Congestion bottleneck point determination method, device, computer equipment and storage medium

By generating the road condition distribution map of each section in the road network topology map, selecting multiple congested road sections and calculating their contribution scores, the accuracy and effectiveness of determining road congestion bottleneck points in the existing technology are solved, and efficient Traffic management.

CN113487869BActive Publication Date: 2024-11-29TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202110786798.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-07-12
Publication Date
2024-11-29
Estimated Expiration
2041-07-12

AI Technical Summary

Technical Problem

The prior art has problems with accuracy and effectiveness when determining the bottleneck points of road congestion, and it is difficult to quickly and accurately find out the direct causes of traffic congestion.

Method used

By obtaining road conditions data at different times and road network topology maps built based on road network data, the road conditions distribution maps of each section in the road network topology map are generated at different times, and the target road sections that have been congested in at least two periods are selected to determine their matching Contribution scores of the congestion cluster and determine the congestion bottleneck points based on the scores.

Benefits of technology

It realizes accurate and effective determination of road congestion bottlenecks, improves the accuracy and efficiency of traffic management, and can quickly find congestion bottlenecks in large-area road networks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113487869B_ABST
    Figure CN113487869B_ABST
Patent Text Reader

Abstract

This application relates to intelligent transportation, and particularly to a method, apparatus, computer equipment, and storage medium for determining congestion bottlenecks. The method includes: acquiring traffic condition data for different time periods and a road network topology map constructed based on the road network data; generating a traffic condition distribution map of each road segment in the road network topology map for different time periods based on the traffic condition data; the traffic condition distribution map reflecting whether each road segment in the road network topology map is congested in different time periods; selecting target road segments in the road network topology map that are congested in at least two of the time periods according to the traffic condition distribution map; determining the congestion contribution score of the target road segment in the set of congested road segments in the road network topology map; and determining congestion bottlenecks based on the target road segments whose congestion contribution scores meet preset conditions. This method can improve the accuracy of determining congestion bottlenecks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, computer equipment and storage medium for determining a congestion bottleneck point. Background Art

[0002] Traffic congestion is an important issue in urban management. With the rapid increase in the number of cars in cities, urban traffic congestion is becoming increasingly serious. The road congestion bottleneck is the direct cause of traffic congestion. The congestion bottleneck will cause large-scale traffic congestion during peak traffic hours, seriously affecting the normal operation of urban traffic. Therefore, how to quickly and accurately find the congestion bottleneck on the road is a prerequisite for solving the traffic congestion problem.

[0003] In traditional technology, the location of congestion bottlenecks on roads is mainly determined by judging whether there is a congested section on the road based on traffic data, and then using the congested section as the congestion bottleneck; or using the congested section reported by the public as the congestion bottleneck. However, the above schemes for determining congestion bottlenecks have problems with accuracy and effectiveness. Summary of the invention

[0004] Based on this, it is necessary to provide a congestion bottleneck point determination method, device, computer equipment and storage medium to address the above technical problems, which can accurately and effectively find the congestion bottleneck point.

[0005] A method for determining a congestion bottleneck point, the method comprising:

[0006] Obtain traffic data at different time periods and a road network topology map based on the road network data;

[0007] Based on the traffic data, a traffic distribution map of each road section in the road network topology map at different time periods is generated; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods;

[0008] Among the road sections of the road network topology map, select a target road section that is congested in at least two of the time periods according to the road condition distribution map;

[0009] Determining a congestion contribution score of the target road section to a congestion cluster formed based on each congested road section in the road network topology map;

[0010] The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches a preset condition.

[0011] In one embodiment, the method further comprises:

[0012] Determine the congestion start time and congestion end time of the congestion cluster;

[0013] Determining the duration of the congestion cluster based on the congestion start time and the congestion end time;

[0014] Determine a congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

[0015] In one embodiment, the traffic condition distribution map includes a traffic condition spatiotemporal distribution map; and generating the traffic condition distribution map of each road section in the road network topology map at different time periods based on the traffic condition data includes:

[0016] Based on the traffic data, a traffic spatial distribution diagram of each road section in the road network topology diagram in each different time period is generated respectively;

[0017] The road condition spatial distribution map of each different time period is combined according to the time dimension to obtain the road condition spatiotemporal distribution map.

[0018] A device for determining a congestion bottleneck point, the device comprising:

[0019] An acquisition module is used to obtain road condition data of different time periods and a road network topology map constructed based on the road network data;

[0020] A generating module, for generating a traffic distribution map of each road section in the road network topology map at different time periods based on the traffic data; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods;

[0021] A selection module, configured to select, from each road section of the road network topology map, a target road section that is congested in at least two of the time periods according to the road condition distribution map;

[0022] A score determination module, used to determine the congestion contribution score of the target road section to the congestion cluster formed based on each congested road section in the road network topology map;

[0023] The bottleneck point determination module is also used to determine the congestion bottleneck point based on the target road section whose congestion contribution score meets the preset conditions.

[0024] In one embodiment, the bottleneck point determination module is further used to:

[0025] sorting the target road sections according to the congestion contribution scores;

[0026] Among the sorted target sections, select the target section whose ranking meets the preset conditions;

[0027] A congestion bottleneck point is determined based on the selected target road section.

[0028] In one embodiment, the road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; the bottleneck point determination module is further used to:

[0029] When the road network is a linear road network, the selected target road section is determined as a congestion bottleneck point;

[0030] When the road network is a mesh road network, the key road nodes associated with the selected target road section are found, and the key road nodes are used as the congestion bottleneck points; or, the selected target road section is clustered, and the target road section intersection points are obtained according to the clustering results, and the target road section intersection points are determined as the congestion bottleneck points.

[0031] In one embodiment, the bottleneck point determination module is further used to:

[0032] When the road network is a linear road network, determining the distance between each selected target road section;

[0033] Connecting target road sections whose distances to each other are less than a preset distance;

[0034] The connected target road section is determined as a congestion bottleneck point.

[0035] In one embodiment, the apparatus further comprises:

[0036] A clustering module, used for clustering the congested road sections in the road network topology map that are congested at different time periods to obtain at least two congestion clusters;

[0037] The score determination module is further used for:

[0038] Determine, among the at least two congestion clusters, a target congestion cluster to which the target road segment belongs;

[0039] Determine a congestion contribution score of the target road segment in the target congestion cluster.

[0040] In one embodiment, the score determination module is further used to:

[0041] Determine the frequency of congestion occurring on the target road section in different time periods; the frequency is a positive integer greater than or equal to two;

[0042] Determining a product value between the frequency and the length of the target section;

[0043] Determine the sum of the product values ​​of the length of each congested road section in the target congestion cluster and the frequency of congestion occurring in each congested road section in the target congestion cluster in different time periods;

[0044] A congestion contribution score is determined based on the product value and the sum value.

[0045] In one embodiment, the clustering module is further used to:

[0046] For the first time period, searching for the congested road sections in the road network topology graph that are congested in the first time period and are interconnected, and taking the interconnected congested road sections as sub-congestion clusters;

[0047] For time periods other than the first one, the congested road sections in the road network topology map that are congested and interconnected in the current time period other than the first one are superimposed with the sub-congestion cluster, until all congested road sections corresponding to the time period other than the first one and interconnected are superimposed with the sub-congestion cluster, to obtain at least two congestion clusters.

[0048] In one embodiment, the apparatus further comprises:

[0049] An acquisition module, used for acquiring congestion clusters formed based on each congested road section in the road network topology map;

[0050] A frequency determination module, used to determine the frequency of congestion occurring in each congested road section within the congestion cluster at different time periods;

[0051] a product value determination module, used to determine a product value between the frequency and the length of the corresponding congested road section in the congestion cluster;

[0052] The cluster size determination module is used to use the sum of the determined product values ​​as the cluster size of the congestion cluster, and the cluster size is used to measure the congestion degree of each congested road section in the congestion cluster.

[0053] In one embodiment, the apparatus further comprises:

[0054] A duration determination module, used to determine the duration of the congestion cluster;

[0055] A screening module, used for screening the congested clusters according to the cluster size and the duration;

[0056] A merging module, used for merging the congested clusters after screening to obtain a merged congested cluster;

[0057] The score determination module is further used to determine the congestion contribution score of the target road section to the merged congestion cluster to which it belongs.

[0058] In one embodiment, the apparatus further comprises:

[0059] A time determination module, used to determine the congestion start time and congestion end time of the congestion cluster;

[0060] A duration determination module, configured to determine the duration of the congestion cluster based on the congestion start time and the congestion end time;

[0061] A type determination module is used to determine the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

[0062] In one embodiment, the road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; the device also includes:

[0063] The cumulative value determination module is used to determine the cumulative value of the length corresponding to the congested road sections in the congestion cluster that are congested during each congestion period when the road network is a linear road network; or, when the road network is a mesh road network, determine the cumulative value of the length corresponding to the congested road sections in the congestion cluster that are congested during each congestion period and belong to the same road; and determine the congestion range of the congestion bottleneck point corresponding to the congestion cluster based on the cumulative value of the length corresponding to the congestion cluster.

[0064] In one embodiment, the traffic condition distribution map includes a traffic condition spatiotemporal distribution map; the generating module is further used to:

[0065] Based on the traffic data, a traffic spatial distribution diagram of each road section in the road network topology diagram in each different time period is generated respectively;

[0066] The road condition spatial distribution map of each different time period is combined according to the time dimension to obtain the road condition spatiotemporal distribution map.

[0067] In one embodiment, the bottleneck point determination module is further used to:

[0068] Taking the target road section whose congestion contribution score reaches the preset condition as the root node, expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes, and obtaining a road section node tree composed of the root node and the expanded adjacent nodes;

[0069] Based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree, the congestion bottleneck point is determined.

[0070] In one embodiment, the bottleneck point determination module is further used to:

[0071] Taking the target road section whose congestion contribution score reaches the preset condition as the root node, and expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes;

[0072] When there is an adjacent node whose congestion contribution score is greater than a preset value among the expanded adjacent nodes, the expanded adjacent node is used as an intermediate node, and a target intermediate node is selected from the intermediate nodes to continue expanding until there is no adjacent node whose congestion contribution score is greater than the preset value among the expanded adjacent nodes;

[0073] When there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, the road segment node tree is constructed based on the root node and all the expanded adjacent nodes.

[0074] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0075] Obtain traffic data at different time periods and a road network topology map based on the road network data;

[0076] Based on the traffic data, a traffic distribution map of each road section in the road network topology map at different time periods is generated; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods;

[0077] Among the road sections of the road network topology map, select a target road section that is congested in at least two of the time periods according to the road condition distribution map;

[0078] Determining a congestion contribution score of the target road section to a congestion cluster formed based on each congested road section in the road network topology map;

[0079] The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches a preset condition.

[0080] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0081] Obtain traffic data at different time periods and a road network topology map based on the road network data;

[0082] Based on the traffic data, a traffic distribution map of each road section in the road network topology map at different time periods is generated; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods;

[0083] Among the road sections of the road network topology map, select a target road section that is congested in at least two of the time periods according to the road condition distribution map;

[0084] Determining a congestion contribution score of the target road section to a congestion cluster formed based on each congested road section in the road network topology map;

[0085] The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches a preset condition.

