City traffic state monitoring and regional signal control system for cyber-physical system

Abstract

The application discloses a kind of city traffic state monitoring and regional signal control system for information physical fusion system, the system includes: traffic data multi-source fusion management subsystem, through data acquisition and processing, the physical entity of road, vehicle, traffic signal equipment in physical layer is carried out data mapping, data storage and transmission;Traffic state monitoring subsystem, through city road network modeling, road congestion degree monitoring, realize in information layer to road state real-time monitoring;Traffic congestion area identification subsystem, in information layer to the congestion area of road is identified and dynamically adjusted;Regional signal control subsystem, in information layer identification boundary control node, key node in area, generate signal light timing scheme, and feedback control is carried out to vehicle and traffic signal equipment in physical layer, and relieve road traffic congestion.The application can relieve regional congestion condition by the perception, analysis, decision and control to city road traffic.

Description

Technical Field

[0001] This invention relates to the field of traffic congestion identification and signal control technology, specifically to urban traffic condition monitoring and regional signal control systems for cyber-physical systems. Background Technology

[0002] As cities gradually expand, the number of vehicles and travel demand grow rapidly, and traffic congestion during peak hours becomes increasingly serious. Regional traffic congestion, spreading from specific points to larger areas, is hindering urban development and the improvement of people's living standards.

[0003] To address congestion control in road networks within smart city cyber-physical systems, traffic congestion is often identified through traffic state analysis and traffic indicator prediction, while traffic signal control is optimized using traffic performance index optimization. Seepage theory, a fundamental ideal model describing phase transitions and critical phenomena, typically simulates the gradual addition or removal of nodes or edges in a network. The statistical characteristics of the network during this process are used to understand its behavior and reveal its geometric and functional properties. Seepage theory provides an analytical method for studying the dynamic transition of traffic flow from free flow to congested flow at the network scale, and is therefore applied to urban traffic networks. Macroscopic Fundamental Graph (MFD) theory is a tool for macroscopic modeling of traffic flow in urban road networks. This theory demonstrates the relationships between macroscopic traffic flow parameters in the road network, such as vehicle density and traffic volume. These relationships can be modeled using MFD fitting curves, enabling the assessment of traffic flow dynamics in urban road networks and subsequent boundary control. Summary of the Invention

[0004] The purpose of this invention is to provide an urban traffic condition monitoring and regional signal control system for cyber-physical systems (CPS) to address the challenges of real-time monitoring of road networks within CPS frameworks and the difficulty in effectively alleviating traffic congestion by controlling a limited number of controlled nodes during feedback control. This invention, for CPS systems, constructs a multi-source traffic data fusion management subsystem, a traffic condition monitoring subsystem, a traffic congestion area identification subsystem, and a regional signal control subsystem. It combines seepage theory with monitoring area identification and congestion area identification to address the difficulty of real-time monitoring of road networks at the information layer. It utilizes a macroscopic basic graph and the weighted seepage centrality of the network to model and analyze congested areas, reducing the number of controlled nodes. Furthermore, it generates regional signal control schemes at the information layer based on road carrying capacity and controls physical layer entities through feedback control. This increases vehicle mobility within and outside the area, improving regional traffic flow and efficiency.

[0005] The specific technical solution for implementing this invention is as follows:

[0006] A city traffic condition monitoring and regional signal control system for a cyber-physical fusion system, the cyber-physical fusion system comprising a physical layer and an information layer, wherein the physical layer includes physical entities such as roads, vehicles, and traffic signal control equipment, as well as traffic congestion phenomena generated by the dynamic processes between these physical entities, characterized in that the information layer includes a traffic data multi-source fusion management subsystem for data mapping of physical entities in the physical layer, a traffic condition monitoring subsystem for monitoring road congestion levels, a traffic congestion area identification subsystem for dynamically identifying congested areas, and a regional signal control subsystem for generating control strategies and feeding back control to physical entities; wherein:

[0007] The traffic data multi-source fusion management subsystem is used to map and store the physical entities of roads, vehicles, and traffic signal equipment at the physical layer through data collection and processing; and to transmit the road topology, road attribute parameters, vehicle speed, vehicle quantity data, and intersection vehicle queue data extracted from urban road network data, vehicle positioning data, and intersection drone videos to the traffic status monitoring subsystem, traffic congestion area identification subsystem, and regional signal control subsystem.

