A method for dividing urban mesoscopic signal control units based on breadth search

Abstract

This application relates to the fields of urban traffic management and signal control, and discloses a method for dividing urban mesoscale signal control units based on breadth-based search. The method comprises the following steps: collecting traffic data from the road network, calculating the optimal signal timing scheme, and fine-tuning the signal cycle based on downstream spillover risk; screening effective coordinated turns by calculating the coordination benefits of each turn and generating signal coordination trunk lines; utilizing a breadth-based search (BFS) algorithm to generate signal control units, which are rationally divided into single-point signal control units, trunk signal control units, and regional signal control units; and finally, calculating coordination directions and phase differences to ensure signal coordination and green wave passage, thereby optimizing the signal control scheme for the road network. This method can improve the capacity of urban traffic, reduce traffic congestion, and enhance overall traffic flow.

Description

Technical Field

[0001] The present invention relates to the technical field of urban traffic management and signal control, and in particular to a method for dividing urban mesoscopic signal control units based on breadth search. Background Art

[0002] Existing urban traffic signal control systems typically rely on fixed signal timing schemes, lacking real-time response and optimization to changes in traffic flow and road network topology. Most traditional signal control methods are unable to dynamically adjust signal cycles based on changes in traffic demand, leading to congestion during peak hours or under special circumstances, reducing road network efficiency.

[0003] Existing signal control systems often rely on a single intersection signal timing scheme, neglecting coordination between intersections and overall traffic flow control on arterial routes. This is particularly true when multiple intersections are involved, as the lack of an effective green wave coordination strategy often results in poor traffic flow and impacts overall traffic efficiency. Signal control system configurations often fail to account for the synergies between individual intersections, leading to frequent conflicts between signal units, significant wasted travel time, and further exacerbating traffic congestion.

[0004] Furthermore, existing signal control systems are relatively simplistic, often focusing on single intersections. This approach lacks consideration for overall signal control capabilities across arterials and regions. When multiple intersecting arterials exist within the same area, existing signal control systems often fail to effectively coordinate signal timing across these arterials, leading to signal interference and mutual impact between these arterials, hindering the full potential of traffic flow.

[0005] Therefore, existing technologies lack an effective signal coordination and signal control unit division method, making it difficult to flexibly respond to traffic flow changes in complex road network structures, further limiting the capacity and smoothness of urban transportation networks. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention provides a method for dividing urban mesoscopic signal control units based on breadth search, which solves the problem that the existing traffic signal control system cannot effectively coordinate multiple intersection signals and optimize the overall traffic capacity of the road network.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for dividing urban mesoscopic signal control units based on breadth search, comprising the following steps:

[0008] Obtain multiple intersections and their traffic information in the road network;

[0009] Calculating an optimal signal timing plan for each intersection based on the traffic information;

[0010] Calculate the queue length of each intersection's downstream section and adjust the release time of the intersection based on the remaining downstream capacity;

[0011] Calculate the coordination benefit of each turn and select turns with benefits greater than zero as coordination candidates;

[0012] Expand the coordination path based on the breadth search algorithm to form a signal coordination trunk line;

[0013] Divide the signal control unit according to the trunk line topology relationship;

[0014] The signal phase difference is calculated based on the main coordination direction of each trunk line to optimize the signal timing.

[0015] Preferably, the step of obtaining multiple intersections in the road network and their traffic information includes:

[0016] Obtain intersection traffic demand, intersection upstream and downstream topology, intersection signal control configuration, and historical record data;

[0017] Data fusion is performed with turns as the minimum granularity to generate fields including intersection number, turn number, number of lanes, lane length, signal control stage number, actual intersection cycle, turn green-signal ratio, actual turn flow rate and demand flow rate, upstream and downstream turns, and corresponding traffic flow ratios.

[0018] Preferably, the step of calculating the optimal signal timing scheme for each intersection based on the traffic information includes:

[0019] Calculate the optimal signal cycle based on the number of intersection stages, stage switching loss time and saturation:

[0020]

[0021] Among them, C(I) is the optimal signal period, N ph is the number of intersection stages, T loss is the single-stage switching loss time, X(I) is the total saturation of the intersection;

[0022] For each signal control stage, calculate its saturation:

[0023]

[0024] Among them, X ph is the stage saturation, DR(T rep ) is the demand flow rate of the representative lane in the stage, FR(T rep ) is the capacity of the representative lane at the stage, that is, the upper limit of the traffic volume that the lane can pass under the current signal timing scheme, that is, the lane saturation flow rate multiplied by the lane green signal ratio;

[0025] Based on the stage saturation and the total saturation of the intersection, the green-signal ratio of each stage is allocated according to the principle of balance:

[0026]

[0027] Among them, G(T) is the green-to-trust ratio of the stage;

[0028] Calculate the green light time in each stage based on the green-to-signal ratio and intersection cycle:

[0029] T ph =G(T)×(C(I)-N ph ×T loss )

[0030] Among them, T ph Green light time for the stage.

