A Method for Highway Congestion Source Tracing and Control Based on Microscopic Traffic Simulation

Abstract

This invention discloses a method for tracing and controlling highway congestion based on microscopic traffic simulation. The method includes the following steps: S1. Acquiring the real road network and adjusting and correcting it using automated means, analyzing checkpoint data to identify peak-hour traffic flow distribution, reconstructing peak-hour traffic conditions, and importing the peak-hour traffic flow into simulation software; S2. Establishing evaluation indicators to determine the optimal segmentation length for road segments; S3. Dividing the target road segment into discrete segments based on the optimal segmentation length determined in step S2, and locating and tracing congested road segments based on the travel time traffic index algorithm; S4. Setting customized variable speed limits and ramp control combined highway management measures based on the congested road segments identified in step S3. This invention, based on road segment congestion analysis, enables precise control of congested road segments, improving the overall operational efficiency of highways and enhancing the overall operational status and service level of highways.

Description

Technical Field

[0001] This invention relates to the field of traffic management and control technology, and in particular to a method for tracing the source of road congestion and managing it. Background Technology

[0002] Highway interchanges and ramp entrances / exits are the main areas of traffic congestion, and the negative impacts of traffic congestion are particularly pronounced. To ensure normal highway operation, improve operational efficiency, and reduce the frequency of accidents and congestion, it is necessary to conduct real-time monitoring of severely congested sections of highways and implement road segment control measures.

[0003] Currently, highway management departments primarily rely on video surveillance supplemented by vehicle flow detectors for highway road monitoring. However, the monitoring area of ​​a single video surveillance device can only extend up to 2 kilometers. For highway management units that often manage thousands of kilometers of road, this limitation is significant and cannot achieve comprehensive monitoring. Furthermore, the typical distance between two adjacent vehicle flow detectors is tens of kilometers, hundreds of kilometers, or even more, making the frequency, timeliness, and density of data collection insufficient for real-time monitoring. Therefore, to facilitate testing of the effectiveness of control measures and save on development costs, a digital twin simulation vehicle operation platform based on a simulation environment has become an effective method for control testing. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a method for tracing the source of highway congestion and controlling it based on microscopic traffic simulation. By using simulation and deduction, the method analyzes and judges the road network congestion situation, implements corresponding control measures, and achieves the goal of alleviating highway congestion and improving the road network operation status.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] The present invention proposes a method for tracing the source of highway congestion and controlling its management based on microscopic traffic simulation, which includes the following steps:

[0007] S1. Obtain the real road network and adjust and correct it through automated means. Analyze the checkpoint data to identify the traffic flow distribution during peak hours, restore the traffic flow during peak hours, and import the peak hour traffic flow into the simulation software after aggregation.

[0008] S2. Based on the characteristics of road segment sub-units and combined with the evaluation index analysis of road segmentation, determine the road segmentation length.

[0009] S3. Based on the segmentation length determined in step S2, the target road segment is divided into discrete road segments. Based on the travel time traffic index algorithm, the travel time ratio of each road segment is extracted to locate and trace the source of congested road segments.

[0010] S4. Based on the congested road sections identified in step S3, set up customized variable speed limits and ramp control measures to improve the overall operational efficiency of the highway.

[0011] Furthermore, step S1 specifically includes the following steps:

[0012] Step S101: Obtain the road network of the study area based on OpenStreetMap, and generate the simulation road network file .net.xml and the visualization configuration file .poly.xml.

[0013] Step S102: Match the checkpoints to the simulated road network based on their latitude and longitude to determine their location within the network. Statistically analyze the traffic flow changes at each checkpoint during different time periods to obtain the peak-hour distribution of road segments. Aggregate the traffic flow at each checkpoint within a given time period. Based on the time distribution characteristics of the checkpoints, use the shortest path principle to connect the vehicles between checkpoints according to their license plate numbers to obtain the peak-hour traffic flow distribution. Then, convert this data into a .rou.xml file and import it into the simulation software.