[0008] The traffic condition monitoring subsystem uses road topology, road attribute parameters, vehicle speed, and vehicle quantity data obtained from the traffic data multi-source fusion management subsystem to model the urban road network, calculate traffic condition monitoring road segments, and monitor the congestion level of urban roads in real time.

[0009] The traffic congestion area identification subsystem, when the traffic condition monitoring subsystem detects that the congestion level of urban roads exceeds a set threshold, calculates the road congestion index and congestion connectivity areas in the road network, identifies real-time congestion areas in the road network and makes dynamic adjustments, and constructs an MFD model of the dynamic congestion area based on historical traffic data; thus realizing the detection of road congestion areas.

[0010] The regional signal control subsystem uses congested areas detected by the traffic congestion area identification subsystem to identify boundary control nodes and key control nodes within the area. Based on the intersection queuing data provided by the traffic data multi-source fusion management subsystem, and according to the real-time status of the number of vehicles, road capacity, and queue length in the area, it generates and controls the signal timing scheme for each controlled node. This enables feedback control of the physical entities of signal control equipment and vehicles, thereby alleviating road traffic congestion.

[0011] In the above scheme, the traffic condition monitoring subsystem specifically includes:

[0012] Urban road network modeling:

[0013] The city roads are constructed as a road network G(t) = (V, E, W) t Let V be the set of nodes representing road intersections, E be the set of edges representing road segments formed between intersections, and W be the edge weight. t This refers to the real-time road congestion index for each road segment in the road network; the congestion index is represented by road vehicle speed and number of vehicles. The congestion index for road segment e is:

[0014]

[0015] in, v represents the average vehicle speed on any road segment e at time t; e，max n represents the maximum permitted speed on any road segment e; e (t) represents the number of vehicles in road e at time t; n e，lane Let be the number of lanes on road e; let le be the length of road e. The average length occupied by the vehicle is set to 5.5 meters;

[0016] Calculation of monitored road sections:

[0017] The historical congestion indices of each road segment in the road network are weighted and summed to calculate the comprehensive congestion index of each road, thus constructing a high-frequency congestion network G. fre The overall congestion index of road segment e The calculation method is as follows:

[0018]

[0019] Among them, θ1, θ2, and θ3 are three road-level critical values ​​obtained by KMeans clustering analysis of historical congestion index values; The weights for the four levels of road congestion are assigned, with a focus on strengthening the influence of the severe congestion level.

[0020] For network G fre Seepage occurs, and the critical seepage value ρ c The seepage results at that time are for the subnet The seepage result corresponding to θ3 is subnet G. high To reduce monitoring costs, the edges in the top 3 connected segments with the most nodes in a network with fewer nodes are selected as the congestion level monitoring area M.

[0021] Road congestion monitoring: When the number of road segments in region M with a congestion index exceeding θ1 exceeds 50% of the roads in the monitoring area, the road network is considered to be congested as a whole; this enables real-time monitoring of the entire road network.

[0022] In the above scheme, the traffic congestion area identification subsystem specifically includes:

[0023] Road congestion area identification: Calculate the congestion connectivity areas and congestion boundaries of the road network, set the seepage threshold ρ∈[0,1], gradually increase the value of ρ starting from 0, and delete the congestion index C in the road network. e For the edge connecting (t) < ρ, the critical seepage value is ρ. c The percolation threshold is the value of the second largest connected component in the network when its size is maximized; the value ρ is close to the critical percolation value. c At that time, multiple connected areas formed in the road network are designated as congested areas B. c Congested area B c A node with non-congested neighbor nodes is used as the congestion boundary;

[0024] Dynamic area adjustment: When the congestion area detection interval is higher than 15 minutes, the congestion area is dynamically adjusted by calculating the congestion status of the adjacent edges of the congestion boundary;

[0025] MFD modeling: This involves creating a macroscopic basic graph model of the congested area, subtracting the number of completed trips from the cumulative number of vehicles within the area, and calculating the optimal cumulative number of vehicles N for the region. B .