[0086] The above-mentioned method, device, computer equipment and storage medium for determining congestion bottleneck points construct a road network topology map based on road network data, and generate a road condition distribution map of each road section in the road network topology map at different time periods based on the road condition data of different time periods, so that it can be judged whether each road section in the road network topology map is congested at different time periods according to the road condition distribution map. According to the road condition distribution map, a target road section that is congested in at least two time periods can be selected, and the congestion contribution score of the congestion cluster formed by each congested road section in the road network topology map can be determined. The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches the preset condition, so that the congestion bottleneck point can be accurately determined, and the accuracy of determining the congestion bottleneck point is improved. In addition, since the road condition distribution map is a road condition map for each road section in the road network topology map, it can reflect whether each road section in a large area road network is congested at different time periods, so the congestion bottleneck point can be quickly found in the large area road network, and can be applied to the application scenario of finding congestion bottleneck points in large and medium-sized traffic road networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 A diagram showing an application environment of a method for determining a congestion bottleneck point in an embodiment;

[0088] Figure 2 It is a flowchart of a method for determining a congestion bottleneck point in one embodiment;

[0089] Figure 3 is a schematic diagram of a road condition distribution map in one embodiment;

[0090] Figure 4a A schematic diagram of a road condition distribution map in another embodiment;

[0091] Figure 4b A schematic diagram of a congestion bottleneck point in one embodiment;

[0092] Figure 5a It is a flowchart of a method for determining a congestion bottleneck point in one embodiment;

[0093] Figure 5b It is a flowchart of a method for determining a congestion bottleneck point in one embodiment;

[0094] Figure 5c is a schematic diagram of a target road section in a road network in one embodiment;

[0095] Figure 5d is a schematic diagram of a road section node tree in one embodiment;

[0096] Figure 5e A schematic diagram of a road segment node tree in an adjacent road network area in one embodiment;

[0097] Figure 6a A schematic diagram of a road network topology diagram in an embodiment;

[0098] Figure 6b A schematic diagram of a road network topology diagram in another embodiment;

[0099] Figure 7a is a schematic diagram of a target road section in one embodiment;

[0100] Figure 7b A schematic diagram of a congestion bottleneck point in one embodiment;

[0101] Figure 8 A schematic diagram of a flow chart of a method for determining a congestion contribution score in one embodiment;

[0102] Fig. 9 A schematic diagram of a congestion range of a linear road network in one embodiment;

[0103] Fig.10a A schematic diagram of a congested road section within a time period of 0-10 minutes in one embodiment;

[0104] Fig.10b This is a schematic diagram of a congested road section within a 10-20 minute period in one embodiment;

[0105] Fig.10c This is a schematic diagram of a congested road section within a 20-30 minute period in one embodiment;

[0106] Fig.10d A schematic diagram of a congested road section within a 30-40 minute period in one embodiment;

[0107] Fig.11 A schematic diagram of a flow chart of a method for determining a congestion range in one embodiment;

[0108] Fig.12 It is a structural block diagram of a device for determining a congestion bottleneck point in one embodiment;

[0109] Fig.13 It is a structural block diagram of a device for determining a congestion bottleneck point in another embodiment;

[0110] Fig.14 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0111] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0112] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. Basic artificial intelligence technologies generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, mechatronics and other technologies. Artificial intelligence software technologies mainly include computer vision technology, speech processing technology, natural language processing technology, as well as machine learning / deep learning, autonomous driving, smart transportation and other major directions.

[0113] The solution provided in the embodiments of the present application involves the intelligent transportation technology of artificial intelligence, which is specifically described by the following embodiments:

[0114] The congestion bottleneck point determination method provided in this application can be applied to Figure 1 In the application environment shown. In the application environment, the server 102 communicates with the vehicle 104 traveling in the road network through the network, and the vehicle 104 sends the road condition data of different time periods to the server 102. The server 102 obtains the road condition data of different time periods and the road network topology map constructed based on the road network data; based on the road condition data, a road condition distribution map of each road section in the road network topology map at different time periods is generated; the road condition distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods; among each road section in the road network topology map, a target road section that is congested in at least two time periods is selected according to the road condition distribution map; the congestion contribution score of the target road section to the congestion cluster formed based on each congested road section in the road network topology map is determined; and the congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches the preset condition.

[0115] Among them, server 102 can be an independent physical server or a server cluster composed of multiple service nodes in the blockchain system. The service nodes form a peer-to-peer (P2P) network. The P2P protocol is an application layer protocol running on the Transmission Control Protocol (TCP) protocol.

[0116] In addition, server 102 can also be a server cluster composed of multiple physical servers, and can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), as well as big data and artificial intelligence platforms.

[0117] The vehicle 104 may be any vehicle traveling on a road network and equipped with a positioning system, including taxis, buses, etc. The positioning system may be a GPS (Global Positioning System) positioning system, a Beidou positioning system, a GLONASS (Global Navigation Satellite System) positioning system, a Galileo positioning system, etc. The server 102 and the vehicle 104 may be connected via a communication connection such as a network, and this application does not limit this.

[0118] In one embodiment, Figure 2 As shown, a method for determining a congestion bottleneck point is provided, and the method is applied to Figure 1 The server in the example is used to illustrate the following steps:

[0119] S202, obtaining road condition data of different time periods and a road network topology map constructed based on the road network data.

[0120] The time period may be a time period of a preset duration, and the preset duration may be, for example, 5 minutes, 10 minutes, etc. Different time periods may be time periods that are continuous in time or discontinuous in time. For example, the server divides 24 hours into 288 different time periods, and the duration of each time period is 5 minutes. For another example, the server divides one hour into 6 time periods, namely, 0-9min time period, 10-19min time period, 20-29min time period, 30-39min time period, 40-49min time period, and 50-59min time period.

[0121] The road condition data is data reflecting the traffic conditions of vehicles in the road network, and may include at least one of a traffic congestion index, an average driving speed, and a vehicle flow rate.

[0122] A road network topology map may be a topology map used to represent the connection relationship between road sections in a road network, and may be composed of edges, endpoints, and intersections, etc. Among them, the edges in the road network topology map are used to represent road sections in the road network, the endpoints are used to represent the starting point or end point of the road section, and the intersections are used to represent the intersections of two or more roads at the intersection.

[0123] In one embodiment, the server may divide the roads in the road network into multiple sections, and obtain the traffic data of each section at different time periods. For example, the server may divide the 2km long expressway A into 40 50m long sections, and obtain the traffic data of each section at different time periods.

[0124] In one embodiment, vehicles traveling on the road network obtain location information through a positioning device, and then send the location information to a server through a wireless communication network. The server calculates the road condition data of each road section based on the location information uploaded by each vehicle.

[0125] In one embodiment, after S202, the server may also pre-process the traffic data, wherein the pre-processing of the traffic data includes: removing redundant data in the traffic data, filtering the traffic data, and removing abnormal data and incomplete data in the traffic data.

[0126] Among them, road network data is data that reflects the topological relationship between various road sections in the road network, including the location coordinates, length, number of lanes and coordinates of intersections of each road section.

[0127] In one embodiment, the server may also obtain road network data from a database corresponding to the electronic map.

[0128] S204, based on the traffic data, generating a traffic distribution map of each road section in the road network topology map at different time periods; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods.

[0129] The server may determine whether each road section is congested at different time periods according to at least one of the traffic congestion index, average driving speed and vehicle flow of each road section at different time periods. For example, at different time periods, the server determines that the road section with a congestion index greater than the index threshold in the road network map is congested; or the server determines that the road section with an average driving speed less than a preset speed threshold is congested; or the server determines that the road section with a congestion index greater than the index threshold and a vehicle flow greater than the vehicle flow threshold in the road network topology map is congested; or the server performs weighted calculation on the average driving speed and vehicle flow of each road section, and determines whether each road section is congested according to the result of the weighted calculation.

[0130] The traffic distribution map is a distribution map used to reflect whether each road section in the road network topology map is congested at different times. It can be a two-dimensional plane map, such as Figure 3 In addition, it can also be a three-dimensional map. The server marks the road sections that are congested in each period in the road network topology map in the traffic distribution map. For example, the congested road sections are marked with different colors from the non-congested road sections. Figure 3 .

[0131] In one embodiment, when the road network is a linear road network, the server generates a two-dimensional road condition distribution map of each road section in the linear road network at different time periods based on the road condition data. In the road condition distribution map of the linear road network, the horizontal axis represents the distance between each point in the road and the starting point, and the vertical axis represents the time period.

[0132] In one embodiment, the road condition distribution map includes a road condition spatiotemporal distribution map; based on the road condition data, generating the road condition distribution map of each road section in the road network topology map at different time periods includes: based on the road condition data, generating the road condition spatial distribution map of each road section in the road network topology map at each different time period; combining the road condition spatial distribution map of each different time period according to the time dimension to obtain the road condition spatiotemporal distribution map.

[0133] Among them, the road condition spatial distribution map is a road condition distribution map used to reflect the road condition distribution map of the road sections where congestion occurs within a certain period of time. When the road network is a linear road network, the road condition spatial distribution map is a one-dimensional spatial distribution map, and when the road network is a mesh road network, the road condition spatial distribution map is a two-dimensional spatial distribution map. Among them, the road condition spatiotemporal distribution map is a distribution map used to reflect the distribution of road sections where congestion occurs in the time dimension and the space dimension.

[0134] In one embodiment, when the road network is a mesh road network, the server generates a two-dimensional road condition spatial distribution map corresponding to each time period according to the road condition data of each time period, and then combines the two-dimensional road condition spatial distribution map of each time period according to the time dimension to obtain a three-dimensional road condition spatiotemporal distribution map. In the three-dimensional road condition spatiotemporal distribution map, the two horizontal coordinate axes in each two-dimensional road condition spatial distribution map represent the distance between each point in the road with different directions and the starting point of the corresponding road, and the coordinate axis perpendicular to the two-dimensional road condition spatial distribution map represents the time period.

[0135] In one embodiment, the server first constructs a spatiotemporal distribution graph, wherein the horizontal axis in the spatiotemporal distribution graph represents the distance between each point on the road and the starting point, and the vertical axis represents the time period. Then, in the spatiotemporal distribution graph, the road sections that are congested and the road sections that are not congested are marked with different colors to obtain the spatiotemporal distribution graph of the road conditions of each road section in different time periods. For example, Figure 3 The spatiotemporal distribution diagram of the traffic conditions of road A in the linear road network. The horizontal axis represents the distance between each section of road A and the starting point of road A, and the vertical axis represents the time period. The server marks the sections with congestion at different time periods with black lines, and marks the sections without congestion at different time periods with white lines.

[0136] In addition, the server may also mark the road sections where congestion occurs with solid lines, and mark the road sections where congestion does not occur with dotted lines.

[0137] S206 , among the road sections, select a target road section that is congested in at least two time periods according to the traffic distribution map.

[0138] Among them, at least two time periods may be adjacent time periods or non-adjacent time periods in different time periods. For example, among the 144 different time periods, at least two time periods may be the 3rd to 15th time periods, or the 20th time period and the 35th time period. Correspondingly, the target road section may refer to a congested road section that is congested in at least two time periods, such as a congested road section that is congested in the 3rd to 15th time period, or a congested road section that is congested in the 20th time period and the 35th time period.

[0139] In one embodiment, S206 may specifically include: the server determines the road sections where congestion occurs in each time period according to the traffic distribution map, and selects the target road section where congestion occurs in at least two time periods from each road section.