[0014] Furthermore, step S2 specifically includes the following steps:

[0015] Step S201: In the simulation model, a flow detector is added at regular intervals along the study road segment. Each area at regular intervals along the study road segment is considered a sub-unit of the road segment. The simulation model is run, and the peak hourly flow data of each sub-unit of the simulated road network is output. Based on the output flow data, the flow changes of the sub-units of the road segment are analyzed, and whether there is a problem of excessive flow differences between the sub-units before and after the sub-units, in order to prepare for the implementation of step S202.

[0016] Step S202: Enumerate the segment lengths within a range. Using the flow detector set in step S201, output the flow corresponding to each segment length. Check the flow distribution of the segmented road sections. If it conforms to a negative binomial distribution, it can be retained. Compare the evaluation indicators under different segment lengths. It is found that when the segment length is 100m, the flow change between different road sections is small and there is no sudden change. Therefore, the optimal segment length should be greater than 100m. After comprehensive consideration, the final segment length is selected.

[0017] Furthermore, the characteristics of the road segment sub-units mentioned in step S2 include their peak hour traffic flow, whether they are interchanges, toll stations, ramps, or other non-main roads, and their road structure and linearity; the evaluation index system includes indicators of the degree of road segment feature display, segmentation accuracy, and post-processing workload.

[0018] Furthermore, in step S201, a flow detector will be added every 100m along the studied road section.

[0019] Furthermore, in step S202, the segmentation lengths from 100m to 1000m are enumerated.

[0020] Furthermore, step S3 specifically includes the following steps:

[0021] Step S301: For the main line, calculate the total length of the main line, and then combine it with the segmentation length of the research road segment selected in step S2 to determine the number of main line segmentation sections.

[0022] Step S302: Based on the three parameters of the total length of the main line, the segment length, and the number of segments of the main line, determine the latitude and longitude position of each segment in the simulated road network, and deploy them in each segment. Detector, measures different sections Average travel speed of vehicles during peak hours ;

[0023] Step S303: Calculate the traffic index for different sections based on the traffic index algorithm. Traffic congestion status based on travel speed :

[0024]

[0025]

[0026] In the formula, It is a section of road In time The travel speed ratio; In free-flow state, road segment travel speed; In actual circumstances, the road section In time Average travel speed;

[0027] Step S304: Based on traffic congestion status indicators Determine different sections The congestion index is used to determine the location of the congested section, and then the control strategy in step S4 is implemented.

[0028] Furthermore, different control strategies are set according to the location of the congestion:

[0029] (1) When the congestion occurs on the main road, a variable speed limit control strategy and a dynamic lane control strategy are adopted. The variable speed limit control strategy is based on the traffic congestion section statistics obtained in step S3, and the optimal speed limit value is selected through repeated simulation iterations. The dynamic lane control strategy updates the lane control strategy of the studied road section at a certain time interval Δt. market share When the threshold is exceeded, the open lane control strategy will be selected.

[0030] (2) When the congestion is located at the ramp, the ramp entrance and exit control strategy is adopted. The main contents of this strategy include: setting up stepped solid and dashed lines on the lanes near the ramp entrance and exit, setting dashed lines in front of the entrance and exit to allow lane changes, and setting up lanes near the entrance and exit to prohibit vehicles from changing lanes at will, so that vehicles entering the ramp can change lanes in advance and alleviate the congestion near the ramp entrance.

[0031] Furthermore, the aforementioned dynamic lane control strategy will time intervals. Set to 10 minutes.

[0032] Furthermore, in the aforementioned variable speed limit control strategy, the calculation formula and constraints for the variable speed limit value of the road segment are as follows:

[0033]

[0034]

[0035]

[0036]

[0037] in: It is the difference between the real-time occupancy rate and the optimal occupancy rate. The magnitude of the difference in occupancy rate shows the vehicle distribution and operating status of the current road segment. It is the expected flow rate downstream of the controlled area; Variable speed limit coefficient; It is the maximum speed limit for the road.