[0026] In the above scheme, the regional signal control subsystem specifically includes:

[0027] Boundary control node identification:

[0028] By calculating node v∈V B The downstream section capacity is A v The number of vehicles is n v Outflow capacity is S v Calculate the carrying capacity R of the downstream road section v Controllability:

[0029] R v =A v -n v +S v (3)

[0030]

[0031] Where T is the time interval; For the downstream neighbor node v of the boundary node v nb The signal light cycle is expressed in seconds; P is the set of phases in the four-phase signal light timing that allow downstream traffic to pass. The green light duration corresponding to phase p; t loss The time it takes for the first vehicle to start and cross the stop line after the green light turns on is 2.3 seconds; t pass The average time for vehicles to pass the stop line is expressed in seconds per cubic meter (pcu), typically taken as 2.5 seconds for a platoon of small cars; λ is a reduction factor, taken as 0.9.

[0032] If R v >0.4A v If the downstream still has carrying capacity, the node is controllable; otherwise, the node is uncontrollable.

[0033] Key node identification within the region:

[0034] Node v in G fre The set of incoming edges on E is v，in The number of incoming edges to node v is |E v，in If |, then the seepage weight of that node is:

[0035]

[0036] Calculate the flow centrality of node v:

[0037]

[0038] Where |V| is the total number of nodes; s and r are the start and end points of the path, and v is the number of nodes on the path from s to r; σ sr σ is the total number of shortest paths from the starting point s to the ending point r. sr (v) is the number of paths through v; x represents the node seepage weight; the seepage centrality is sorted from largest to smallest, and the nodes with the highest ranking in the region are selected as the key control nodes in the region.

[0039] Signal timing light scheme generation:

[0040] For controlled nodes at the regional boundary, the minimum value among the upstream road segment flow change, downstream traffic flow carrying capacity change, and predicted completion volume is taken as the restricted input flow. This ensures that the flow does not exceed the road capacity and takes into account the regional completion capacity, thus avoiding excessive vehicle accumulation within the region. Based on the road's traffic diversion rate, the corresponding green light phase duration adjustment value is calculated, while keeping the traffic light cycle unchanged.

[0041] For key control nodes within the area, the green light phase duration for each phase is calculated based on the minimum value between the vehicle queue length of each road segment controlled by the intersection and the capacity of the downstream road segment of the node, while keeping the signal light cycle constant.

[0042] Signal control implementation: The calculated signal phase duration of the controlled signal is configured to the traffic lights at the location of the controlled node. By changing the signal timing information of the congested area, feedback control from the information layer to the physical layer is achieved, thereby improving the road traffic congestion.

[0043] The beneficial effects of this invention are:

[0044] This invention proposes an urban traffic condition monitoring and regional signal control system oriented towards a cyber-physical system. Within the cyber-physical system framework, physical entities such as roads and vehicles are mapped to the information layer for real-time monitoring and strategy analysis of road conditions. Signal configuration strategies are then fed back to traffic signal equipment in the physical layer, regulating vehicle flow within and outside congested areas to alleviate urban traffic congestion. A weighted seepage process is incorporated to reduce the monitoring range and lower the cost of monitoring the road network. Furthermore, the critical characteristics of the road network seepage process are utilized to calculate the road congestion connectivity area using a congestion index combining traffic speed and road occupancy. The invention combines boundary control based on a macroscopic basic graph with key node control within the region. By introducing the seepage centrality of complex networks, critical analysis of intersections within the region is performed to identify auxiliary control nodes and generate control strategies, effectively improving the region's traffic performance and efficiency. Attached Figure Description

[0045] Figure 1 This is a system framework diagram of the urban traffic condition monitoring and regional signal control system of the present invention, which is oriented towards cyber-physical fusion systems;

[0046] Figure 2 This is a graph showing the variation of the sizes of the largest and second largest connected segments of the road network with the seepage threshold in an embodiment of the present invention;

[0047] Figure 3 This is a location diagram of the three nodes with the greatest seepage centrality within the congested area in this embodiment of the invention;

[0048] Figure 4 This is a graph showing the change in the number of boundary-controlled nodes over time in an embodiment of the present invention. Detailed Implementation

[0049] To make the technical problems to be solved, the technical solutions, and the advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0050] This invention provides an urban traffic condition monitoring and regional signal control system for cyber-physical systems (CPS). Its purpose is to address the challenges of real-time monitoring of road networks within CPS frameworks, and the difficulty in effectively alleviating traffic congestion by controlling a limited number of controlled nodes during feedback control. For CPS systems, it constructs a multi-source traffic data fusion management subsystem, a traffic condition monitoring subsystem, a traffic congestion area identification subsystem, and a regional signal control subsystem. It combines seepage theory with monitoring area identification and congestion area identification to address the difficulty of real-time monitoring of road networks at the information layer. It utilizes a macroscopic basic graph and the weighted seepage centrality of the network to model and analyze congested areas, reducing the number of controlled nodes. Furthermore, it generates regional signal control schemes at the information layer based on road carrying capacity and controls physical layer entities through feedback control. This increases vehicle mobility and traffic efficiency within and outside the area, thereby improving regional control effectiveness.