[0140] For example, Figure 4a As shown, Figure 4a The following is a traffic distribution map of Road A in a linear road network. The line segment between the two endpoints represents a 100-meter section of Road A. A solid line segment indicates that the section is congested during the corresponding time period, and a dotted line segment indicates that the section is not congested during the corresponding time period. For example, the line segment numbered 4 indicates that the section 100-200 meters of Road A is congested during the 30-40 minute period. Figure 4a According to the traffic distribution map shown, the server can determine that the section of 100-200 meters is congested in 5 time periods, the section of 200-300 meters is congested in 2 time periods, the section of 300-400 meters is congested in 3 time periods, the section of 600-700 meters is congested in 1 time period, and the section of 700-800 meters is congested in 2 time periods. Therefore, the target sections selected by the server are the four congested sections of 100-200 meters, 200-300 meters, 300-400 meters, and 700-800 meters, which are congested in at least two time periods.

[0141] S208, determining the congestion contribution score of the target road section to the congestion cluster formed based on each congested road section in the road network topology map.

[0142] The congested road section is a road section in the road network topology that is congested in one or more time periods. The congestion contribution score is used to indicate the congestion contribution of the target road section in the set of congested road sections. The congestion contribution score can be expressed in numbers, fractions or percentages, for example, the congestion contribution score can be 5, 30%, or 2 / 3.

[0143] Among them, a congestion cluster is a collection of congested road sections. The congested road sections in the same congestion cluster are related in time or space. The temporal relationship can be that the congestion periods are adjacent, or the congestion periods are within the same time range; the spatial relationship can be that they are adjacent in space or can be connected, or belong to the same spatial range. Figure 4aAs shown, the congestion cluster is a set of congested road sections 1-10.

[0144] In one embodiment, the server may determine the congestion contribution score according to the frequency of congestion occurring in the target road section at different time periods, so that the congestion contribution score of the target road section is proportional to the frequency of congestion occurring in different time periods. Figure 4a As shown, the 100-200 meter section was congested in 5 different time periods, the 200-300 meter section was congested in 2 different time periods, the 300-400 meter section was congested in 3 different time periods, the 600-700 meter section was congested in 1 time period, and the 700-800 meter section was congested in 2 different time periods. Among them, the 100-200 meter section had the highest frequency of congestion in different time periods, and the server determined the highest congestion contribution score for the 100-200 meter section.

[0145] S210, determining a congestion bottleneck point based on a target road section whose congestion contribution score meets a preset condition.

[0146] The preset condition is used to measure whether the target road section can be used to determine the congestion bottleneck point. Specifically, the preset condition can be a preset score or a preset score range, such as whether the congestion contribution score of the target road section reaches the preset score, or whether it is within the preset score range. The preset score can be a preset score, or the average of all congestion contribution scores, or the weighted sum of the congestion contribution scores.

[0147] Congestion bottlenecks are intersections, sections or areas in the road network where congestion often occurs, such as road intersections, forks, tunnels and areas formed by multiple sections. Congestion bottlenecks have various types of congestion, including morning peak bottlenecks, all-day bottlenecks and evening peak bottlenecks. For example, Figure 4b Shown are the congestion bottlenecks identified by the server.

[0148] In one embodiment, S210 may specifically include: the server determines the target road section as a congestion bottleneck point; or, the server may also determine the area to which the target road section belongs as a congestion bottleneck point; or, the server may also determine the target road section and the road section connected to the target road section as a congestion bottleneck point; or, the server may also determine the intersection between the target road sections as a congestion bottleneck point.

[0149] After S210, the server may also determine the type of congestion bottleneck point, and specific steps may include: the server determines the congestion start time and congestion end time of the congestion cluster; determines the duration of the congestion cluster based on the congestion start time and congestion end time; and determines the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration. Since the congested sections in the congestion cluster are distributed in different time periods, the server can determine the duration of the congestion cluster through the congestion start time and congestion end time of the congestion cluster, and then determine the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration, so that the congestion bottleneck points can be classified and controlled according to the congestion types corresponding to the congestion bottleneck points, thereby improving the efficiency of controlling the congestion bottleneck points.

[0150] Among them, the server can classify the congestion bottleneck points into morning peak bottleneck points, all-day bottleneck points, evening peak bottleneck points and other congestion types according to the time range to which the congestion cluster duration belongs. For example, 24 hours are divided into 144 time periods, each time period is 10 minutes. If the duration of the congestion cluster is within the range of 200min-600min, the server will determine the congestion bottleneck point corresponding to the congestion cluster as the morning peak bottleneck point; if the duration of the congestion cluster is within the range of 800min-1200min, the server will determine the congestion bottleneck point corresponding to the congestion cluster as the evening peak bottleneck point; if the duration of the congestion cluster exceeds 8 hours, the server will determine the congestion bottleneck point corresponding to the congestion cluster as the all-day bottleneck point. For example, if Figure 3 The bottleneck points throughout the day, the bottleneck points during the morning rush hour, and the bottleneck points during the evening rush hour are shown in .

[0151] In the above embodiment, a road network topology map is constructed based on the road network data, and a road condition distribution map of each road section in the road network topology map at different time periods is generated based on the road condition data of different time periods, so that it can be judged whether each road section in the road network topology map is congested at different time periods according to the road condition distribution map. According to the road condition distribution map, a target road section that is congested in at least two time periods can be selected, and the congestion contribution score of the congestion cluster formed by each congested road section in the road network topology map can be determined. The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches the preset condition, so that the congestion bottleneck point can be accurately determined, and the accuracy of determining the congestion bottleneck point is improved. In addition, since the road condition distribution map is a road condition map for each road section in the road network topology map, it can reflect whether each road section in a large area road network is congested at different time periods, so the congestion bottleneck point can be quickly found in the large area road network, and it can be applied to the application scenario of finding congestion bottleneck points in large and medium-sized traffic road networks.

[0152] In one embodiment, Figure 5a As shown, S210 may specifically include the following steps:

[0153] S502a, sorting the target road sections according to the congestion contribution scores.

[0154] The target road sections may be sorted in descending order according to the congestion contribution scores, or in descending order according to the congestion contribution scores.

[0155] S504a, selecting a target road section whose ranking meets a preset condition from among the sorted target road sections.

[0156] Among them, the preset condition can be used to indicate whether the sorted ranking reaches the preset ranking; or, when sorting in descending order of congestion contribution scores, the preset condition can also be that the sorted ranking is within a preset ranking range, for example, the sorted ranking is within the top 20% ranking range.

[0157] S506a: Determine a congestion bottleneck point based on the selected target road section.

[0158] In one embodiment, S506a may specifically include: the server determines the selected target road section as a congestion bottleneck point; or, the server may also determine the area to which the selected target road section belongs as a congestion bottleneck point; or, the server may also determine the selected target road section and the road section connected to the selected target road section as a congestion bottleneck point; or, the server may also determine the intersection between the target road sections as a congestion bottleneck point.

[0159] In the above embodiment, the server sorts the target road sections according to the size of the congestion contribution score, and determines the congestion bottleneck points based on the target road sections whose rankings meet the preset conditions, so that the congestion bottleneck points can be determined more accurately and reasonably.

[0160] In one embodiment, Figure 5b As shown, S210 may specifically include the following steps:

[0161] S502b, taking the target road segment whose congestion contribution score reaches the preset condition as the root node, expanding the adjacent nodes layer by layer to the remaining target road segments as child nodes, and obtaining a road segment node tree composed of the root node and the expanded adjacent nodes.

[0162] The remaining target road segments are target road segments other than the target road segment corresponding to the root node. The remaining target road segments include target road segments adjacent to the target road segment corresponding to the root node and target road segments not adjacent to the target road segment corresponding to the root node.

[0163] The layer-by-layer expansion of adjacent nodes starts from the root node, and the target road segment corresponding to the expanded child node is adjacent to the target road segment corresponding to the parent node of the child node. Figure 5cAs shown in FIG. 1 , there is a target road section AK in the road network. If the target road section whose congestion contribution score reaches the preset condition is the target road section B, the server uses the target road section B as the root node. Since the target roads A, C and D are adjacent to the target road section B, the server expands the root node based on the target roads A, C and D to obtain the adjacent nodes A, C and D of the root node, and then expands the adjacent nodes A, C and D, and so on. The expanded road section node tree is as follows: Figure 5d shown.

[0164] In one embodiment, S502b specifically includes: taking the target road section whose congestion contribution score reaches the preset condition as the root node, and expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes; when there is an adjacent node with a congestion contribution score greater than a preset value among the expanded adjacent nodes, taking the expanded adjacent nodes as intermediate nodes, and selecting target intermediate nodes from the intermediate nodes to continue expanding until there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes; when there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, forming a road section node tree based on the root node and all expanded adjacent nodes.

[0165] Among them, the intermediate node is a node with a parent node and a child node in the road segment node tree. The server can select some target intermediate nodes that meet the conditions from the intermediate nodes and continue to expand the target intermediate nodes. The selected target intermediate node can be an intermediate node whose congestion contribution score is greater than a preset score, or an intermediate node whose length in the corresponding target road segment is greater than a preset length value.

[0166] The server first selects the target road segment whose congestion contribution score meets the preset conditions as the root node, and expands from the root node to the adjacent nodes layer by layer. The target road segment corresponding to the expanded adjacent node is adjacent to the target road segment corresponding to the parent node of the adjacent node. When there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, the expansion is stopped, and the server constructs a road segment node tree based on the root node and the target road segments corresponding to all expanded adjacent nodes.

[0167] In one embodiment, after S502b, it also includes: combining all the segment node trees to obtain a segment node forest; in the segment node forest, merging the segment node trees corresponding to the overlapping target segments on the boundaries of different road network areas to obtain a segment node forest after merging the segment node trees.

[0168] The road network area is an area obtained by dividing the road network topology map according to the division rules. The division rules may be, for example, division according to Geohash coding, which is an address coding method that uses a string to represent the longitude and latitude of each point in the road network topology map. The server may divide a large-scale road network topology map according to Geohash coding to obtain multiple road network areas. For example, the server may divide the road network topology map of a city-level city according to the Geohash coding of each coordinate point in the road network topology map of the city-level city to obtain multiple road network areas.

[0169] The merging is to combine multiple segment node trees into one segment node tree. Figure 5e As shown, the segment node tree 2 in the road network area A is composed of target segments 1, 2, 3, 4, 5, and 6, and the segment node tree 1 in the road network area B is composed of target segments 6, 7, 8, 9, and 10. The segment node tree 1 and the segment node tree 2 have an overlapping target segment 6. The server merges the segment node tree 1 and the segment node tree 2 into a segment node tree composed of target segments 1-10.

[0170] In one embodiment, in the road segment node forest, the road segment node trees corresponding to the overlapping target road segments on the boundary of the adjacent road network area are determined; the overlapping target road segments in the determined road segment node trees are deduplicated; and the road segment node trees obtained after the deduplication processing are merged to obtain a road segment node forest after the merged road segment node trees.

[0171] The deduplication process is to remove the overlapping target road segments so that there are no overlapping target road segments in the merged road segment node tree. Figure 5e As shown, the segment node tree 1 and the segment node tree 2 have overlapping target segments 6 on the boundaries of adjacent road network areas. The server performs deduplication processing on the target segments 6 in the segment node tree 1 and the segment node tree 2, retains the target segment 6 in only one segment node tree (assuming that the target segment 6 in the segment node tree 2 is retained), and merges the segment node tree 1 and the segment node tree 2 obtained after the deduplication processing.

[0172] In one embodiment, after obtaining the segment node forest after merging the segment node trees, it also includes: for each segment node tree in the segment node forest, the server determines the corresponding congestion bottleneck points based on the target segments corresponding to the root node to the leaf nodes in the segment node tree.