[0038] The present invention adopts the above technical solution, and its significant technical effects compared with the prior art are as follows:

[0039] 1. This invention utilizes a digital twin simulation platform for vehicle operation within a simulated environment, facilitating the testing of control measures' effectiveness and saving R&D costs. Furthermore, it imports periodic real checkpoint data into the simulated road network, accurately recreating the network's actual operating status and facilitating subsequent congestion analysis and source tracing.

[0040] 2. This invention integrates a road segmentation congestion tracing model and a traffic collaborative management model, thereby enabling the targeted addition of corresponding control measures. These measures are then implemented based on variable speed limits, ramp control, and vehicle right-of-way restrictions, achieving the goals of saving traffic scheduling resources, alleviating highway congestion, and improving the overall road network operation. (See attached figures for details.)

[0041] Figure 1 This is a schematic diagram of the overall implementation process of the present invention;

[0042] Figure 2 This is a traffic flow time distribution statistical chart of the present invention;

[0043] Figure 3 This is a schematic diagram of the road network segmentation points according to the present invention;

[0044] Figure 4 The present invention relates to a ramp entrance and exit control strategy. Detailed Implementation

[0045] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0046] To achieve the above objectives, this invention proposes a method for tracing the source of highway congestion and controlling its management based on microscopic traffic simulation, which improves the efficiency and accuracy of traffic management in congested sections of highways. For example... Figure 1 As shown in the overall implementation flowchart, the highway congestion source analysis and control method based on microscopic traffic simulation described in this invention includes the following steps:

[0047] S1. Obtain the real road network and adjust and correct it using automated methods. Analyze traffic flow information including license plate numbers, vehicle route points, vehicle types, and checkpoint data for the lanes where vehicles are located. Identify the traffic flow distribution during peak hours and reconstruct the traffic flow situation during peak hours. After aggregation, import the peak hour traffic flow into simulation software. This embodiment analyzes the congested section of the Beijing-Shanghai Expressway south of the Dongqiao Interchange and north of the West Exit. The specific steps are as follows:

[0048] First, the road network of a section of the Beijing-Shanghai Expressway was obtained using OpenStreetMap, generating a simulation road network file (.net.xml) and a visualization configuration file (.poly.xml). Then, the different road level (type) attributes of different road edges were obtained from the simulation road network file (.net.xml). Roads were filtered based on these attributes, eliminating redundant road networks other than expressways. Simultaneously, ramp road edges were added using the NetEdit editor based on CAD drawings. Finally, checkpoints were matched to the simulation road network based on their latitude and longitude to determine their locations within the network. Traffic flow changes at each checkpoint at different times were statistically analyzed to obtain data such as... Figure 2 The peak-hour distribution of the road segment shown is obtained by aggregating the traffic flow of each checkpoint during a peak period. Based on the time distribution characteristics of the checkpoints, the shortest path principle is used to connect the vehicles between the checkpoints according to their license plate numbers, and the complete travel routes of different vehicles are inferred. This allows us to obtain the traffic flow distribution during the peak period and convert it into a .rou.xml traffic flow information file for input into the simulation road network.

[0049] S2. Based on the characteristics of the road segment sub-units, including peak-hour traffic flow, whether it is an interchange, toll station, ramp, or other non-main road, and its road structure and linearity, the road segment length is determined by combining the evaluation indicators for road segmentation. The specific steps are as follows:

[0050] Step S201: Consider each 100m segment of the road under study as a sub-unit. The total length of the simulated road network is approximately 4km, therefore, it is divided into 40 sub-units. Flow detectors are installed at the corresponding road segment cross-sections. Run the simulation model and output the peak hourly flow data for each sub-unit of the simulated road network. Based on the output flow data, analyze the flow changes within the sub-units and whether there are excessively large differences in flow between consecutive sub-units, preparing for step S202.