[0051] Example

[0052] All results in this embodiment were obtained in an experimental environment using an 8-core 64-bit Windows 10 operating system, Python, and the SUMO simulation tool.

[0053] The simulation experiment road network G is set as a 20×20 grid network, where network nodes represent road intersections and edges represent road segments connecting intersections. The distance between any two neighboring nodes in the network is set to 300 meters, and the number of lanes is a random integer between {1, 2, 3}. The microscopic traffic simulation software SUMO is used to simulate the road traffic process of 30,000 vehicles with a travel demand within a 4-hour simulation period, where the starting point and ending point of vehicle travel are set on random road segments with an interval of more than 1,000 meters.

[0054] See Figure 1 , Figure 1 The system framework for this embodiment is as follows:

[0055] A city traffic condition monitoring and regional signal control system oriented towards cyber-physical fusion systems.

[0056] The cyber-physical fusion system comprises a physical layer and an information layer. The physical layer includes physical entities such as roads, vehicles, and traffic signal control equipment, as well as traffic congestion phenomena generated by the dynamic processes between these physical entities. The information layer includes a traffic data multi-source fusion management subsystem for mapping data to the physical entities in the physical layer, a traffic state monitoring subsystem for monitoring road congestion levels, a traffic congestion area identification subsystem for dynamically identifying congested areas, and a regional signal control subsystem for generating control strategies and feeding back control to the physical entities. Wherein:

[0057] The traffic data multi-source fusion management subsystem is used to map and store the physical entities of roads, vehicles, and traffic signal equipment at the physical layer through data collection and processing; and to transmit the road topology, road attribute parameters, vehicle speed, vehicle quantity data, and intersection vehicle queue data extracted from urban road network data, vehicle positioning data, and intersection drone videos to the traffic status monitoring subsystem, traffic congestion area identification subsystem, and regional signal control subsystem.

[0058] The traffic condition monitoring subsystem uses road topology, road attribute parameters, vehicle speed, and vehicle quantity data obtained from the traffic data multi-source fusion management subsystem to model the urban road network, calculate traffic condition monitoring road segments, and monitor the congestion level of urban roads in real time.

[0059] The traffic congestion area identification subsystem, when the traffic condition monitoring subsystem detects that the congestion level of urban roads exceeds a set threshold, calculates the road congestion index and congestion connectivity areas in the road network, identifies real-time congestion areas in the road network and makes dynamic adjustments, and constructs an MFD model of the dynamic congestion area based on historical traffic data; thus realizing the detection of road congestion areas.

[0060] The regional signal control subsystem uses congested areas detected by the traffic congestion area identification subsystem to identify boundary control nodes and key control nodes within the area. Based on the intersection queuing data provided by the traffic data multi-source fusion management subsystem, and according to the real-time status of the number of vehicles, road capacity, and queue length in the area, it generates and controls the signal timing scheme for each controlled node. This enables feedback control of the physical entities of signal control equipment and vehicles, thereby alleviating road traffic congestion.

[0061] Specifically, the traffic condition monitoring subsystem includes:

[0062] Urban road network modeling:

[0063] The city roads are constructed as a road network G(t) = (V, E, W) t Let V be the set of nodes representing road intersections, E be the set of edges representing road segments formed between intersections, and W be the edge weight. t This refers to the real-time road congestion index for each road segment in the road network; the congestion index is represented by road vehicle speed and number of vehicles. The congestion index for road segment e is:

[0064]

[0065] in, v represents the average vehicle speed on any road segment e at time t; e，max n represents the maximum permitted speed on any road segment e; e(t) represents the number of vehicles in road e at time t; n e，lane l is the number of lanes on road e; e Let e ​​be the length of road e; The average length occupied by the vehicle is set to 5.5 meters;

[0066] Calculation of monitored road sections:

[0067] The historical congestion indices of each road segment in the road network are weighted and summed to calculate the comprehensive congestion index of each road, thus constructing a high-frequency congestion network G. fre The overall congestion index of road segment e The calculation method is as follows:

[0068]

[0069] Among them, θ1, θ2, and θ3 are three road-level critical values ​​obtained by KMeans clustering analysis of historical congestion index values; The weights for the four levels of road congestion are assigned, with a focus on strengthening the influence of the severe congestion level.