[0031] Preferably, the step of calculating the length of the detention queue at the downstream section of each intersection and adjusting the release time of the intersection based on the downstream remaining traffic capacity includes:

[0032] Calculate the queue length in the downstream lane based on the flow difference and signal cycle of the downstream turn:

[0033]

[0034] Among them, L queue (D) is the queue length of the downstream lane, DR(D) is the demand flow rate of the downstream turn, FR(D) is the actual flow rate of the downstream turn, C(D) is the signal cycle of the downstream intersection, L std The standard distance between queuing vehicles;

[0035] Calculate the capacity of the downstream free lane based on the total length of the downstream road section, the length of the delayed queue, and the safety distance:

[0036] L free (D)=L c (D)-L queue (D)-S safe

[0037] Among them, L free (D) is the capacity of the downstream free lane, L c (D) is the lane length of the downstream section, S safe is the safe distance to the exit; if the following conditions are met, it is determined that there is an overflow risk downstream:

[0038] L free (D)<0

[0039] or

[0040] L queue (D)≥L c(D)-S safe

[0041] Based on the length of the downstream free lane, calculate the downstream maximum green light capacity, that is, the maximum green light time allowed for release at this intersection:

[0042]

[0043] Among them, G max (D) is the maximum green light capacity downstream; T std is the intersection saturation flow rate headway;

[0044] For downstream turns with overflow risks, adjust the green light time at the corresponding stage of the intersection:

[0045] T p ′ h =min(G max (D),T ph -S)+S

[0046] Among them, T p ′ h is the green light time of the corresponding stage at this intersection, T ph is the green light time of the stage, and S is the reserved buffer time;

[0047] Recalculate the total intersection cycle based on the adjusted green light time:

[0048] C ′ (I) = ∑T p ′ h

[0049] Among them, C ′ (I) Adjusted intersection signal cycle.

[0050] Preferably, the step of calculating the coordination benefit of each turn and screening the turns with benefits greater than zero as coordination candidates includes:

[0051] For each turn, if its required flow rate is greater than the actual flow rate, the turn is considered to be in an overflow state and does not participate in green wave coordination. Overflow turns are eliminated from all turns, and the remaining non-overflowing turns form the coordination candidate set.

[0052] For each candidate turn in the coordination candidate set, calculate its coordination benefit:

[0053]

[0054] Where FR(T) is the turning capacity in the optimal single intersection solution, W coor(T) is the proportion of coordinated traffic flow in turning, FR(I) is the total traffic capacity of the intersection in the optimal single intersection solution, N ph is the total number of signal stages at the intersection;

[0055] The turns with coordination benefits greater than zero are retained and enter the final coordination candidate set.

[0056] Preferably, the step of expanding the coordination path based on the breadth search algorithm to form a signal coordination trunk line includes: step 1: initializing the new trunk line, including:

[0057] (1) In the candidate turn set TurnSet valid In the process, choose the turn with the largest coordination benefit max As the starting point of the trunk line, add the turn set to be expanded TurnSet expand ;

[0058] (2) Assign a unique identifier Corr to the current trunk id ;

[0059] (3) Set the intersection relation field INTER of the trunk line to null, indicating that the current trunk line does not intersect with other trunk lines;

[0060] Step 2: Trunk expansion

[0061] Step 2.1, from TurnSet expand Select any one of the turns current ;

[0062] Step 2.2, Turn current Perform coordination access verification. If the following conditions are met, the Turn is discarded. current And return to step 2.1:

[0063] Current Main Line Corr id It has intersected with other main lines;

[0064] and Turn current Intersection with other main lines;

[0065] step2.3, Turn current Join the current trunk line and mark it as the trunk line coordinated turn coor ;

[0066] Step 2.4, if Turn current Corr with other trunk lines i ′ d If they intersect, update the current trunk intersection relationship attributes:

[0067] INTER=Corr i ′ d

[0068] step2.5, turn current Upstream Turn up and downstream turn down To expand:

[0069] Turn up and Turn down Add to the set to be expanded TurnSet expand ;

[0070] Step 2.6: Turn Set from Candidate valid Remove Turn current All turns in the current stage;

[0071] step2.7, repeat steps 2.1 to 2.7 until TurnSet expand is empty, completing the current trunk generation; step 3: the generated trunk Corr id Stored in the trunk set, if the candidate turn set TurnSet valid If it is not empty, return to step 1 and generate the next trunk line; otherwise, terminate the process.