[0173] S504b: Determine the congestion bottleneck point based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree.

[0174] In one embodiment, S504b may include: connecting the target road sections corresponding to the root node to the leaf nodes end to end, and determining the connected route as a congestion bottleneck point.

[0175] In another embodiment, S504b may further include: determining a branch with the most layers in the road segment node tree, and determining the target road segments corresponding to the root node to the leaf nodes in the branch with the most layers as congestion bottleneck points.

[0176] In the above embodiment, the target road section whose congestion contribution score reaches the preset conditions is taken as the root node, and the adjacent nodes are expanded layer by layer to the remaining target road sections as child nodes to obtain a road section node tree composed of the root node and the expanded adjacent nodes, and the congestion bottleneck point is determined based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree, so that the determined congestion bottleneck point is more in line with the actual situation of congestion, thereby improving the accuracy of determining the congestion bottleneck point.

[0177] In one embodiment, the road network includes a linear road network and a mesh road network. A linear road network is a road network where each road intersects at a three-dimensional intersection and has no intersections in a two-dimensional plane. For example, a linear road network may be a highway network, an express road network, etc. A mesh road network is a road network where each road has intersections in a two-dimensional plane.

[0178] The server uses the link as the basic unit to form a road network topology map of the linear road network. The link is composed of an ordered sequence of coordinates and is used to represent a road segment of a preset length in the road network. The starting and ending points of the road segment are the endpoints of the link. Since each road in the linear road network has no intersection in the two-dimensional plane, the road network topology map corresponding to each road in the linear road network is a one-dimensional linear map. For example, Figure 6a The figure shows the road network topology of Road A and Road B in a linear road network. Both Road A and Road B are composed of connecting edges connected end to end. Each connecting edge represents a 5m long road segment, and the connecting edges are separated by endpoints. There is no intersection between the road network topology of Road A and Road B.

[0179] The server uses multiple connection edges connected in sequence to form a connection segment (Segment) as the basic unit to form a network topology map of the mesh network. Since each connection segment intersects at the intersection, the network topology map corresponding to the mesh network is a two-dimensional mesh map. For example, Figure 6b The figure shows a road network topology diagram of a mesh road network, which includes road A, road B, road C and road D, and each road intersects at an intersection. In the road network topology diagram of the mesh road network, the server uses a connection segment composed of connecting edges to represent a road segment between two intersections on a road.

[0180] In one embodiment, the road network topology map includes a topology map constructed based on road network data corresponding to different road networks; S506 may specifically include: when the road network is a linear road network, determining the selected target road section as a congestion bottleneck point; when the road network is a mesh road network, finding the key road nodes associated with the selected target road section, and using the key road nodes as congestion bottleneck points; or, clustering the selected target road sections, obtaining the target road section intersection points according to the clustering results, and determining the target road section intersection points as congestion bottleneck points.

[0181] Among them, key road nodes are sections in the road network that play a key role in restricting the smooth passage of vehicles, such as ramps, tunnel entrances, narrow sections, and river crossings.

[0182] The key road nodes associated with the selected target road section may be: key road nodes that are spatially associated with the selected target road section, for example, key road nodes connected to the target road section; or key road nodes within a preset range around the selected target road section; or key road nodes that belong to the same road as the selected target road section.

[0183] In one embodiment, the server determines the key road nodes according to the road network topology map corresponding to the mesh road network. Then, the key road nodes associated with the selected target road section are searched in the road network topology map according to a search algorithm. The search algorithm includes but is not limited to a sequential search method, a binary search method, a binary tree sorting search method, a hash table search method, and a block search method.

[0184] The target road section intersection is an intersection formed at the intersection of two or more target road sections, including but not limited to a T-shaped intersection, a Y-shaped intersection, a cross-shaped intersection, and a roundabout intersection.

[0185] In one embodiment, the server clusters the target road sections that meet the preset conditions through a clustering algorithm to obtain a set of multiple target road sections, and then determines the target road section intersection in each set, and determines the target road section intersection as the congestion bottleneck point. The clustering algorithm includes but is not limited to a partitioning method (e.g., K-Means algorithm), a hierarchical method (e.g., BIRCH algorithm), a density algorithm (e.g., DBSCAN algorithm), a graph theory clustering method, a grid algorithm (STING algorithm), and a model algorithm.

[0186] In the above embodiment, the road network includes a linear road network and a mesh road network. For a linear road network, the target road section selected by the server is determined as a congestion bottleneck point; and for a mesh road network, the server finds the key road nodes or clusters the target road section intersection points, and determines the key road nodes or the target road section intersection points as congestion bottleneck points. Thus, the server can determine the congestion bottleneck points according to the characteristics of different road networks, and the determined congestion bottleneck points are more accurate. In addition, since the key road nodes and the target road section intersection points are nodes in the road network that are prone to congestion, the server determines the obtained key road nodes and the target road section intersection points as congestion bottleneck points, which improves the rationality of the determined congestion bottleneck points.

[0187] In one embodiment, the above-mentioned step of determining the selected target road section as a congestion bottleneck point when the road network is a linear road network may specifically include: when the road network is a linear road network, determining the distance between each selected target road section; connecting the target road sections whose distances are less than a preset distance; and determining the connected target road sections as congestion bottleneck points.

[0188] The distance between each selected target road section may be: the distance between coordinate points selected from each target road section. The selected coordinate point may be the midpoint of each target road section, or may be the end point of each target road section.

[0189] In one embodiment, the step of connecting the target road segments whose distances are less than the preset distance may specifically include: the server connects the two target road segments whose distances are less than the preset distance and the road segment between the target road segments to form a road segment. Figure 7a As shown, the server connects section 4 to section 6, section 6 to section 8, section 9 to section 11, and section 11 to section 13, and the following is obtained: Figure 7b The connected target segment is shown in .

[0190] In one embodiment, when the road network is a linear road network, the step of determining the selected target road section as the congestion bottleneck point may specifically include: when the road network is a linear road network, the server determines whether each selected target road section meets the discrete condition, and if so, the server connects the target road sections whose distances are less than a preset distance, and determines the congestion bottleneck point according to the road sections obtained after the connection. The discrete condition may be, for example, that the variance of the midpoint coordinates of each target road section reaches a preset variance value, or the discrete condition may be that the average difference of the midpoint coordinates of each target road section reaches a preset difference value, etc.

[0191] In the above embodiment, since congestion will spread along the road line when a linear road network is congested, connecting target road sections whose distances are less than a preset distance and determining the connected target road sections as congestion bottleneck points can make the determined congestion bottleneck points more consistent with the actual situation of congestion, thereby improving the rationality of the determined congestion bottleneck points. In addition, when the road network is a linear road network, if each target road section meets the discrete condition, it means that the target road sections are distributed more dispersedly. When congestion occurs, since the road sections in the linear road network will affect each other, the congested road sections will spread along the linear road. The server determines the road sections obtained after connection as congestion bottleneck points, which is more consistent with the actual situation of the linear road network when congestion occurs, and the determined congestion bottleneck points are more reasonable.

[0192] In one embodiment, Figure 8 As shown, S208 may include the following steps:

[0193] S802: Cluster the congested road sections that are congested at different time periods in the road network topology map to obtain at least two congestion clusters.

[0194] Clustering can be a process of dividing a collection of physical objects or abstract objects into multiple classes consisting of similar objects. The server clusters the congested sections in the road network topology map that are congested at different times by clustering the congested sections in time and space, and dividing the congested sections that are related in time or space into the same congestion cluster. The cluster generated by clustering is a collection of a group of data objects that are similar to objects in the same cluster and different from objects in other clusters. The above-mentioned data objects may refer to congested sections.

[0195] A congestion cluster can be a collection of congested sections. The congested sections in the same congestion cluster are related in time or space. The temporal relationship can be that the congestion periods are adjacent, or that the congestion periods are within the same time range; the spatial relationship can be that they are spatially adjacent or connected, or belong to the same spatial range.

[0196] When the road network is a linear road network, the congested sections that occur congestion in each time period are distributed in a one-dimensional space, so the congestion clusters obtained by the server by clustering the congested sections in different time periods are two-dimensional congestion clusters; when the road network is a mesh road network, the congested sections that occur congestion in each time period are distributed in a two-dimensional space, so the congestion clusters obtained by the server by clustering the congested sections in different time periods are three-dimensional congestion clusters.

[0197] S804: Determine, among at least two congestion clusters, a target congestion cluster to which the target road segment belongs.

[0198] The server clusters the congested road sections that are congested at different time periods to obtain at least two congestion clusters, each of which includes a plurality of congested road sections that are temporally or spatially associated.

[0199] In one embodiment, the server may determine the target congestion cluster to which the target road segment belongs according to the coordinate value of the target road segment.

[0200] In one embodiment, before S804, the process further includes: the server optimizes at least two congestion clusters, and the optimization method may specifically include: the server determines the center point of each congestion cluster, and deletes the congested road sections in each congestion cluster whose distance from the center point exceeds a preset distance. Since the congested road sections far from the center point have a weaker correlation with other congested road sections, the server optimizes each congestion cluster to improve the accuracy of the obtained congestion cluster.

[0201] S806: Determine the congestion contribution score of the target road section in the target congestion cluster.

[0202] In one embodiment, the server may determine the congestion contribution score based on the frequency of congestion occurring in the target road section at different time periods in the target congestion cluster. Specifically, the server determines the frequency of congestion occurring in the target road section at different time periods in the target congestion cluster, and inputs the determined frequency into a score calculation function with the frequency as a parameter to obtain the congestion contribution score. The score calculation function may be a linear function with the frequency as a parameter, or a nonlinear function with the frequency as a parameter.

[0203] In the above embodiment, the server clusters the congested sections in the road network topology map that are congested at different time periods to obtain at least two congestion clusters, and determines the congestion contribution scores of the target sections in the target congestion clusters to which they belong. Therefore, the server can determine the contribution of each target section to the congestion level of the congested section according to the congestion contribution score, and determine the target section that contributes more to the congestion level as the congestion bottleneck point, so that the congestion bottleneck point can be determined more accurately.

[0204] In one embodiment, S802 may specifically include: for the first time period, searching for congested road sections in the road network topology map that are congested in the first time period and interconnected, and taking the interconnected congested road sections as sub-congestion clusters; for non-first time periods, superimposing congested road sections in the road network topology map that are congested in the current non-first time period and interconnected with the sub-congestion clusters, until all congested road sections corresponding to all non-first time periods and interconnected are superimposed with the sub-congestion clusters, to obtain at least two congestion clusters.

[0205] Among the interconnected congested sections, any point on each congested section can be interconnected with any point on other congested sections, and the interconnected congested sections can form a strong connection. Superimposing the interconnected congested sections with the sub-congested clusters can be a collection of new congested sections by combining the interconnected congested sections with the congested sections in the sub-congested clusters.

[0206] In another embodiment, S802 may also specifically include: for the first time period, the server determines the strong connection composed of the congested road sections in the road network topology map that are congested in the first time period, and uses the above strong connection as a sub-congestion cluster, wherein the strong connection is a collection of interconnected congested road sections. For non-first time periods, the server determines the strong connection in the road network topology map of the current time period, and then traverses the congested road section A in each strong connection, queries the sub-congestion cluster to which the congested road section A belongs in the previous time period, and merges the sub-congestion cluster with the strong connection corresponding to the congested road section A into a new sub-congestion cluster, until all the strong connections composed of the congested road sections corresponding to the non-first time periods are superimposed on the sub-congestion cluster to which the congested road section belongs in the previous time period, and at least two congestion clusters are obtained.