[0070] For network G fre Seepage occurs, and the critical seepage value ρ c The seepage results at that time are for the subnet The seepage result corresponding to θ3 is subnet G. high To reduce monitoring costs, the edges in the top 3 connected segments with the most nodes in a network with fewer nodes are selected as the congestion level monitoring area M.

[0071] Road congestion monitoring: When the number of road segments in region M with a congestion index exceeding θ1 exceeds 50% of the roads in the monitoring area, the road network is considered to be congested as a whole; this enables real-time monitoring of the entire road network.

[0072] Specifically, the traffic congestion area identification subsystem includes:

[0073] Road congestion area identification: Calculate the congestion connectivity areas and congestion boundaries of the road network, set the seepage threshold ρ∈[0,1], gradually increase the value of ρ starting from 0, and delete the congestion index C in the road network. e For the edge connecting (t) < ρ, the critical seepage value is ρ. c The percolation threshold is the value of the second largest connected component in the network when its size is maximized; the value ρ is close to the critical percolation value. c At that time, multiple connected areas formed in the road network are designated as congested areas B. c Congested area B c A node with non-congested neighbor nodes is used as the congestion boundary;

[0074] Dynamic area adjustment: When the congestion area detection interval is higher than 15 minutes, the congestion area is dynamically adjusted by calculating the congestion status of the adjacent edges of the congestion boundary;

[0075] MFD modeling: This involves creating a macroscopic basic graph model of the congested area, subtracting the number of completed trips from the cumulative number of vehicles within the area, and calculating the optimal cumulative number of vehicles N for the region. B .

[0076] Specifically, the regional signal control subsystem includes:

[0077] Boundary control node identification:

[0078] By calculating node v∈V B The downstream section capacity is A v The number of vehicles is n v Outflow capacity is S v Calculate the carrying capacity R of the downstream road section v Controllability:

[0079] R v =A v -n v +S v (3)

[0080]

[0081] Where T is the time interval; For the downstream neighbor node v of the boundary node v nb The signal light cycle is expressed in seconds; P is the set of phases in the four-phase signal light timing that allow downstream traffic to pass. The green light duration corresponding to phase p; t loss The time it takes for the first vehicle to start and cross the stop line after the green light turns on is 2.3 seconds; t pass The average time for vehicles to pass the stop line is expressed in seconds per cubic meter (pcu), typically taken as 2.5 seconds for a platoon of small cars; λ is a reduction factor, taken as 0.9.

[0082] If R v >0.4A v If the downstream still has carrying capacity, the node is controllable; otherwise, the node is uncontrollable.

[0083] Key node identification within the region:

[0084] Node v in G fre The set of incoming edges on E is v，in The number of incoming edges to node v is |E v，in If |, then the seepage weight of that node is:

[0085]

[0086] Calculate the flow centrality of node v:

[0087]

[0088] Where |V| is the total number of nodes; s and r are the start and end points of the path, and v is the number of nodes on the path from s to r; σ sr σ is the total number of shortest paths from the starting point s to the ending point r. sr (v) is the number of paths through v; x represents the node seepage weight; the seepage centrality is sorted from largest to smallest, and the nodes with the highest ranking in the region are selected as the key control nodes in the region.

[0089] Signal timing light scheme generation:

[0090] For controlled nodes at the regional boundary, the minimum value among the upstream road segment flow change, downstream traffic flow carrying capacity change, and predicted completion volume is taken as the restricted input flow. This ensures that the flow does not exceed the road capacity and takes into account the regional completion capacity, thus avoiding excessive vehicle accumulation within the region. Based on the road's traffic diversion rate, the corresponding green light phase duration adjustment value is calculated, while keeping the traffic light cycle unchanged.

[0091] For key control nodes within the area, the green light phase duration for each phase is calculated based on the minimum value between the vehicle queue length of each road segment controlled by the intersection and the capacity of the downstream road segment of the node, while keeping the signal light cycle constant.