[0051] Step S202: Enumerate segmentation lengths from 100m to 1000m, check whether the traffic flow distribution of the segmented road sections conforms to a negative binomial distribution, and compare the evaluation indicators under different segmentation lengths. It is found that when the segmentation length is 100m, the traffic flow changes between different road sections are small, and there are no sudden changes. Therefore, the optimal segmentation length should be greater than 100m. After comprehensive consideration, the final segmentation length is selected. and conduct road network such as Figure 3 The road segmentation is shown in the figure. An evaluation index system for road segmentation is established, with key indicators including the degree of road segment feature representation, segmentation accuracy, and post-processing workload.

[0052] The degree of road segment feature representation refers to whether road segment segmentation can distinguish different types of structures, different numbers of lanes, different lane widths, etc. In this case, the primary consideration is whether different types of structures can be distinguished. The specific feature calculation formula is as follows:

[0053]

[0054] In the formula, Indicators for displaying road segment characteristics Refers to the set of ramps in the studied road segment. Ramps that are separated from the main line segment and belong to the same section. The cross length (i.e., the length of the ramp within the study section that includes the main line). ramp The length.

[0055] Segmentation accuracy and post-processing workload are two corresponding indicators. Shorter segment lengths result in more segmented road segments, increasing the workload for later analysis, but allowing for more precise differentiation of congestion between different segments. Conversely, longer segment lengths result in a longer study area, making it easier to mask differences between segments, leading to lower accuracy but less workload. The specific formulas for these indicators are:

[0056]

[0057] In the formula, It is the set of all enumerated segment lengths. For accuracy indicators, It refers to a segment length of , It refers to a segment length of The number of road sections at that time.

[0058] According to traffic flow theory, the number of vehicles arriving on a road segment within a certain time interval is treated as a random number. Its statistical regularity can be described by three types of discrete distributions depending on the situation: a Poisson distribution for free-flowing conditions (no interaction between vehicles), a binomial distribution for congested conditions, and a negative binomial distribution for other conditions (such as those affected by peak traffic periods). Considering that the output is the traffic flow during peak hours, a negative binomial regression model is needed to test whether the average traffic flow of the segmented road segments conforms to a negative binomial distribution.

[0059] The final results are shown in Table 1. When the segmentation length is 500m, the accuracy and workload are well balanced, and the results are also more in line with the negative binomial distribution. Therefore, 500m was selected as the segmentation length.

[0060] surface Relationship between different segment lengths and evaluation index system

[0061]

[0062] S3, as determined in step S2 The target road segment is divided into discrete segments. Based on the travel time traffic index algorithm, the travel time ratio of each segment is extracted to locate and trace the source of congestion. Specifically, the steps include:

[0063] Step S301: For the main line section, calculate the total length of the main line. Then, in conjunction with step S2, the segment length of the research road section is selected. Determine the number of segments in the main line. .

[0064] Step S302: The total length of the main line can be obtained through step S301. 11500m, segment length Set to 500m, number of segments There are 23 [sections / areas]. These three parameters can be used to determine the latitude and longitude of each segment in the simulated road network, and [the network] can be deployed in each segment accordingly. Detector, measures different sections Average travel speed of vehicles during peak hours .

[0065] Step S303: Calculate the traffic index for different sections based on the traffic index algorithm. Traffic congestion status based on travel speed :

[0066]

[0067]

[0068] In the formula, It is a section of road In time The travel speed ratio; In free-flow state, road segment travel speed; In actual circumstances, the road section In time The average travel speed.

[0069] Based on the travel speed of each section, with 10 minutes as a research time unit, the traffic index of each analysis section was calculated using the method based on the travel speed ratio. Finally, the traffic flow status results for 3 hours were obtained. The correspondence between different ratio values ​​and traffic status is shown in Table 2.

[0070] surface Correspondence between different ratio values ​​and traffic conditions

[0071]

[0072] Step S304: Based on traffic congestion status indicators Determine different sections The congestion index was used to determine the location of congested sections. The average traffic index for each studied road segment was obtained by averaging the traffic flow status at different time units within the same segment, as shown in Table 3. It can be seen that the main highway section near the Dongqiao Interchange experiences mild congestion, with a congested section length of approximately 3.5 kilometers. Considering the upstream and downstream sections, the congestion is attributed to reduced capacity due to deceleration at the interchange ramps, resulting in discontinuous traffic flow. Based on this conclusion, corresponding control measures can be implemented.