[0072] Preferably, the step of dividing the signal control units according to the trunk line topology relationship includes:

[0073] The signal control unit is divided according to the trunk topology structure, and the signal control unit includes:

[0074] Single-point signal control unit: an independent intersection, not part of any trunk line;

[0075] Trunk signal control unit: a single trunk line without crossing;

[0076] Regional signal control unit: a control area consisting of multiple cross-links;

[0077] Merge trunk lines based on the intersection situation to form regional signal control units.

[0078] Preferably, the step of calculating the signal phase difference based on the main coordination direction of each trunk line includes:

[0079] For the intersections within the signal control unit, the coordinated turn at the trunk intersection is selected as the regional coordinated turn;

[0080] The signal control phase where the coordinated turn occurs is defined as the coordination phase of the intersection;

[0081] Based on the travel time and queue length of the coordinated turn, the phase difference is calculated:

[0082]

[0083] Among them, L turn is the length of the turning section, V is the average speed of the section, T ph Green light time for the stage, W coor (T) is the proportion of coordinated traffic that is turned.

[0084] Preferably, for the intersections within the signal control unit, the step of selecting the coordinated turning at the trunk intersection as the regional coordinated turning includes:

[0085] If the intersection belongs to two or more intersecting trunk roads, the main coordinated turn is determined by the priority of the trunk road generation order.

[0086] The present invention also provides a device for dividing urban mesoscopic signal control units based on breadth search, comprising:

[0087] Data acquisition module, used to obtain intersection traffic information and calculate relevant flow data;

[0088] Single intersection signal optimization module, used to calculate the optimal signal timing plan for each intersection;

[0089] Downstream overflow control module, used to adjust the release time of the intersection based on the downstream traffic capacity;

[0090] Coordination benefit calculation module, used to calculate the coordination benefit of each turn and screen candidate turns;

[0091] A trunk generation module, configured to expand the coordination path using breadth search to form a signal coordination trunk;

[0092] A signal control unit division module is used to divide the signal control unit;

[0093] The coordination direction and phase difference calculation module is used to calculate the coordination direction and signal phase difference to achieve green wave coordinated control.

[0094] The present invention provides a method for dividing urban mesoscopic signal control units based on breadth search, which has the following beneficial effects:

[0095] 1. This invention optimizes urban traffic signal control, achieves coordinated intersection signaling and green wave traffic flow, and reduces traffic congestion. By rationally allocating signal cycles and green light times, it can effectively avoid traffic bottlenecks and improve the traffic capacity of the entire urban road network.

[0096] 2. The present invention effectively divides the urban road network into signal control units through an algorithm based on breadth search. It can scientifically divide intersections into single-point signal control units, trunk signal control units and regional signal control units according to the traffic demand, topology and intersection relationship of different intersections and trunk lines, thereby accurately controlling and optimizing signal timing.

[0097] 3. This invention effectively avoids overflow risks on downstream roads by calculating the length of the downstream queue and adjusting signal timing. By adjusting the signal cycle and green light duration, it ensures smooth traffic flow on downstream roads, reduces traffic congestion and queues, and ensures smooth traffic flow.

[0098] 4. This invention calculates the coordination benefit of each turn, ensuring that only beneficial coordinated turns are selected for green wave coordination. By optimizing trunk line generation and expansion through the breadth-based search algorithm (BFS), signal coordination trunk lines can effectively cover multiple intersections and trunk lines, improving the coordination of the entire traffic network and optimizing signal control solutions.

[0099] 5. This invention can handle complex intersections and road network topologies, rationally planning the relationship between trunk lines and regional signal control units. By accurately recording and merging the intersection relationships of trunk lines, it achieves signal coordination for multiple intersecting trunk lines, thus avoiding deadlock and ensuring efficient coordination within the region. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 Schematic diagram of the method flow of the present invention;

[0101] Figure 2 Schematic diagram of the method flow of the present invention;

[0102] Figure 3 Schematic diagram of the device structure of the present invention.

[0103] Among them, 10, data acquisition module; 20, single intersection signal optimization module; 30, downstream overflow control module; 40, coordination benefit calculation module; 50, trunk generation module; 60, signal control unit division module; 70, coordination direction and phase difference calculation module. DETAILED DESCRIPTION

[0104] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0105] Please see the attached Figure 1 -Attached Figure 2The present invention provides a method for dividing urban mesoscopic signal control units based on breadth search, which aims to optimize the urban traffic signal control system, improve traffic flow efficiency, reduce the number of vehicle stops and waits, and enhance the coordination ability of arterial lines.

[0106] The present invention adopts a breadth-first search (BFS) algorithm to ensure the global optimality of signal control unit division, and fully considers the coordination relationship of intersection signals, the impact of detention queues, the coordination requirements of main lines and the phase optimization strategy to achieve efficient green wave coordinated control.