[0092] Signal control implementation: The calculated signal phase duration of the controlled signal is configured to the traffic lights at the location of the controlled node. By changing the signal timing information of the congested area, feedback control from the information layer to the physical layer is achieved, thereby improving the road traffic congestion.

[0093] See Figure 2 , Figure 2 This is a graph showing the variation of the sizes of the largest and second largest connected segments of the road network with the seepage threshold in this embodiment. When the size of the second largest connected segment |SC| reaches its maximum value, the resulting seepage threshold is ρ. c =0.66, in The largest connected component corresponding to the time is the largest connected region B that is most severely affected. c .

[0094] See Figure 3 , Figure 3This is a location map of the three nodes with the highest seepage centrality within the congested area in this embodiment. Based on the historical traffic flow weighting of the road segments in the experimental area of ​​this embodiment, and the analysis of the top three nodes with the highest seepage centrality, the darker the edge color, the more congested the area. The intersection corresponding to the node with the highest seepage centrality is used as the signal control target within the area.

[0095] See Figure 4 , Figure 4 This is a graph showing the change in the number of controlled boundary nodes over time in this embodiment. Starting from the 8th time slice, the cumulative number of vehicles monitored in the area reaches the optimal cumulative threshold of 1500 vehicles, after which boundary signal control is activated. The number of controlled boundary nodes fluctuates over time, and the fluctuation trend is similar to the traffic flow trend in the inflow area. That is, when the inflow is low, it indicates that the congestion level in the inflow area is high, so the number of nodes that cannot be effectively controlled in the boundary node group increases, and the number of controlled nodes decreases.

Claims

1. A city traffic condition monitoring and regional signal control system for a cyber-physical fusion system, the cyber-physical fusion system comprising a physical layer and an information layer, the physical layer comprising physical entities such as roads, vehicles, and traffic signal control equipment, as well as traffic congestion phenomena generated by the dynamic processes between physical entities, characterized in that, The information layer includes a traffic data multi-source fusion management subsystem for mapping data to physical entities in the physical layer, a traffic state monitoring subsystem for monitoring road congestion levels, a traffic congestion area identification subsystem for dynamically identifying congested areas, and a regional signal control subsystem for generating control strategies and feeding back control to physical entities; wherein: The traffic data multi-source fusion management subsystem is used to map and store the physical entities of roads, vehicles, and traffic signal equipment at the physical layer through data collection and processing; and to transmit the road topology, road attribute parameters, vehicle speed, vehicle quantity data, and intersection vehicle queue data extracted from urban road network data, vehicle positioning data, and intersection drone videos to the traffic status monitoring subsystem, traffic congestion area identification subsystem, and regional signal control subsystem. The traffic condition monitoring subsystem uses road topology, road attribute parameters, vehicle speed, and vehicle quantity data obtained from the traffic data multi-source fusion management subsystem to model the urban road network, calculate traffic condition monitoring road segments, and monitor the congestion level of urban roads in real time. The traffic congestion area identification subsystem, when the traffic condition monitoring subsystem detects that the congestion level of urban roads exceeds a set threshold, calculates the road congestion index and congestion connectivity areas in the road network, identifies real-time congestion areas in the road network and makes dynamic adjustments, and constructs an MFD model of the dynamic congestion area based on historical traffic data; thus realizing the detection of road congestion areas. The regional signal control subsystem uses congested areas detected by the traffic congestion area identification subsystem to identify boundary control nodes and key control nodes within the area. Based on the intersection queuing data provided by the traffic data multi-source fusion management subsystem, and according to the real-time status of the number of vehicles, road capacity, and queue length in the area, it generates and controls the signal timing scheme for each controlled node. This enables feedback control of the physical entities of signal control equipment and vehicles, thereby alleviating road traffic congestion. Key node identification within the region: node On the Internet The set of incoming edges on is ,node The number of incoming edges is Then the seepage weight of this node is: (5) compute nodes Flow centerity: (6) in, is the total number of nodes; s and r are the start and end points of the path, and v is the number of nodes on the path from s to r; It is the total number of shortest paths from the starting point s to the ending point r. Through The number of paths; Indicates the node percolation weight; The overall congestion index of road segment e; sorting the seepage centrality from largest to smallest, and selecting the top-ranked nodes in the area as key control nodes in the area; The historical congestion indices of each road segment in the road network are weighted and summed to calculate the comprehensive congestion index of each road, thus constructing a high-frequency congestion network. .