[0207] In the above embodiment, the server superimposes the congested sections that are congested and interconnected in the current period with the sub-congested clusters formed in the previous period to cluster the congested sections, obtain at least two congested clusters, and can determine the congestion contribution score of each target section according to the clustered congested clusters, and finally determine the congestion bottleneck point, so that the congestion bottleneck point can be determined more accurately. In addition, the server forms strong connections between the congested sections in the current period, which is equivalent to clustering the congested sections in the spatial dimension, and then superimposes the strong connections to which the congested sections belong in the current period with the sub-congested clusters to which the congested sections belong in the previous period, which is equivalent to clustering the congested sections in the time dimension. The correlation of the congested sections in the spatial dimension and the time dimension is fully utilized, and the congested sections in the obtained congested clusters have correlation in the two dimensions of time and space, which can better reflect the actual situation of congestion, so that the congestion bottleneck point can be determined more accurately and effectively.

[0208] In one embodiment, before S208, it also includes: obtaining a congestion cluster formed based on each congested road section in the road network topology map; determining the frequency of congestion in each congested road section in the congested cluster at different time periods; determining the product value between the frequency and the length of the corresponding congested road section in the congested cluster; and using the sum of the determined product values ​​as the cluster size of the congested cluster, and the cluster size is used to measure the congestion degree of the congested road section in the congested cluster.

[0209] The frequency of congestion in different time periods of each congested road section may be the number of time periods in which congestion occurs in the congested road section in all different time periods. Figure 4aAs shown, the congested sections numbered 1-10 form a congestion cluster, the target section of 100-200 meters is congested in 5 different time periods, and the frequency of congestion in the congested section in different time periods is 5.

[0210] Among them, the cluster size is used to measure the congestion degree of the congested road section within the congestion cluster. The congestion degree is a parameter used to represent the spatiotemporal impact range of the congestion cluster, which includes the spatial distribution range of the congested road sections in the congestion cluster and the temporal duration range of the congested road sections in the congestion cluster. The longer the length of the congested road section in the congestion cluster, the higher the congestion degree, and the longer the congestion duration of the congested road section in the congestion cluster, the higher the congestion degree.

[0211] The server determines the product value between the frequency and the length of the corresponding congested road section in the congestion cluster by multiplying the number of time periods in which the congested road section is congested at different time periods by the length of the congested road section to obtain the product value. Figure 4a As shown, the frequency of congestion in the congested section of 100-200 meters is 5 and the length is 100, so the product value obtained by the server is 500.

[0212] In one embodiment, the server determines the product between the frequency and the length of the corresponding congested road section in the congestion cluster, which may include: the server obtains the congestion index marking value of each congested road section in different time periods, and then calculates the product of the congestion index marking value of each congested road section in different time periods and the length of the road section, and uses the obtained product value as the product value between the frequency and length of congestion in each congested road section in the congestion cluster in different time periods.

[0213] For example, the server determines the product value between the frequency and the length of the corresponding congested road section in the congestion cluster by formula (1), where s0 is the determined product value and T is the set of all different time periods from the start time to the end time of the congestion cluster. i,j is the mark value of the congestion index of congested road section i in time period j. When congested road section i is congested in time period j, c i,j is 1, when the target road section i is not congested during the j period, c i,j is 0. i is the length of road section i.

[0214]

[0215] In one embodiment, the server uses the above sum as the cluster size of the congestion cluster. The cluster size can be calculated in the following way: the server first determines the product value between the length of each congested road section in the congestion cluster and the frequency of congestion of each congested road section in different time periods, and then sums the obtained product values ​​to obtain the cluster size of the congestion cluster.

[0216] In another embodiment, the cluster size can also be calculated in the following manner: the server first calculates the product of the congestion index mark value corresponding to each road segment in the congestion cluster and the road segment length, and then sums the product values ​​obtained for each road segment to obtain the cluster size of the congestion cluster. For example, the server calculates the cluster size of the congestion cluster according to formula (2). Where s is the obtained cluster size, L is the congestion cluster composed of each road segment, and T is the set of all time periods from the start time to the end time of the congestion cluster. i,j is the mark value of the congestion index of road section i in time period j. When road section i is a congested road section in time period j, c i,j is 1, when the road section i is a smooth road section in the j period, c i,j is 0. i is the length of road section i.

[0217]

[0218] In one embodiment, the duration of each congestion cluster is determined; the congestion clusters are screened according to the cluster size and duration; the screened congestion clusters are merged to obtain a merged congestion cluster; S208 may include: determining the congestion contribution score of the target road segment to the merged congestion cluster to which it belongs.

[0219] The duration is the time interval between the congestion start time and the congestion end time of the congestion cluster. The congestion start time is the earliest time when the congested road section appears in the congestion cluster. The congestion end time is the latest time when the congested road section appears in the congestion cluster. For example, Figure 3 As shown in the figure, the congested section in congestion cluster A appears at 400 minutes at the earliest and 1350 minutes at the latest, with a duration of 950 minutes.

[0220] In one embodiment, filtering the congested clusters according to the cluster size and duration may be that the server deletes the congested clusters whose cluster size does not reach the quantitative threshold, or deletes the congested clusters whose duration does not reach the time threshold, or deletes the congested clusters whose cluster size does not reach the quantitative threshold and whose duration does not reach the time threshold at the same time.

[0221] In one embodiment, based on the merging of filtered congested clusters, the merged congested clusters can be obtained by the server merging congested clusters whose interval time does not reach a preset time, or merging congested clusters whose spatial interval does not reach a preset distance, or merging congested clusters whose interval time does not reach a preset time and whose spatial interval does not reach a preset distance.

[0222] In the above embodiment, the server screens and merges the congestion clusters to obtain the congestion clusters that have a greater impact on the congestion level in the road network, making the obtained congestion clusters more reasonable and more practical. Then, the server determines the target congestion cluster to which the target road section belongs in the merged congestion clusters, and then determines the congestion contribution score, thereby improving the accuracy of the congestion contribution score.

[0223] In one embodiment, S806 may specifically include: determining the frequency of congestion in the target section at different time periods; the frequency is a positive integer greater than or equal to two; determining the product value between the frequency and the length of the target section; determining the sum of the products of the length of each congested section in the target congestion cluster and the frequency of congestion in each congested section in the target congestion cluster at different time periods; and determining the congestion contribution score based on the product value and the sum value.

[0224] The frequency of congestion on the target road section in different time periods may be the number of time periods in which congestion occurs on the target road section in all different time periods. Figure 4a As shown, the target section of 100-200 meters in Road A was congested in 5 different time periods, and the frequency of congestion in the target section in different time periods was 5.

[0225] The server determines the product value between the frequency and the length of the target road segment by multiplying the number of time periods when the target road segment is congested at different time periods by the length of the target road segment to obtain the product value. Figure 4a As shown, the target road section is a road section of 100-200 meters, the frequency of congestion in the target road section is 5, and the length is 100, so the product value obtained by the server is 500.

[0226] In one embodiment, the server determines the product value between the frequency and the length of the target road section, which may include: the server obtains the congestion index mark value of the target road section in different time periods, and then calculates the product value of the congestion index mark value of the target road section in different time periods and the length of the road section, and uses the obtained product value as the product value between the frequency of congestion in the target road section in different time periods and the length of the target road section.

[0227] For example, the server determines the product value between the frequency and the length of the target road segment by formula (1), where s0 is the determined product value and T is the set of all different time periods from the start time to the end time of the target congestion cluster. i,j is the mark value of the congestion index of target road section i in time period j. When target road section i is congested in time period j, c i,j is 1, when the target road section i is not congested during the j period, c i,j is 0. i is the length of road section i.

[0228]

[0229] In one embodiment, the server first determines the frequency of congestion in the target road section at different time periods, determines the product of the frequency and the length of the target road section, and then determines the sum of the products of the length of each congested road section in the target congestion cluster and the frequency of congestion in each congested road section in the target congestion cluster at different time periods, and determines the congestion contribution score based on the ratio of the product value to the sum.

[0230] In one embodiment, the server first calculates the product of the congestion index mark value corresponding to each road section in the congestion cluster and the road section length, and sums the product values ​​obtained for each road section to obtain the cluster size of the congestion cluster. Then, the product value between the congestion index mark value corresponding to the target road section in different time periods and the length of the target road section is determined. Finally, the server calculates the ratio of the above product value to the cluster size, and uses the ratio as the congestion contribution score. For example, the server calculates the congestion contribution score according to formula (3), where R i is the congestion contribution score of the i-th target road segment, L is the congestion cluster composed of each road segment, T is the set of all time periods from the start time to the end time of the congestion cluster, c i,j is the mark value of the congestion index of road section i in time period j. When road section i is a congested road section in time period j, c i,j is 1, when the road section i is a smooth road section in the j period, c i,j is 0.

[0231] l i is the length of section i

[0232]

[0233] In another embodiment, the server inputs the product value and the sum value into a calculation model, and obtains the congestion contribution score through the calculation model. The calculation model may be a CNN (Convolutional Neural Networks) model, an RNN (Residual Neural Networks), a DL (Deep Learning) model, etc.

[0234] In the above embodiment, the server determines the congestion contribution score based on the product value corresponding to the target road section and the sum of the product values ​​corresponding to each congested road section, so that the determined congestion contribution score is proportional to the frequency of congestion occurring in the target road section at different time periods, thereby making the determined congestion contribution score more accurate and reasonable.

[0235] After determining the congestion bottleneck point, the server can also determine the congestion range of the congestion bottleneck point. Specifically, when the road network is a linear road network, the length accumulation value corresponding to the congested road sections in the congestion cluster that are congested during each congestion period is determined, wherein the road network topology map includes a topology map constructed based on road network data corresponding to different road networks; or, when the road network is a mesh road network, the length accumulation value corresponding to the congested road sections in the congestion cluster that are congested during each congestion period and belong to the same road is determined; according to the length accumulation value corresponding to the congestion cluster, the congestion range of the congestion bottleneck point corresponding to the congestion cluster is determined.

[0236] Among them, the congestion range is the distribution range of the congested sections in the congestion cluster in the spatial dimension. When the road network is a linear road network, since the roads in the linear road network have no intersections in the two-dimensional plane, the congested sections corresponding to each time period in the congestion cluster are distributed in one-dimensional space. The server calculates the length accumulation value corresponding to each congested section in one spatial dimension, and determines the congestion range based on the length accumulation value calculated in one spatial dimension. When the road network is a mesh road network, each road in the mesh road network has an intersection in the two-dimensional plane, and there are multiple roads in the road network in the two-dimensional plane. The congested sections corresponding to each time period in the congestion cluster are distributed in two-dimensional space. The server calculates the length accumulation value corresponding to each congested section belonging to the same road in two spatial dimensions, and determines the congestion range based on the length accumulation value calculated in the two spatial dimensions.

[0237] Among them, the above-mentioned determining the congestion range of the congestion bottleneck point corresponding to the congestion cluster according to the length accumulation value corresponding to the congestion cluster may specifically include: the server determines the congestion range according to the maximum length accumulation value, or may also determine the congestion range according to the average value of the length accumulation values ​​corresponding to each time period.

[0238] For example, when the road network is a linear road network, Fig. 9 As shown, congested sections 1-12 form a congestion cluster. For the congested sections in the congestion cluster, the corresponding length accumulation values ​​in different time periods from 30 minutes to 70 minutes are 100 meters, 300 meters, 400 meters, 300 meters and 100 meters respectively. Among them, the congested sections corresponding to the maximum length accumulation value of 400 meters are distributed in the range of 100 meters to 500 meters. The server determines that the congestion range is 100 meters to 500 meters.