[0107] like Figure 1 As shown, the method for dividing urban mesoscopic signal control units based on breadth search may include the following steps:

[0108] S1. Obtain multiple intersections and their traffic information in the road network;

[0109] S2. Calculate the optimal signal timing plan for each intersection based on traffic information;

[0110] S3. Adjust the release time of this intersection based on the remaining traffic capacity downstream;

[0111] S4. Calculate the coordination benefit of each turn and select turns with benefits greater than zero as coordination candidates;

[0112] S5. Expand the coordination path based on the breadth search algorithm to form a signal coordination trunk line;

[0113] S6. Divide the signal control unit according to the trunk line topology relationship;

[0114] S7. Calculate the signal phase difference based on the main coordination direction of each trunk line to optimize the signal timing.

[0115] The following is a detailed description of each step in the method of the present invention, which comprehensively explains the specific implementation principles, technical details and processes of each step.

[0116] Step S1 collects and integrates traffic data from the road network, providing the foundation for subsequent signal control unit division and signal optimization. Through data association and calculation, a traffic feature dataset with turn-based granularity is generated, containing topological information, traffic demand, signal configuration information, and traffic flow status.

[0117] First, data on intersection traffic demand, upstream and downstream topology, intersection signal control configuration, and historical records are collected. Data sources include fixed detection equipment (such as geomagnetic detectors and video detectors), floating vehicle data (FCD), manual survey data, and signal control system logs. After preprocessing, the data is fused at the turn-by-turn granularity to form the following key fields:

[0118] Intersection ID:

[0119] Each intersection is assigned a unique number that identifies the smallest control unit of the signal control unit. This number is used to define the intersection to which the turn belongs and establish the upstream and downstream topological relationship.

[0120] Movement ID:

[0121] Each turn is assigned a unique number, serving as the smallest perceptual unit for traffic state data fusion. The number is composed of "intersection number + entry lane group + target lane group" to ensure uniqueness.

[0122] Lane Count of Movement:

[0123] Calculate the number of lanes included in the turn. For mixed lanes, convert according to the usage ratio of each type of vehicle:

[0124]

[0125] in:

[0126] N lanes (M) is the number of valid lanes included in the turn;

[0127] W i is the mixed traffic ratio of the i-th lane (e.g., for a left-turn-through mixed lane, if the left-turn ratio is 40%, then W = 0.4);

[0128] N i is the total number of lanes of type i.

[0129] Lane Length of Movement:

[0130] Record the lane length involved in the turn and compare it with the queue length to assess the overflow risk. The conditions for queue overflow are as follows:

[0131] When the queue length for the turn is greater than the lane length for the turn, queue overflow may occur, affecting the signal optimization strategy.

[0132] Phase ID of Movement:

[0133] Each turn corresponds to a major release phase in the signal control scheme. If a turn involves multiple phases, the major release phase is selected.

[0134] Actual Cycle Length of the intersection:

[0135] During the detection period, the signal cycle (in seconds) of the intersection is defined as the time it takes for the traffic light to execute once in the phase chain sequence.

[0136] This data can be used to calculate the steering green-to-signal ratio, phase difference adjustment, etc.

[0137] Turning to the actual Green Split of Movement:

[0138] The green light time ratio of the turn in actual signal control:

[0139]

[0140] in:

[0141] T green (M) is the actual green light time for the turn (in seconds);

[0142] C(I) is the intersection cycle (seconds).

[0143] Actual Flow Rate per Lane:

[0144] The actual flow rate of this diversion during the statistical period:

[0145]

[0146] in:

[0147] V pass (M) is the number of vehicles passing through the green light time; T std is the headway time of the intersection saturation flow rate;

[0148] T green (M) is the green light time (seconds).

[0149] Average demand flow rate per lane for turning traffic:

[0150] Calculate the demand flow rate when there is a queue delay at this turn:

[0151]

[0152] in:

[0153] Q init (M) is the number of vehicles initially queued at the green light;

[0154] C(I) is the intersection cycle (seconds).

[0155] When q demand (M)>qactual (M) indicates that the turn capacity is limited and the signal timing or coordination strategy needs to be adjusted.

[0156] Upstream Intersection of Movement:

[0157] According to the topological relationship of the signal-controlled intersection, the upstream intersection of the turn is determined. Generally, each turn corresponds to only one upstream intersection.

[0158] Upstream Movement of Movement:

[0159] At the upstream intersection, determine the main source of the turn. If there are multiple possible sources at the upstream intersection, select the turn with the largest traffic contribution:

[0160] M up (M) = arg max M ′ W coor (M ′ )

[0161] in:

[0162] M up (M) is the upstream main coordinated turn of the turn;

[0163] W coor (M ′ ) is the flow contribution ratio of upstream diversion.