2. The urban traffic condition monitoring and regional signal control system for cyber-physical fusion systems according to claim 1, characterized in that, The traffic condition monitoring subsystem specifically includes: Urban road network modeling: Constructing urban roads into a road network Node set Represents a road intersection, edge set Represents the road segment formed between intersections, edge weight This refers to the real-time road congestion index for each road segment within the road network; the congestion index is represented by road vehicle speed and the number of vehicles. The congestion index is: (1) in, Represents any road segment The average vehicle speed at time t; Represents any road segment Maximum permitted speed; For the road at time t The number of vehicles in the area; For roads The number of lanes; For roads Length; The average length occupied by the vehicle is set to 5.5 meters; Calculation of monitored road sections: Section Comprehensive congestion index The calculation method is as follows: (2) in, , , These are three road-level critical values ​​obtained from KMeans cluster analysis of historical congestion index values; The weights for the four levels of road congestion are assigned, with a focus on strengthening the influence of the severe congestion level. For the network Seepage occurs; critical seepage value The seepage results at that time are for the subnet The corresponding seepage result is the subnet. To reduce monitoring costs, the edges between the top three connected components with the most nodes in a network with fewer nodes are selected as the congestion level monitoring area. ; Road congestion monitoring: When The region's congestion index exceeds When the number of road segments exceeds 50% of the roads in the monitoring area, the road network is considered to be congested as a whole; this enables real-time monitoring of the entire road network.

3. The urban traffic condition monitoring and regional signal control system for cyber-physical fusion systems according to claim 1, characterized in that, The traffic congestion area identification subsystem specifically includes: Road congestion area identification: Calculate the congested connectivity areas and congestion boundaries of the road network, and set a seepage threshold. [0,1], starting from 0 and gradually increasing The value of the congestion index is determined and removed from the road network. The connection edge, the critical value of seepage The percolation threshold is the value at which the size of the second largest connected component in the network is maximized; it will approach the critical percolation value. At that time, multiple connected areas formed in the road network are considered as the identified congestion areas. Congested areas A node with non-congested neighbor nodes is used as the congestion boundary; Dynamic area adjustment: When the congestion area detection interval is higher than 15 minutes, the congestion area is dynamically adjusted by calculating the congestion status of the adjacent edges of the congestion boundary; MFD modeling: This involves creating a macroscopic basic graph model of the congested area, subtracting the number of trips completed within the area, to calculate the optimal cumulative number of vehicles in the region. .

4. The urban traffic condition monitoring and regional signal control system for cyber-physical fusion systems according to claim 1, characterized in that, The regional signal control subsystem specifically includes: Boundary control node identification: Through compute nodes The downstream section capacity is The number of vehicles is Outflow capacity is Calculate the carrying capacity of the downstream road section Controllability: (3) (4) Where T is the time interval; Boundary nodes downstream neighbor nodes The traffic light cycle is measured in seconds. The set of phases in a four-phase traffic light timing system that allow downstream traffic to pass; The green light duration corresponds to phase p; The time it takes for the first vehicle to start and cross the stop line after the green light turns on is 2.3 seconds. The average time for a vehicle to cross the stop line is expressed in seconds per unit unit (pcu). For a platoon of small cars, this is typically taken as 2.5 seconds. The reduction factor is set to 0.9; like If the downstream still has carrying capacity, the node is controllable; otherwise, the node is uncontrollable. Signal timing light scheme generation: For controlled nodes at the regional boundary, the minimum value among the upstream road segment flow change, downstream traffic flow carrying capacity change, and predicted completion volume is taken as the restricted input flow. This ensures that the flow does not exceed the road capacity and takes into account the regional completion capacity, thus avoiding excessive vehicle accumulation within the region. Based on the road's traffic diversion rate, the corresponding green light phase duration adjustment value is calculated, while keeping the traffic light cycle unchanged. For key control nodes within the area, the green light phase duration for each phase is calculated based on the minimum value between the vehicle queue length of each road segment controlled by the intersection and the capacity of the downstream road segment of the node, while keeping the signal light cycle constant. Signal control implementation: The calculated signal phase duration of the controlled signal is configured to the traffic lights at the location of the controlled node. By changing the signal timing information of the congested area, feedback control from the information layer to the physical layer is achieved, thereby improving the road traffic congestion.

Images