[0239] For example, when the road network is a mesh road network, Figures 10a to 10dThe following are the distribution diagrams of congested sections in the congestion cluster in two-dimensional space in four different time periods within 10min-40min. In the 0-10min period, the maximum cumulative length value in the x-direction dimension is 300; in the y-direction dimension, the maximum cumulative length value is 300. In the 20min period, the maximum cumulative length value in the x-direction dimension is 200; in the y-direction dimension, the maximum cumulative length value is 300. In the 30min period, the maximum cumulative length value in the x-direction dimension is 400; in the y-direction dimension, the maximum cumulative length value is 400. In the 40min period, the maximum cumulative length value in the x-direction dimension is 400; in the y-direction dimension, the maximum cumulative length value is 500. The server determines the congestion range in the x direction based on the accumulated value of the maximum length in the x direction in different time periods, and determines the congestion range in the y direction based on the accumulated value of the maximum length in the y direction in different time periods. The determined congestion range is 100 to 400 meters in the x direction and 100 to 500 meters in the y direction.

[0240] In the above embodiment, the server determines the length accumulation value and determines the congestion range of the congestion bottleneck point according to the length accumulation value, so as to accurately determine the congestion location, facilitate the control of the congestion bottleneck point, and improve the efficiency of the control of the congestion bottleneck point.

[0241] In one embodiment, as shown in Figure 11, the method for determining a congestion bottleneck point includes the following steps:

[0242] S1102, obtaining road condition data of different time periods and a road network topology map constructed based on the road network data.

[0243] S1104, based on the traffic data, generating a traffic distribution map of each road section in the road network topology map at different time periods; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods.

[0244] S1106: Select, from among the road sections, a target road section that is congested in at least two time periods according to the traffic distribution map.

[0245] S1108, clustering the congested road sections that are congested at different time periods in the road network topology map to obtain at least two congestion clusters.

[0246] S1110, determining a target congestion cluster to which the target road section belongs among at least two congestion clusters;

[0247] S1112: Determine the congestion contribution score of the target road section in the target congestion cluster.

[0248] S1114, determining a congestion bottleneck point based on the target road section whose congestion contribution score meets a preset condition.

[0249] S1116, determining the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration of the congestion cluster, and determining the congestion range of the congestion bottleneck point corresponding to the congestion cluster according to the accumulated value of the length corresponding to the congestion cluster.

[0250] The specific steps of S1102 to S1116 above can be referred to Figure 2 Implementation and Figure 8 Example.

[0251] The present application also provides a traffic control application scenario, which applies the above-mentioned congestion bottleneck point determination method, and specifically includes the following contents:

[0252] The server obtains the topological data of each road section in the road network and generates a road network topology map based on the topological data of each road section. When the road network is a linear road network, the server generates a road network topology map based on the connection edge as the basic unit; when the road network is a mesh road network, the server generates a road network topology map based on the connection segment as the basic unit.

[0253] The vehicles traveling in the road network obtain location information through the positioning system and upload the location information to the server. The server preprocesses the location information to ensure the integrity and correctness of the location information. Then, the server calculates the preprocessed location information to obtain the traffic data, and generates a traffic distribution map based on the traffic data to reflect whether each road section in the road network topology map is congested at different times, and displays it on the screen. The server clusters the congested sections in the road network topology map that are congested at different times according to the traffic distribution map to obtain at least two congestion clusters. The server selects the target road section whose congestion contribution score meets the preset conditions from the at least two congestion clusters, and determines the congestion bottleneck point based on the target road section. In addition, the server can also determine the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the congestion start time and congestion end time of the congestion cluster, and when the road network is a linear road network, determine the length accumulation value corresponding to the congested road sections in the congestion cluster that are congested in each congestion period; or when the road network is a mesh road network, determine the length accumulation value corresponding to the congested road sections in the congestion cluster that are congested in each congestion period and belong to the same road; according to the length accumulation value corresponding to the congestion cluster, determine the congestion range of the congestion bottleneck point corresponding to the congestion cluster. The server performs traffic control on the congestion bottleneck point according to the determined congestion type and congestion range.

[0254] It should be understood that although Figure 2 The steps in the flowcharts of , 6, 8, and 11 are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2At least part of the steps in , 6, 8, and 11 may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily one by one, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0255] In one embodiment, Fig.12 As shown, a device for determining a congestion bottleneck point is provided. The device can adopt a software module or a hardware module, or a combination of the two to become a part of a computer device. The device specifically includes: an acquisition module 1202, a generation module 1204, a selection module 1206, a score determination module 1208 and a bottleneck point determination module 1210, wherein:

[0256] An acquisition module 1202 is used to acquire road condition data of different time periods and a road network topology map constructed based on the road network data;

[0257] The generating module 1204 is used to generate a traffic distribution map of each road section in the road network topology map at different time periods based on the traffic data; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods;

[0258] A selection module 1206 is used to select, from each road section in the road network topology map, a target road section that is congested in at least two time periods according to the road condition distribution map;

[0259] The score determination module 1208 is used to determine the congestion contribution score of the congestion cluster formed by the congested road sections of the target road section in the road network topology map;

[0260] The bottleneck point determination module 1210 is further configured to determine a congestion bottleneck point based on a target road section whose congestion contribution score meets a preset condition.

[0261] The above-mentioned method, device, computer equipment and storage medium for determining congestion bottleneck points construct a road network topology map based on road network data, and generate a road condition distribution map of each road section in the road network topology map at different time periods based on the road condition data of different time periods, so that it can be judged whether each road section in the road network topology map is congested at different time periods according to the road condition distribution map. According to the road condition distribution map, a target road section that is congested in at least two time periods can be selected, and the congestion contribution score of the congestion cluster formed by each congested road section in the road network topology map can be determined. The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches the preset condition, so that the congestion bottleneck point can be accurately determined, and the accuracy of determining the congestion bottleneck point is improved. In addition, since the road condition distribution map is a road condition map for each road section in the road network topology map, it can reflect whether each road section in a large area road network is congested at different time periods, so the congestion bottleneck point can be quickly found in the large area road network, and can be applied to the application scenario of finding congestion bottleneck points in large and medium-sized traffic road networks.

[0262] In one embodiment, the bottleneck point determination module 1210 is further configured to:

[0263] Sort the target road sections according to the congestion contribution scores;

[0264] Among the sorted target sections, select the target section whose ranking meets the preset conditions;

[0265] Determine the congestion bottleneck point based on the selected target road section.

[0266] In one embodiment, the road network topology map includes a topology map constructed based on road network data corresponding to different road networks; the bottleneck point determination module 1210 is further used to:

[0267] When the road network is a linear road network, the selected target road section is determined as the congestion bottleneck point;

[0268] When the road network is a mesh road network, the key road nodes associated with the selected target road section are found, and the key road nodes are used as congestion bottlenecks; alternatively, the selected target road section is clustered, and the intersection of the target road section is obtained according to the clustering result, and the intersection of the target road section is determined as the congestion bottleneck.

[0269] In one embodiment, the bottleneck point determination module 1210 is further configured to:

[0270] When the road network is a linear road network, determine the distance between each selected target road segment;

[0271] Connect target road sections whose distances to each other are less than a preset distance;

[0272] The connected target road section is determined as the congestion bottleneck point.

[0273] In one embodiment, Fig.13 As shown, the device also includes:

[0274] A clustering module 1212 is used to cluster the congested road sections in the road network topology map that are congested in different time periods to obtain at least two congestion clusters;

[0275] The score determination module 1208 is further configured to:

[0276] Determine, among the at least two congestion clusters, a target congestion cluster to which the target road segment belongs;

[0277] Determine a congestion contribution score of the target road segment in the target congestion cluster.

[0278] In one embodiment, the score determination module 1208 is further configured to:

[0279] Determine the frequency of congestion occurring on the target road section in different time periods; the frequency is a positive integer greater than or equal to two;

[0280] Determining a product value between the frequency and the length of the target section;

[0281] Determine the sum of the product values ​​of the length of each congested road section in the target congestion cluster and the frequency of congestion occurring in each congested road section in the target congestion cluster in different time periods;

[0282] A congestion contribution score is determined based on the product value and the sum value.

[0283] In one embodiment, the clustering module 1212 is further configured to:

[0284] For the first time period, searching for the congested road sections in the road network topology graph that are congested in the first time period and are interconnected, and taking the interconnected congested road sections as sub-congestion clusters;

[0285] For time periods other than the first one, the congested road sections in the road network topology map that are congested and interconnected in the current time period other than the first one are superimposed with the sub-congestion cluster, until all congested road sections corresponding to the time period other than the first one and interconnected are superimposed with the sub-congestion cluster, to obtain at least two congestion clusters.

[0286] In one embodiment, the apparatus further comprises:

[0287] An acquisition module 1202 is used to acquire a congestion cluster formed based on each congested road section in the road network topology map;

[0288] A frequency determination module 1214, configured to determine the frequency of congestion occurring in each congested road section within the congestion cluster at different time periods;

[0289] A product value determination module 1216, configured to determine a product value between the frequency and the length of the corresponding congested road section in the congestion cluster;

[0290] The cluster size determination module 1218 is used to use the sum of the determined product values ​​as the cluster size of the congestion cluster, and the cluster size is used to measure the congestion degree of each congested road section in the congestion cluster.

[0291] In one embodiment, the apparatus further comprises:

[0292] Duration determination module 1220, used to determine the duration of the congestion cluster;

[0293] A screening module 1222, configured to screen the congested clusters according to the cluster size and the duration;

[0294] A merging module 1224, configured to merge the filtered congested clusters to obtain a merged congested cluster;

[0295] The score determination module 1208 is further configured to determine the congestion contribution score of the target road segment to the merged congestion cluster to which it belongs.

[0296] In one embodiment, the apparatus further comprises:

[0297] A time determination module 1226, configured to determine a congestion start time and a congestion end time of the congestion cluster;

[0298] A duration determination module 1220, configured to determine the duration of the congestion cluster based on the congestion start time and the congestion end time;

[0299] The type determination module 1228 is configured to determine the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

[0300] In one embodiment, the road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; the device also includes:

[0301] The accumulated value determination module 1230 is used to determine the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period when the road network is a linear road network; or, when the road network is a mesh road network, to determine the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period and belong to the same road; and determine the congestion range of the congestion bottleneck point corresponding to the congestion cluster based on the accumulated length values ​​corresponding to the congestion cluster.

[0302] In one embodiment, the traffic condition distribution map includes a traffic condition spatiotemporal distribution map; the generating module 1204 is further used to:

[0303] Based on the traffic data, a traffic spatial distribution diagram of each road section in the road network topology diagram in each different time period is generated respectively;

[0304] The road condition spatial distribution map of each different time period is combined according to the time dimension to obtain the road condition spatiotemporal distribution map.

[0305] In one embodiment, the bottleneck point determination module 1210 is further configured to:

[0306] Taking the target road section whose congestion contribution score reaches the preset condition as the root node, the adjacent nodes are expanded layer by layer to the remaining target road sections as child nodes, and a road section node tree consisting of the root node and the expanded adjacent nodes is obtained;

[0307] Determine the congestion bottleneck point based on the target road sections corresponding to the root node to the leaf node in the road section node tree.