[0164] Coordinated Flow Ratio of Movement:

[0165] Calculate the proportion of flow in this turn that comes from the upstream turn:

[0166]

[0167] in:

[0168] Q coor (M) is the flow from upstream coordinated diversion;

[0169] Q total (M) is the total flow of the steering.

[0170] This ratio is used in subsequent signal coordination calculations.

[0171] Downstream Intersection of Movement:

[0172] According to the topological relationship of the signal-controlled intersection, the downstream intersection of the turn is determined. Generally, there is only one downstream intersection.

[0173] Downstream Movement of Movement:

[0174] At the downstream intersection, determine the main destination of the turn. If there are multiple possible downstream turns, choose the one with the largest flow contribution:

[0175] M down (M) = arg max M′ W coor (M ′ )

[0176] in:

[0177] M down (M) is the downstream main coordinated steering of this steering.

[0178] This step completes the data collection, fusion and calculation, providing the necessary input data for subsequent signal control unit division, trunk generation and signal optimization. The logical relationship between each field is as follows:

[0179] The intersection number and the turn number define the spatial topological relationship;

[0180] Traffic flow data is used to assess capacity and coordination needs;

[0181] Upstream and downstream steering correlation is used to calculate coordination ratios and support signal optimization;

[0182] Green-to-signal ratio, flow rate and demand flow rate are used to determine overflow risk and optimize phase allocation.

[0183] After the data structure is reasonably organized, it will be passed as input to the subsequent steps for signal control unit division and signal optimization.

[0184] Regarding step S2, this step calculates the optimal signal cycle and green light time in each stage for each intersection based on the traffic demand of the intersection, signal control stage information and road capacity to optimize traffic efficiency.

[0185] First, the optimal signal cycle is calculated based on the number of intersection stages, stage switching loss time, and total saturation. The calculation formula for the intersection cycle is as follows:

[0186]

[0187] in:

[0188] C(I) is the optimal signal period (seconds);

[0189] Nph is the number of intersection stages;

[0190] T loss is the single-stage switching loss time (seconds);

[0191] X(I) is the total saturation of the intersection (dimensionless value).

[0192] The intersection saturation is calculated by summing the saturations at each stage:

[0193]

[0194] in:

[0195] X ph is the saturation of a single stage.

[0196] For each signal control stage, its saturation is calculated and defined as follows:

[0197]

[0198] in:

[0199] DR(T rep ) is the demand flow rate of the representative lane in the stage (vehicles / hour);

[0200] FR(T rep ) is the saturation flow rate of the representative lane in the stage (vehicles / hour).

[0201] According to the principle of balance, the green credit ratio of each stage is calculated:

[0202]

[0203] in:

[0204] G(T) is the green-to-signal ratio of the stage (dimensionless value).

[0205] Finally, the green light time of each stage is calculated based on the green-to-signal ratio and signal cycle:

[0206] T ph =G(T)×(C(I)-N ph ×T loss )

[0207] in:

[0208] T ph Green light time for the stage (seconds).

[0209] The calculation logic in the above steps ensures that the signal cycle matches traffic demand, improving intersection capacity and reducing delays. The signal timing calculation results will be used for subsequent signal control unit division and signal coordination optimization.

[0210] As for step S3, this step aims to determine whether there is an overflow risk by evaluating the length of the queue on the downstream road section, and adjust the signal timing of this intersection based on this to optimize traffic efficiency and alleviate congestion on the downstream road section.

[0211] First, calculate the length of the downstream lane's delayed queue. The calculation formula for the delayed queue length is as follows:

[0212]

[0213] in:

[0214] L queue (D) is the length of the queue in the downstream lane (meters);

[0215] DR(D) is the demand flow rate of downstream steering (vehicles / hour);

[0216] FR(D) is the actual flow rate of the downstream diversion (vehicles / hour);

[0217] C(D) is the signal cycle (seconds) of the downstream intersection;

[0218] L std is the standard queuing vehicle distance (meters).

[0219] Next, calculate the remaining lane capacity of the downstream road segment. The calculation formula for the remaining lane capacity is as follows:

[0220] L free (D)=L c (D)-L queue (D)-S safe

[0221] in:

[0222] L free (D) is the downstream free lane capacity (m);

[0223] L c (D) is the total lane length of the downstream section (m);

[0224] L queue (D) is the length of the queue in the downstream lane (meters);

[0225] S safe is the exit safety distance (meters).

[0226] If the following conditions are met, it is determined that there is a risk of overflow downstream:

[0227] L free (D)<0

[0228] or

[0229] L queue (D)≥L c (D)-S safe

[0230] in:

[0231] L free (D) is the downstream free lane capacity (m);

[0232] L queue (D) is the length of the queue in the downstream lane (meters);

[0233] L c (D) is the total lane length of the downstream section (m);

[0234] S safe is the exit safety distance (meters).

