and two intersections, removing three critical intersections significantly slows down the rate of infection spread and significantly accelerates the network recovery process. After more than 15 time steps, the number of infected nodes in the network after node removal has essentially reached zero, while the number of infected nodes in the baseline network is still decreasing. This indicates that as more critical intersections are removed, the network's control over infection propagation is significantly improved.

[0065] The overall trend from the three graphs shows a positive correlation between infection speed and the number of infected nodes; the faster the initial infection speed, the higher the number of infected nodes. As the number of removed critical intersections increases, the infection propagation speed decreases significantly, leading to a gradual reduction in the peak number of infections and accelerating the recovery process. Removing more critical intersections slows the infection propagation speed, reduces the infection range, and results in faster network recovery.

[0066] Furthermore, to verify the effectiveness of the intersection ranking algorithm based on the ranking model in suppressing traffic congestion propagation, this invention compares several commonly used network analysis algorithms, including degree centrality, betweenness centrality, PageRank, and Congestion Index. These algorithms provide different perspectives on identifying key network nodes, representing multi-dimensional analysis methods from the number of connections, mediation effects, probability propagation to congestion characteristics.

[0067] Figure 12 The study demonstrates how well each algorithm controls the number of infected nodes after removing key intersections, under the same network structure. Figure 13 We then further focus on the two superior algorithms, DEA and congestion index, to study the infection propagation situation after considering network structure information.

[0068] exist Figure 12In the study, the initial infection propagation trends of the five algorithms differed significantly. The degree centrality and betweenness centrality algorithms showed a rapid increase in the number of infected nodes with high peak values, indicating that these methods were ineffective in suppressing the initial spread of infection. In contrast, the DEA and congestion index algorithms showed lower infection peak values ​​and faster decay rates, demonstrating better congestion control and recovery speed. DEA and the congestion index performed better in identifying key intersections. To further verify the effectiveness of the optimal algorithm, Figure 13 The changes in the number of infected nodes after considering network structure information were compared separately between the model of this invention and the congestion index algorithm.

[0069] This step not only depicts how the absence of intersections alters the speed and scope of congestion propagation, but also identifies key nodes that play a decisive role in traffic resilience from a dynamic perspective by comparing network recovery times.

[0070] The results show that the algorithm has a more significant inhibitory effect in the early stages of infection propagation, with a lower peak number of infected nodes compared to the congestion index algorithm, and a faster recovery speed, eliminating the infection in a shorter time. This indicates that by combining network structure information, the algorithm has a greater advantage in identifying key intersections and improving network stability, and can more effectively control the spread of traffic congestion in the network.

[0071] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this invention.

[0072] The above describes a road intersection sorting method and system provided by one or more embodiments of the present invention. Based on the same idea, the present invention also provides a corresponding road intersection sorting system, such as... Figure 14 As shown.

[0073] Figure 14 A schematic diagram 1400 of a key intersection identification method and system provided by the present invention includes: The trajectory data preprocessing module 1401 acquires traffic trajectory data; based on the traffic trajectory data, it constructs an undirected weighted traffic network by using intersections as points, road segments as edges, and road segments and lengths as edge weights. The intersection importance assessment module 1402 is used to extract the topology features in the undirected weighted traffic network and input them into the trained intersection importance assessment model to obtain the operating efficiency value of each intersection; based on the operating efficiency value, each intersection is ranked by importance. The congestion propagation simulation module 1403 identifies multiple key intersections based on the importance of each intersection; it then removes a specified number of key intersections in sequence and calculates the traffic congestion index after removing the key intersections; based on the traffic congestion index after removing the intersections, it determines the impact of the key intersections on the traffic congestion situation.

[0074] For specific limitations regarding the road intersection sorting method system, please refer to the limitations of the road intersection sorting method described above, which will not be repeated here. Each module in the aforementioned road intersection sorting system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0075] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 A method for sorting road intersections is provided.

[0076] The present invention also provides Figure 15 The schematic diagram of the computer device shown is as follows: Figure 15 As shown, at the hardware level, this computer device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above. Figure 1 A method for sorting road intersections is provided.

[0077] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

Claims

1. A method of sequencing a road intersection, characterized by, The method comprises the following steps: obtaining traffic trajectory data; calculating traffic operation characteristics according to the traffic trajectory data; taking intersections of the traffic trajectory data as network nodes and road segments as connecting edges, and determining the weighted values of the road segments according to the lengths and the number of lanes of the road segments; constructing a weighted undirected traffic network according to the network nodes, the connecting edges and the weighted values of the road segments, and extracting network structure characteristics of the intersections from the weighted undirected traffic network; calculating operation efficiency values of the intersections according to the network structure characteristics and the traffic operation characteristics; ranking the importance of the intersections according to the operation efficiency values to obtain an importance ranking result of the intersections.