[0073] surface Average traffic index of different sections

[0074]

[0075] S4. Based on the congested road sections identified in step S3, implement customized variable speed limits and ramp control measures to improve the overall operational efficiency of highways. This includes the following:

[0076] (1) When the congestion is located on the main road, a variable speed limit control strategy and a dynamic lane control strategy are adopted; the variable speed limit control strategy is to implement a variable speed limit control strategy on the congested section based on the traffic congestion section statistics obtained in step S3, and select the optimal speed limit value of the road through repeated simulation iterations; the dynamic lane control strategy is to set the time interval The lane control strategy for the research section is set to be updated every 10 minutes. market share When the threshold is exceeded, the open lane control strategy will be selected.

[0077] In the variable speed limit control strategy, the calculation formula and constraints for the variable speed limit value of a road segment are as follows:

[0078]

[0079]

[0080]

[0081]

[0082] in: It is the difference between the real-time occupancy rate and the optimal occupancy rate. The magnitude of the difference in occupancy rate shows the vehicle distribution and operating status of the current road segment. It is the expected flow rate downstream of the controlled area; Variable speed limit coefficient; It is the maximum speed limit for the road.

[0083] The following are some of the parameters used in this test: , , , .

[0084] (2) When the congestion occurs at a ramp, a ramp entrance / exit control strategy should be adopted; refer to Figure 4 As shown, the main contents of this strategy include: setting up stepped solid and dashed lines on the lanes near the ramp entrances and exits, setting dashed lines to allow lane changes before the entrances and exits, and setting up lane changes to prohibit vehicles from changing lanes at will on the road sections near the entrances and exits, so that vehicles entering the ramps can change lanes in advance and alleviate the congestion near the ramp entrances.

[0085] The effectiveness of the above control strategies was verified by traffic indicators. The changes in the indicators before and after implementing the variable speed limit control strategy, dynamic lane control strategy, and ramp entrance and exit control strategy are shown in Table 4.

[0086] surface Changes in indicators before and after the implementation of control measures

[0087]

[0088] It can be seen that this method can identify specific congestion points in the entrance and exit areas of highway ramps and implement corresponding collaborative management strategies for these congestion points, thereby alleviating road congestion and improving driving conditions.

[0089] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solutions based on the technical concept proposed in this invention shall fall within the scope of protection of this invention.

Claims

1. A method for tracing the source of highway congestion and controlling its management based on microscopic traffic simulation, characterized in that, Includes the following steps: S1. Obtain the real road network and adjust and correct it through automated means. Analyze the checkpoint data to identify the traffic flow distribution during peak hours and restore the traffic flow during peak hours. After data collection, import the traffic flow during peak hours into the simulation software. S2. Based on the characteristics of road segment sub-units and combined with the evaluation index system for road segmentation, determine the length of road segmentation; S3. Based on the segmentation length determined in step S2, the target road segment is divided into discrete road segments. Based on the travel time traffic index algorithm, the travel time ratio of each road segment is extracted to locate and trace the source of congested road segments. Specifically: Step S301: For the main line, calculate the total length of the main line, and then combine it with the segmentation length of the research road segment selected in step S2 to determine the number of main line segmentation sections. Step S302: Based on the three parameters of the total length of the main line, the segment length, and the number of segments of the main line, determine the latitude and longitude position of each segment in the simulated road network, and deploy them in each segment respectively. Detector, measures different sections Average travel speed of vehicles during peak hours ; Step S303: Calculate the traffic index for different sections based on the traffic index algorithm. Traffic congestion status based on travel speed : In the formula, It is a section of road In time The travel speed ratio; In free-flow state, road segment travel speed; In actual circumstances, the road section In time Average travel speed; Step S304: Based on traffic congestion status indicators Determine different sections The congestion index is used to determine the location of congested areas; S4. Based on the congested road sections identified in step S3, set up customized highway control measures that combine variable speed limits and ramp management.

2. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 1, characterized in that, Step S1 specifically includes the following steps: Step S101: Obtain the road network of the study area based on OpenStreetMap, and generate the simulation road network file .net.xml and the visualization configuration file .poly.xml; Step S102: Match the checkpoints to the simulated road network based on their latitude and longitude, determine the location of the checkpoints in the road network, count the traffic flow changes of each checkpoint at different times, obtain the peak hour distribution of the road segment, aggregate the traffic flow of each checkpoint within a time period, and connect the vehicles between the checkpoints according to the license plate number using the shortest path principle to obtain the peak hour traffic flow distribution, and convert it into a .rou.xml file for import into the simulation software.

3. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 1, characterized in that, Step S2 specifically includes the following steps: Step S201: In the simulation model, a flow detector is added at regular intervals along the study road segment. Each area at regular intervals along the study road segment is regarded as a sub-unit of the road segment. The simulation model is run to output the peak hourly flow data of each sub-unit of the simulated road network. Based on the output flow data, the flow changes of the sub-units of the road segment and whether there is a problem of excessive flow differences between the sub-units are analyzed to prepare for the implementation of step S202. Step S202: Enumerate the segment lengths within a range, and output the flow corresponding to each segment length using the flow detector set in step S201; check the flow distribution of the segmented road segments, and retain those that conform to the negative binomial distribution; compare the evaluation index under different segment lengths, and the selected final segment length should have small flow changes between different road segments and no sudden changes.

4. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 3, characterized in that, In step S201, a flow detector will be added every 100m along the studied road section.

5. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 3, characterized in that, In step S202, the segmentation lengths from 100m to 1000m are enumerated.

6. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 1, characterized in that, The characteristics of the road segment sub-units mentioned in step S2 include peak hour traffic flow, whether it is an interchange, toll station, ramp, or other non-main road, and information on road structure and road linearity; the evaluation index system includes indicators such as the degree of road segment feature display, segmentation accuracy, and post-processing workload.

7. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 1, characterized in that, In step S4, different control strategies are set according to the location of the congestion: (1) When the congestion is located on the main road, a variable speed limit control strategy and a dynamic lane control strategy are adopted; the variable speed limit control strategy is based on the traffic congestion section statistics obtained in step S3, and the optimal speed limit value of the road is selected through repeated simulation iterations; the dynamic lane control strategy is based on the traffic congestion section statistics obtained in step S3, and the optimal speed limit value of the road is selected through repeated simulation iterations; the dynamic lane control strategy is based on the traffic congestion section statistics obtained in step S3, and the optimal speed limit value of the road is selected through repeated simulation iterations at certain time intervals. Next, update the lane control strategy for the studied road section, when the section market share When the threshold is exceeded, the open lane control strategy will be selected. (2) When the congestion is located at the ramp, the ramp entrance and exit control strategy includes: setting up stepped solid and dashed lines on the lanes near the ramp entrance and exit, setting dashed lines in front of the entrance and exit to allow lane changes, and setting up lanes near the entrance and exit to prohibit vehicles from changing lanes at will, so that vehicles entering the ramp can change lanes in advance and alleviate the congestion near the ramp entrance.

8. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 7, characterized in that, The aforementioned dynamic lane control strategy will time interval Set to 10 minutes.

9. The method for highway congestion source analysis and control based on microscopic traffic simulation according to claim 7, characterized in that, In the aforementioned variable speed limit control strategy, the calculation formula and constraints for the variable speed limit value of the road segment are as follows: in: It is the difference between the real-time occupancy rate and the optimal occupancy rate. The magnitude of the difference in occupancy rate shows the vehicle distribution and operating status of the current road segment. It is the expected flow rate downstream of the controlled area; Variable speed limit coefficient; It is the maximum speed limit for the road.

Images