[0235] If the overflow risk condition is met, the maximum green light capacity of the downstream section is calculated as the maximum green light time limit of this intersection. The calculation formula for the maximum green light capacity is as follows:

[0236]

[0237] in:

[0238] G max (D) is the maximum green light capacity of the downstream (seconds);

[0239] L free (D) is the downstream free lane capacity (m);

[0240] L std is the standard queuing vehicle distance (meters).

[0241] For downstream turns with overflow risks, adjust the green light time at the corresponding stage of the intersection. The calculation formula is as follows:

[0242] T p ′ h =min(G max (D),T ph -S)+S

[0243] in:

[0244] T p ′ h is the green light time of the adjusted stage (seconds);

[0245] G max (D) is the maximum green light capacity of the downstream (seconds);

[0246] Tph The green light time of the original stage (seconds);

[0247] S is the reserved buffer time (seconds), which can be set to 10 seconds.

[0248] Finally, the signal cycle of the intersection is recalculated based on the adjusted stage green light time. The calculation formula is as follows:

[0249] C ′ (I) = ∑T p ′ h

[0250] in:

[0251] C ′ (I) is the adjusted intersection signal cycle (seconds);

[0252] T p ′ h The green light time of the adjusted stage (in seconds).

[0253] Through this calculation process, we can adjust the intersection's signal timing, reduce the risk of overflow on downstream roads, and improve traffic flow. Adjusting the signal cycle and green time effectively optimizes signal control at the intersection and avoids severe overflow congestion on downstream roads.

[0254] As for step S4, this step is used to evaluate the coordination benefits of intersection turns. By judging the overflow of each turn, the turns without overflow are screened out, their coordination benefits are calculated, and the turns with benefits greater than zero are screened as coordination candidates.

[0255] First, for each turn, if its demand flow rate DR(T) is greater than the actual flow rate FR(T), the turn is determined to be in an overflow state and cannot participate in green wave coordination. The judgment conditions for overflow turns are as follows:

[0256] DR(T)>FR(T)

[0257] in:

[0258] DR(T) is the required flow rate of steering (vehicles / hour);

[0259] FR(T) is the actual flow rate of the steering (vehicles / hour).

[0260] If the turn is in an overflow state, the turn is removed from the coordination candidate set, and the remaining turns that are not overflowing form a new coordination candidate set.

[0261] Next, for each turn in the coordination candidate set, calculate its coordination benefit R coor (T). The formula for calculating coordination benefits is as follows:

[0262]

[0263] in:

[0264] R coor (T) is the coordination benefit of steering;

[0265] FR(T) is the turning capacity in the optimal single intersection solution (vehicles / hour);

[0266] W coor (T) is the proportion of coordinated flow in the turn (dimensionless value);

[0267] FR(I) is the total traffic capacity of the intersection in the optimal single intersection solution (vehicles / hour);

[0268] N ph is the number of stages in the intersection signal cycle.

[0269] After calculating the coordination benefit of each turn, the turns with benefits greater than zero are selected as the final coordination candidate set. These coordination candidate turns have good coordination benefits and can achieve higher traffic efficiency in the green wave coordination scheme.

[0270] Ultimately, through screening and optimization, all turns with positive coordination benefits are retained as final coordination solution candidates. The coordination benefits of these turns help improve traffic flow, optimize signal timing, achieve green wave traffic, and enhance intersection capacity.

[0271] Regarding step S5, this step uses the breadth-based search algorithm (BFS) to expand the coordination path, form a signal coordination trunk line, and optimize the signal timing plan. The specific steps are as follows:

[0272] Step 1: Initialize the Shinkansen. The initialization process includes the following:

[0273] 1. From candidate to set TurnSet valid In the process, choose the turn with the largest coordination benefit max As the starting point of the trunk line, add it to the turn set to be expanded TurnSet expand .

[0274] 2. Assign a unique identifier Corr to the current trunk id .

[0275] 3. Initialize the intersection relationship field INTER of the trunk line to null, indicating that the current trunk line does not intersect with other trunk lines.

[0276] Step 2: Start the trunk extension process:

[0277] 2.1. From the to-be-expanded set to the TurnSet expand Select any one of the turns current .

[0278] 2.2. Turn current Perform deadlock loop judgment. If the following conditions are met, abandon Turn current And return to step 2.1:

[0279] Current Main Line Corr id It has intersected with other main lines;

[0280] and Turn current Coordinated turns belonging to other main lines.

[0281] 2.3、Turn current Join the current trunk line and mark it as the trunk line coordinated turn coor .