2. A method of sequencing a road intersection, characterized by, The method further comprises the following steps: determining a plurality of key intersections according to the importance ranking result of the intersections; removing a specified number of key intersections from the weighted undirected traffic network in sequence, and calculating traffic congestion indexes of the weighted undirected traffic network after each removal of a key intersection; determining the relationship between the key intersections and the traffic congestion propagation speed, propagation range and traffic network recovery ability according to the traffic congestion indexes of the weighted undirected traffic network after each removal of a key intersection.

3. A method of sequencing a road intersection as claimed in claim 1 wherein, The construction of the weighted undirected traffic network according to the network nodes, the connecting edges and the weighted values of the road segments specifically comprises the following steps: for any two adjacent network nodes, if there is a connecting edge between the two nodes, the two network nodes are connected through the connecting edge and the corresponding weighted value; if there is no connecting edge between the two nodes, the two network nodes are not connected; after connecting all the network nodes through the connecting edges and the corresponding weighted values, a weighted undirected traffic network is obtained; wherein, the expression of the weighted undirected traffic network is: ； where G is an undirected weighted transportation network, is a set of intersections, , is a set of links, is a weighted value of a connecting edge . .

4. A method of sequencing a road intersection as claimed in claim 3 wherein, the network structure characteristics comprise the degree, the strength and the weighted combination index of the nodes; the degree is the number of edges associated with the intersection node, which is used to represent the number of direct connections of the intersection with other nodes in the network, and the calculation formula of the degree is: ； wherein denotes the degree of the intersection , denotes the degree of the intersection and an indicator function whether there is a connection; the strength is a weighted form of the degree, which represents the correlation degree between any intersection node and other intersection nodes, and is used to represent the information propagation ability and control ability of the intersection when congestion occurs, and the calculation formula of the strength is: ； wherein denotes the strength of a node , denotes the length of a road segment between intersections , denotes the number of lanes between intersections , denotes the number of lanes between intersections ; the calculation formula of the weighted combination index is: ； wherein , denotes the weight of the degree and the intensity, , denotes the weighted combination indicator.

5. A method of sequencing a road intersection as claimed in claim 1, wherein, the traffic operation characteristics comprise the peak hour traffic volume, the 95% percentile queue length, the parking rate, the maximum queue time index, the average speed and the average delay.

6. A method of sequencing a road intersection as defined in claim 1, wherein, The operation efficiency value of each intersection is calculated by an super-efficiency DEA model, and the calculation formula is: ； ； ； ； ； In the formula, represents the operating efficiency value of the intersection, The smaller the value of represents the total number of network structure characteristics and traffic operation characteristics, that is, the number of input variables, , represents the parameter for adjusting the input and output deviation, represents the input slack variable, which represents the adjustment of the input variable, and embodies the part of the input that is not fully utilized, is the output slack variable, which represents the adjustment of the output variable, and embodies the part of the output that does not reach the ideal value, represents the decision variable, which represents the weight of other intersections relative to the intersection being evaluated, represents the th input variable of the th intersection, represents the th input variable of the th intersection being evaluated, represents the output decision variable, which is used to measure the output part, represents the th output variable of the th intersection, represents other resource constraint variables, represents the specific resource variable of the th intersection being evaluated.

7. A method of sequencing a road intersection as defined in claim 2, wherein, the calculation formula of the traffic congestion index is: ； wherein denotes a congestion index of a network node i ; denotes a congestion index of an import road segment ; denotes a congestion index of an import road segment ; denotes a length of an import road segment ; denotes a number of import road segments of a network node.

8. A road intersection sequencing system characterized by, The method comprises the following steps: a traffic trajectory data preprocessing module is configured to obtain traffic trajectory data; calculating traffic operation characteristics according to the traffic trajectory data; taking intersections of the traffic trajectory data as network nodes and road segments as connecting edges, and determining the weighted values of the road segments according to the lengths and the number of lanes of the road segments; constructing a weighted undirected traffic network according to the network nodes, the connecting edges and the weighted values of the road segments; extracting network structure characteristics of the intersections from the weighted undirected traffic network; an intersection importance evaluation module is configured to calculate operation efficiency values of the intersections according to the network structure characteristics and the traffic operation characteristics; According to the operation efficiency value, the intersections are ranked in importance to obtain an importance ranking result of the intersections; The congestion propagation simulation module is configured to determine a plurality of key intersections according to the importance ranking result of the intersections; The specified number of key intersections are sequentially removed from the undirected weighted traffic network graph, and a traffic congestion index of the undirected weighted traffic network after each removal of the key intersection is calculated; According to the traffic congestion index of the undirected weighted traffic network after each removal of the key intersection, a relationship between the key intersection and the traffic congestion propagation speed, propagation range and traffic network recovery ability is determined.

Images