[0308] In one embodiment, the bottleneck point determination module 1210 is further configured to:

[0309] The target road section whose congestion contribution score meets the preset conditions is taken as the root node, and the adjacent nodes are expanded layer by layer to the remaining target road sections as child nodes;

[0310] When there is an adjacent node whose congestion contribution score is greater than a preset value among the expanded adjacent nodes, the expanded adjacent node is used as an intermediate node, and a target intermediate node is selected from the intermediate nodes to continue expanding until there is no adjacent node whose congestion contribution score is greater than the preset value among the expanded adjacent nodes;

[0311] When there is no adjacent node with a congestion contribution score greater than a preset value among the expanded adjacent nodes, a road segment node tree is constructed based on the root node and all the expanded adjacent nodes.

[0312] For the specific definition of the congestion bottleneck point determination device, please refer to the definition of the congestion bottleneck point determination method above, which will not be repeated here. Each module in the above-mentioned congestion bottleneck point determination device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0313] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig.14As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store congestion bottleneck point determination data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a congestion bottleneck point determination method is implemented.

[0314] Those skilled in the art will understand that Fig.14 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0315] In one embodiment, a computer device is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: obtaining road condition data of different time periods and a road network topology map constructed based on the road network data; generating a road condition distribution map of each road section in the road network topology map at different time periods based on the road condition data; the road condition distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods; among each road section in the road network topology map, selecting a target road section that is congested in at least two time periods according to the road condition distribution map; determining a congestion contribution score of the target road section to a congestion cluster formed based on each congested road section in the road network topology map; and determining a congestion bottleneck point based on the target road section whose congestion contribution score reaches a preset condition.

[0316] In one embodiment, when the processor executes the computer program, the following steps are also implemented: sorting the target road sections according to the size of the congestion contribution score; selecting the target road sections whose ranking meets the preset conditions among the sorted target road sections; and determining the congestion bottleneck point based on the selected target road sections.

[0317] In one embodiment, the road network topology map includes a topology map constructed based on road network data corresponding to different road networks; when the processor executes the computer program, the following steps are also implemented: when the road network is a linear road network, the selected target road section is determined as a congestion bottleneck point; when the road network is a mesh road network, the key road nodes associated with the selected target road section are found, and the key road nodes are used as congestion bottleneck points; or, the selected target road sections are clustered, and the target road section intersections are obtained according to the clustering results, and the target road section intersections are determined as congestion bottleneck points.

[0318] In one embodiment, when the processor executes the computer program, the following steps are also implemented: when the road network is a linear road network, the distance between each selected target road section is determined; the target road sections whose distances are less than a preset distance are connected; and the connected target road sections are determined as congestion bottleneck points.

[0319] In one embodiment, when the processor executes the computer program, the following steps are also implemented: clustering the congested road sections that are congested at different time periods in the road network topology map to obtain at least two congestion clusters; determining the target congestion cluster to which the target road section belongs among the at least two congestion clusters; and determining the congestion contribution score of the target road section in the target congestion cluster.

[0320] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining the frequency of congestion in the target section at different time periods; the frequency is a positive integer greater than or equal to two; determining the product value between the frequency and the length of the target section; determining the sum of the products of the length of each congested section in the target congestion cluster and the frequency of congestion in each congested section in the target congestion cluster at different time periods; and determining the congestion contribution score based on the product value and the sum value.

[0321] In one embodiment, when the processor executes the computer program, the following steps are also implemented: for the first time period, the congested road sections in the road network topology map that are congested in the first time period and interconnected are searched, and the interconnected congested road sections are used as sub-congestion clusters; for non-first time periods, the congested road sections in the road network topology map that are congested in the current non-first time period and interconnected are superimposed with the sub-congestion clusters, until all the congested road sections corresponding to the non-first time periods and interconnected are superimposed with the sub-congestion clusters, so as to obtain at least two congestion clusters.

[0322] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining a congestion cluster formed based on each congested road section in the road network topology map; determining the frequency of congestion in each congested road section in the congested cluster at different time periods; determining the product value between the frequency and the length of the corresponding congested road section in the congested cluster; and using the sum of the determined product values ​​as the cluster size of the congested cluster, and the cluster size is used to measure the congestion degree of the congested road section in the congested cluster.

[0323] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining the duration of a congestion cluster; screening the congestion clusters according to the cluster size and duration; merging the screened congestion clusters to obtain a merged congestion cluster; determining the congestion contribution score of the target road section to the congestion cluster formed based on the congested road sections in the road network topology map includes: determining the congestion contribution score of the target road section to the merged congestion cluster to which it belongs.

[0324] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining the congestion start time and congestion end time of the congestion cluster; determining the duration of the congestion cluster based on the congestion start time and congestion end time; and determining the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

[0325] In one embodiment, a road network topology map includes a topology map constructed based on road network data corresponding to different road networks; when the processor executes the computer program, the following steps are also implemented: when the road network is a linear road network, the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period are determined; or, when the road network is a mesh road network, the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period and belong to the same road are determined; based on the accumulated length values ​​corresponding to the congestion cluster, the congestion range of the congestion bottleneck point corresponding to the congestion cluster is determined.

[0326] In one embodiment, the road condition distribution map includes a road condition spatiotemporal distribution map; when the processor executes the computer program, the following steps are also implemented: based on the road condition data, a road condition spatial distribution map of each road section in the road network topology map is generated in each different time period; the road condition spatial distribution map of each different time period is combined according to the time dimension to obtain a road condition spatiotemporal distribution map.

[0327] In one embodiment, when the processor executes the computer program, the following steps are also implemented: taking the target road section whose congestion contribution score meets the preset conditions as the root node, expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes, and obtaining a road section node tree composed of the root node and the expanded adjacent nodes; determining the congestion bottleneck point based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree.

[0328] In one embodiment, when the processor executes the computer program, the following steps are also implemented: taking the target road section whose congestion contribution score reaches the preset condition as the root node, and expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes; when there is an adjacent node with a congestion contribution score greater than a preset value among the expanded adjacent nodes, taking the expanded adjacent nodes as intermediate nodes, and selecting target intermediate nodes from the intermediate nodes to continue expanding until there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes; when there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, forming a road section node tree based on the root node and all expanded adjacent nodes.

[0329] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: obtaining traffic data of different time periods and a road network topology map constructed based on the road network data; generating a traffic condition distribution map of each road section in the road network topology map at different time periods based on the traffic condition data; the road condition distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods; among each road section in the road network topology map, selecting a target road section that is congested in at least two time periods according to the road condition distribution map; determining a congestion contribution score of the target road section to a congestion cluster formed based on each congested road section in the road network topology map; and determining a congestion bottleneck point based on the target road section whose congestion contribution score reaches a preset condition.

[0330] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: sorting the target road sections according to the size of the congestion contribution score; selecting the target road sections whose ranking meets the preset conditions among the sorted target road sections; and determining the congestion bottleneck point based on the selected target road sections.

[0331] In one embodiment, the road network topology map includes a topology map constructed based on road network data corresponding to different road networks; when the computer program is executed by the processor, the following steps are also implemented: when the road network is a linear road network, the selected target road section is determined as a congestion bottleneck point; when the road network is a mesh road network, the key road nodes associated with the selected target road section are found, and the key road nodes are used as congestion bottleneck points; or, the selected target road sections are clustered, and the target road section intersections are obtained according to the clustering results, and the target road section intersections are determined as congestion bottleneck points.

[0332] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when the road network is a linear road network, the distance between each selected target road section is determined; the target road sections whose distances are less than a preset distance are connected; and the connected target road sections are determined as congestion bottleneck points.

[0333] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: clustering the congested road sections that are congested at different time periods in the road network topology map to obtain at least two congestion clusters; determining the target congestion cluster to which the target road section belongs among the at least two congestion clusters; and determining the congestion contribution score of the target road section in the target congestion cluster.

[0334] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the frequency of congestion in the target road section at different time periods; the frequency is a positive integer greater than or equal to two; determining the product value between the frequency and the length of the target road section; determining the sum of the products of the length of each congested road section in the target congestion cluster and the frequency of congestion in each congested road section in the target congestion cluster at different time periods; and determining the congestion contribution score based on the product value and the sum value.

[0335] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: for the first time period, the congested road sections in the road network topology map that are congested in the first time period and interconnected are searched, and the interconnected congested road sections are used as sub-congestion clusters; for non-first time periods, the congested road sections in the road network topology map that are congested in the current non-first time period and interconnected are superimposed with the sub-congestion clusters, until all the congested road sections corresponding to the non-first time periods and interconnected are superimposed with the sub-congestion clusters, so as to obtain at least two congestion clusters.

[0336] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a congestion cluster formed based on each congested road section in the road network topology map; determining the frequency of congestion in each congested road section in the congested cluster at different time periods; determining the product value between the frequency and the length of the corresponding congested road section in the congested cluster; and using the sum of the determined product values ​​as the cluster size of the congested cluster, and the cluster size is used to measure the congestion degree of the congested road section in the congested cluster.

[0337] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the duration of a congestion cluster; screening the congestion clusters according to the cluster size and duration; merging the screened congestion clusters to obtain a merged congestion cluster; determining the congestion contribution score of the target road section to the congestion cluster formed based on the congested road sections in the road network topology map includes: determining the congestion contribution score of the target road section to the merged congestion cluster to which it belongs.

[0338] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the congestion start time and congestion end time of the congestion cluster; determining the duration of the congestion cluster based on the congestion start time and congestion end time; and determining the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

[0339] In one embodiment, a road network topology map includes a topology map constructed based on road network data corresponding to different road networks; when the computer program is executed by a processor, the following steps are also implemented: when the road network is a linear road network, the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period are determined; or, when the road network is a mesh road network, the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period and belong to the same road are determined; based on the accumulated length values ​​corresponding to the congestion cluster, the congestion range of the congestion bottleneck point corresponding to the congestion cluster is determined.

[0340] In one embodiment, the road condition distribution map includes a road condition spatiotemporal distribution map; when the computer program is executed by the processor, the following steps are also implemented: based on the road condition data, a road condition spatial distribution map of each road section in the road network topology map is generated in each different time period; the road condition spatial distribution map of each different time period is combined according to the time dimension to obtain a road condition spatiotemporal distribution map.

[0341] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: taking the target road section whose congestion contribution score reaches the preset conditions as the root node, expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes, and obtaining a road section node tree composed of the root node and the expanded adjacent nodes; determining the congestion bottleneck point based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree.

[0342] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: taking the target road section whose congestion contribution score reaches the preset condition as the root node, and expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes; when there is an adjacent node with a congestion contribution score greater than a preset value among the expanded adjacent nodes, taking the expanded adjacent nodes as intermediate nodes, and selecting target intermediate nodes from the intermediate nodes to continue expanding until there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes; when there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, forming a road section node tree based on the root node and all expanded adjacent nodes.

[0343] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0344] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application 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 memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0345] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 specification.