[0282] 2.4、If Turn current Corr with other trunk lines i ′ d If they intersect, update the intersection relationship:

[0283] INTER=Corr i ′ d

[0284] in:

[0285] Corr id is the identifier of the current trunk line;

[0286] Corr i ′ d is the identifier of the intersecting trunk line;

[0287] INTER is an intersection relation field, indicating the intersection of the current trunk line and other trunk lines.

[0288] 2.5. Turn current Upstream Turn up and downstream turn down To expand, Turn up and Turn down Add to the set to be expanded TurnSet expand .

[0289] 2.6. From Candidate to Set TurnSet valid Remove Turn current All turns in the current stage.

[0290] 2.7. Repeat steps 2.1 to 2.7 until TurnSet expand If it is empty, the current trunk generation is completed.

[0291] Step 3: Generate the trunk line Corr id Store in the trunk set. If the candidate turn set TurnSet valid If it is not empty, return to Step 1 and generate the next trunk line; otherwise, terminate the process.

[0292] Through this step, the signal coordination trunk line generation process based on the breadth search algorithm can be realized, the intersection signal timing can be optimized, and the coordination and traffic efficiency of the traffic flow can be improved.

[0293] Regarding step S6, this step divides the signal control units according to the topological relationship of the trunk line. A signal control unit is a set of intersections that can independently coordinate green waves. The types of signal control units include: single-point signal control units, trunk signal control units, and regional signal control units. The specific steps are as follows:

[0294] First, define the types of signal control units. Signal control units are divided according to the topology of the trunk line, mainly including the following types:

[0295] Single-point signal control unit: During the trunk line generation process, intersections that are not assigned to any trunk line are single-point signal control units. This means that these intersections do not participate in the coordination of any trunk line and perform signal control independently.

[0296] Trunk signal control unit: If a trunk line does not intersect with other trunk lines, it forms an independent trunk signal control unit. This type of signal control unit contains a trunk line that does not intersect with other trunk lines and can independently perform green wave coordination.

[0297] Regional signal control unit: When multiple trunk lines intersect to form a region, these trunk lines form a regional signal control unit. Due to the intersection of these trunk lines, a regional signal control unit may contain multiple intersecting trunk lines. These trunk lines cannot be used as independent signal control units for green wave coordination. Regional signal control units need to be merged based on the intersection of the trunk lines to form a coordinated signal control area.

[0298] Next, the trunk lines are merged according to the crossover situation to form a regional signal control unit. The specific steps are as follows:

[0299] Intersection relationship records for trunks: Record the intersection relationship between trunks based on the INTER attribute of each trunk. The INTER attribute indicates the intersection between a trunk and other trunks. If two trunks have an intersection, this intersection information is recorded in their intersection relationship.

[0300] Trunk merging: When there are crossing trunks, these crossing trunks are merged into a regional signal control unit based on the INTER attribute information. The regional signal control unit formed by multiple crossing trunks can coordinate their signal cycles and perform green wave control.

[0301] Signal control unit division: After the merger, the remaining unintersecting trunk lines continue to serve as trunk signal control units, while individual intersections are divided into single-point signal control units. The final signal control units include: single-point signal control units, trunk signal control units, and regional signal control units.

[0302] like Figure 2 As shown in the figure, according to the intersection relationship recorded by the INTER attribute of the trunk line, the trunk lines are associated to generate an area.

[0303] This step allows the intersection set to be divided into different types of signal control units based on the topological relationship of the trunk lines. Each signal control unit independently coordinates green waves according to its type, thereby improving overall traffic flow.

[0304] Regarding step S7, this step is used to determine the coordinated direction and phase difference of each intersection in the signal control unit, and calculate the phase difference based on the travel time and queue length of the coordinated turn. The specific steps are as follows:

[0305] First, determine the coordinated turn. For intersections within a signal control unit, the coordinated turn at the main road intersection is selected as the regional coordinated turn. If the intersection has two or more intersecting main roads, the coordinated turn at the main road with the higher priority is selected as the main coordinated turn at the core coordinated intersection. By default, the first generated main road has a higher priority, that is, the main road with the earlier main road number has a higher priority.

[0306] Next, we define the coordination phase. The signal control phase for the coordinated turn is defined as the coordination phase for the intersection. This phase corresponds to the coordinated signal phase in the intersection signal control scheme and indicates the coordinated signal control scheme for the intersection.

[0307] Then, calculate the coordinated phase difference. The coordinated phase difference is the phase difference of the coordinated steering, reflecting the phase difference between different steerings. The calculation formula is as follows:

[0308]

[0309] in:

[0310] Δφ is the phase difference of coordinated steering (seconds);

[0311] L turn is the length of the turning section (m);

[0312] V is the average driving speed on the turning section (m / s);

[0313] T ph is the green light time (seconds) of this stage;

[0314] W coor (T) is the proportion of coordinated flow in steering (dimensionless value).