[0346] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for determining a congestion bottleneck point, characterized in that: The method comprises: Obtain traffic data at different time periods and a road network topology map based on the road network data; Based on the traffic data, a traffic distribution map of each road section in the road network topology map at different time periods is generated; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods; Among the road sections of the road network topology map, select a target road section that is congested in at least two of the time periods according to the road condition distribution map; Clustering the congested road sections in the road network topology graph that are congested at different time periods to obtain at least two congestion clusters; Determining the congestion contribution score of the target road section to the congestion cluster formed based on each congested road section in the road network topology map, including: determining the target congestion cluster to which the target road section belongs among the at least two congestion clusters; determining the frequency of congestion occurring in the target road section in different time periods; the frequency is a positive integer greater than or equal to 2; determining the product value between the frequency and the length of the target road section; determining the sum of the product values ​​between the length of each congested road section in the target congestion cluster and the frequency of congestion occurring in each congested road section in the target congestion cluster in different time periods; determining the congestion contribution score based on the product value and the sum value; or determining the congestion contribution score of the target road section to the merged congestion cluster to which it belongs, the merged congestion cluster being obtained by merging the filtered congestion clusters, and the filtered congestion clusters being obtained by filtering according to the cluster size and duration of the congestion clusters; The congestion bottleneck point is determined based on the target road section whose congestion contribution score reaches a preset condition.

2. The method according to claim 1, characterized in that The determining of the congestion bottleneck point based on the target road section whose congestion contribution score reaches a preset condition comprises: sorting the target road sections according to the congestion contribution scores; Among the sorted target sections, select the target section whose ranking meets the preset conditions; A congestion bottleneck point is determined based on the selected target road section.

3. The method according to claim 2, characterized in that The road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; The determining of the congestion bottleneck point based on the selected target road section includes: When the road network is a linear road network, the selected target road section is determined as a congestion bottleneck point; When the road network is a mesh road network, the key road nodes associated with the selected target road section are found, and the key road nodes are used as the congestion bottleneck points; or, the selected target road section is clustered, and the target road section intersection points are obtained according to the clustering results, and the target road section intersection points are determined as the congestion bottleneck points.

4. The method according to claim 3, characterized in that When the road network is a linear road network, determining the selected target road section as a congestion bottleneck point includes: When the road network is a linear road network, determining the distance between each selected target road section; Connecting target road sections whose distances to each other are less than a preset distance; The connected target road section is determined as a congestion bottleneck point.

5. The method according to claim 1, characterized in that The method further comprises: Determine the congestion start time and congestion end time of the congestion cluster; Determining the duration of the congestion cluster based on the congestion start time and the congestion end time; Determine a congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

6. The method according to claim 1, characterized in that The traffic condition distribution map includes a traffic condition spatiotemporal distribution map; the generating of the traffic condition distribution map of each road section in the road network topology map at different time periods based on the traffic condition data includes: Based on the traffic data, a traffic spatial distribution diagram of each road section in the road network topology diagram in each different time period is generated respectively; The road condition spatial distribution map of each different time period is combined according to the time dimension to obtain the road condition spatiotemporal distribution map.

7. The method according to claim 1, characterized in that The clustering of the congested road sections in the road network topology graph that are congested in different time periods to obtain at least two congestion clusters includes: For the first time period, searching for the congested road sections in the road network topology graph that are congested in the first time period and are interconnected, and taking the interconnected congested road sections as sub-congestion clusters; For time periods other than the first one, the congested road sections in the road network topology map that are congested and interconnected in the current time period other than the first one are superimposed with the sub-congestion cluster, until all congested road sections corresponding to the time period other than the first one and interconnected are superimposed with the sub-congestion cluster, to obtain at least two congestion clusters.

8. The method according to claim 1, characterized in that The method further comprises: Obtaining a congestion cluster formed based on each congested road section in the road network topology map; Determine the frequency of congestion occurring in each congested road section within the congestion cluster at different time periods; Determining a product value between the frequency and the length of the corresponding congested road section in the congestion cluster; The sum of the determined product values ​​is used as the cluster size of the congestion cluster, and the cluster size is used to measure the congestion degree of the congested road section in the congestion cluster.

9. The method according to claim 8, characterized in that The method further comprises: determining a duration of the congestion cluster; Filtering the congested clusters according to the cluster size and the duration; The filtered congested clusters are merged to obtain a merged congested cluster.

10. The method according to any one of claims 1 to 9, characterized in that The road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; The method further comprises: When the road network is a linear road network, determining the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested during each congestion period; or, When the road network is a mesh road network, determining the accumulated length values ​​corresponding to the congested road sections in the congestion cluster that are congested in each congestion period and belong to the same road; The congestion range of the congestion bottleneck point corresponding to the congestion cluster is determined according to the accumulated value of the length corresponding to the congestion cluster.

11. The method according to claim 1, characterized in that: The determining of the congestion bottleneck point based on the target road section whose congestion contribution score reaches a preset condition comprises: Taking the target road section whose congestion contribution score reaches the preset condition as the root node, expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes, and obtaining a road section node tree composed of the root node and the expanded adjacent nodes; Based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree, the congestion bottleneck point is determined.

12. The method according to claim 11, characterized in that The target road segment whose congestion contribution score reaches the preset condition is taken as the root node, and the adjacent nodes are expanded layer by layer to the remaining target road segments as child nodes, so as to obtain a road segment node tree composed of the root node and the expanded adjacent nodes, including: Taking the target road section whose congestion contribution score reaches the preset condition as the root node, and expanding the adjacent nodes layer by layer to the remaining target road sections as child nodes; When there is an adjacent node whose congestion contribution score is greater than a preset value among the expanded adjacent nodes, the expanded adjacent node is used as an intermediate node, and a target intermediate node is selected from the intermediate nodes to continue expanding until there is no adjacent node whose congestion contribution score is greater than the preset value among the expanded adjacent nodes; When there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, the road segment node tree is constructed based on the root node and all the expanded adjacent nodes.

13. A device for determining a congestion bottleneck point, characterized in that: The device comprises: An acquisition module is used to obtain road condition data of different time periods and a road network topology map constructed based on the road network data; A generating module, for generating a traffic distribution map of each road section in the road network topology map at different time periods based on the traffic data; the traffic distribution map is used to reflect whether each road section in the road network topology map is congested at different time periods; A selection module, configured to select, from each road section of the road network topology map, a target road section that is congested in at least two of the time periods according to the road condition distribution map; A clustering module, used for clustering the congested road sections in the road network topology map that are congested at different time periods to obtain at least two congestion clusters; A score determination module is used to determine, among the at least two congestion clusters, the target congestion cluster to which the target road segment belongs; determine the frequency of congestion occurring in the target road segment in different time periods; the frequency is a positive integer greater than or equal to 2; determine the product value between the frequency and the length of the target road segment; determine the sum of the product values ​​between the length of each congested road segment in the target congestion cluster and the frequency of congestion occurring in each congested road segment in the target congestion cluster in different time periods; determine the congestion contribution score based on the product value and the sum value; or determine the congestion contribution score of the target road segment to the merged congestion cluster to which it belongs, the merged congestion cluster being obtained by merging the filtered congestion clusters, which are obtained by filtering according to the cluster size and duration of the congestion clusters; The bottleneck point determination module is also used to determine the congestion bottleneck point based on the target road section whose congestion contribution score meets the preset conditions.

14. The device according to claim 13, characterized in that The bottleneck point determination module is further used to sort the target road sections according to the size of the congestion contribution score; and select the target road sections whose ranking meets the preset conditions from the sorted target road sections; A congestion bottleneck point is determined based on the selected target road section.

15. The device according to claim 14, characterized in that The road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; The bottleneck point determination module is also used to determine the selected target road section as the congestion bottleneck point when the road network is a linear road network; when the road network is a mesh road network, find the key road nodes associated with the selected target road section and use the key road nodes as the congestion bottleneck point; or cluster the selected target road sections, obtain the target road section intersection points according to the clustering results, and determine the target road section intersection points as the congestion bottleneck points.

16. The device according to claim 15, characterized in that The bottleneck point determination module is also used to determine the distance between each selected target road section when the road network is a linear road network; connect the target road sections whose distances are less than a preset distance; and determine the connected target road sections as congestion bottleneck points.

17. The device according to claim 13, characterized in that The device also includes: A time determination module, used to determine the congestion start time and congestion end time of the congestion cluster; A duration determination module, configured to determine the duration of the congestion cluster based on the congestion start time and the congestion end time; A type determination module is used to determine the congestion type of the congestion bottleneck point corresponding to the congestion cluster according to the duration.

18. The device according to claim 13, characterized in that The traffic condition distribution map includes a traffic condition time-space distribution map; The generation module is also used to generate, based on the road condition data, a road condition spatial distribution map of each road section in the road network topology map in each different time period; and combine the road condition spatial distribution map of each different time period according to the time dimension to obtain the road condition spatiotemporal distribution map.

19. The device according to claim 13, characterized in that The clustering module is also used to search for the congested road sections in the road network topology map that are congested in the first time period and interconnected, and use the interconnected congested road sections as sub-congestion clusters; for non-first time periods, superimpose the congested road sections in the road network topology map that are congested in the current non-first time period and interconnected with the sub-congestion clusters, until all congested road sections corresponding to non-first time periods and interconnected are superimposed with the sub-congestion clusters, so as to obtain at least two congestion clusters.

20. The device according to claim 13, characterized in that The device also includes: An acquisition module is used to acquire a congestion cluster formed based on each congested road section in the road network topology map; A frequency determination module, used to determine the frequency of congestion occurring in each congested road section within the congestion cluster at different time periods; a product value determination module, used to determine a product value between the frequency and the length of the corresponding congested road section in the congestion cluster; The cluster size determination module is used to use the sum of the determined product values ​​as the cluster size of the congestion cluster, and the cluster size is used to measure the congestion degree of the congested road section in the congestion cluster.

21. The device according to claim 20, characterized in that The device also includes: A duration determination module, used to determine the duration of the congestion cluster; A screening module, used for screening the congested clusters according to the cluster size and the duration; The merging module is used to merge the congested clusters after screening to obtain a merged congested cluster.

22. The device according to any one of claims 13 to 21, characterized in that The road network topology map includes a topology map constructed based on the road network data corresponding to different road networks; The device also includes: The cumulative value determination module is used to determine the cumulative value of the length corresponding to the congested road sections in the congestion cluster that are congested during each congestion period when the road network is a linear road network; or, when the road network is a mesh road network, determine the cumulative value of the length corresponding to the congested road sections in the congestion cluster that are congested during each congestion period and belong to the same road; and determine the congestion range of the congestion bottleneck point corresponding to the congestion cluster based on the cumulative value of the length corresponding to the congestion cluster.

23. The device according to claim 13, characterized in that The bottleneck point determination module is further configured to use the target road section whose congestion contribution score meets the preset condition as a root node, and to expand the adjacent nodes layer by layer to the remaining target road sections as child nodes, so as to obtain a road section node tree composed of the root node and the expanded adjacent nodes; Based on the target road sections corresponding to the root node to the leaf nodes in the road section node tree, the congestion bottleneck point is determined.

24. The device according to claim 23, characterized in that The bottleneck point determination module is further configured to use the target road section whose congestion contribution score reaches a preset condition as a root node, and expand adjacent nodes layer by layer to the remaining target road sections as child nodes; When there is an adjacent node whose congestion contribution score is greater than a preset value among the expanded adjacent nodes, the expanded adjacent node is used as an intermediate node, and a target intermediate node is selected from the intermediate nodes to continue expanding until there is no adjacent node whose congestion contribution score is greater than the preset value among the expanded adjacent nodes; When there is no adjacent node with a congestion contribution score greater than the preset value among the expanded adjacent nodes, the road segment node tree is constructed based on the root node and all the expanded adjacent nodes.

25. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 12 are implemented.

26. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

27. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

Citation Information

Patent Citations

  • Road congestion bottleneck point determination method, apparatus, server and storage medium thereof

    CN106960571A

  • Traffic congestion tracing method and device, electronic equipment and storage medium

    CN112863176A