[0315] The coordination phase difference is calculated based on the length of the turning section, driving speed, green time, and the coordinated traffic volume percentage for that turn. This phase difference can be used to adjust signal timing to ensure signal coordination and green wave passage between different signal control units.

[0316] Through this step, the signal coordination of each intersection within the signal control unit can be optimized, traffic conflicts between different intersections can be reduced, and traffic flow can be improved.

[0317] In summary, this paper proposes a method for dividing urban mesoscopic signal control units based on breadth-based search, aiming to optimize urban traffic signal control and improve traffic flow. The following is the overall process of this method:

[0318] 1. Get traffic information:

[0319] Collect traffic demand, upstream and downstream topology, signal configuration information, and historical traffic flow data for each intersection in the road network. This data is collected through fixed detection equipment, floating vehicle data, and manual surveys, and preprocessed into traffic characteristics that can be used for subsequent analysis.

[0320] 2. Calculate the optimal signal timing plan for the intersection:

[0321] The optimal signal cycle at each intersection is calculated based on traffic demand, signal control phases, and road capacity. Based on factors such as the number of phases, phase transition loss time, and phase saturation, the intersection signal cycle is determined and the green-to-signal ratio is allocated to optimize signal control strategies.

[0322] 3. Adjust intersection signal timing to avoid downstream overflow:

[0323] By calculating the length of the downstream queue, we can determine whether there is overflow risk. If overflow occurs, we adjust the intersection's signal timing based on the length of available lanes and the maximum green light capacity to reduce overflow pressure downstream and optimize green wave coordination.

[0324] 4. Calculate the coordination benefits of each intersection and screen coordination candidates:

[0325] Calculate the coordination benefit for each intersection and eliminate overflow turns. For turns that do not overflow, calculate their coordination benefits and select turns with benefits greater than zero as candidate turns for subsequent signal coordination.

[0326] 5. Generate signal coordination trunk:

[0327] A breadth-based search algorithm is used to select the candidate turn with the greatest coordination benefit and generate a signal coordination trunk. Based on trunk topology, intersections between trunks are processed to generate independent trunks and regional signal control units. Trunks are expanded and merged based on intersections to generate regional signal control units.

[0328] 6. Divide the signal control unit:

[0329] Based on the topology of the trunk lines, the road network is divided into different signal control units. These include single-point signal control units, trunk line signal control units, and regional signal control units. Single-point signal control units are independent intersections, trunk line signal control units are non-intersecting trunk lines, and regional signal control units are composed of multiple intersecting trunk lines.

[0330] 7. Determine the coordination direction and phase difference of the signal control unit:

[0331] Determine coordinated turns within the signal control unit and calculate a phase difference based on the turn travel time and queue length. The coordinated phase difference is calculated based on the length of the turn section, the travel speed of the section, and the green light time, ensuring signal coordination and green wave passage between different intersections.

[0332] 8. Output the signal control unit division results:

[0333] The final output is an optimized signal control unit division plan, which provides effective support for urban traffic signal control, ensures the implementation of the green wave coordination plan, and improves traffic smoothness.

[0334] The method of the present invention enables intelligent signal control at intersections in urban traffic networks, optimizing the allocation of signal cycles and green light durations, improving intersection capacity, reducing traffic congestion, and enhancing traffic flow. By rationally dividing signal control units, conflicts between different signal control units are reduced, improving green wave coordination, and ultimately achieving efficient traffic scheduling and optimization.

[0335] The device for dividing city mesoscopic signal control units based on breadth search described below and the method for dividing city mesoscopic signal control units based on breadth search described above can refer to each other.

[0336] Please see the attached Figure 3 The present invention also provides a device for dividing urban mesoscopic signal control units based on breadth search, comprising:

[0337] The data acquisition module 10 is used to obtain intersection traffic information and calculate relevant flow data;

[0338] Single intersection signal optimization module 20, used to calculate the optimal signal timing plan for each intersection;

[0339] Downstream overflow control module 30, used to adjust the release time of the intersection based on the downstream traffic capacity;

[0340] A coordination benefit calculation module 40 is used to calculate the coordination benefit of each turn and screen candidate turns;

[0341] A trunk generation module 50 is configured to expand the coordination path using a breadth search to form a signal coordination trunk;

[0342] The signal control unit division module 60 is used to divide the signal control unit;

[0343] The coordination direction and phase difference calculation module 70 is used to calculate the coordination direction and signal phase difference to achieve green wave coordinated control.

[0344] This device can be used to execute the above method, and its principles and technical effects are similar, so they will not be described here in detail.

[0345] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for dividing urban mesoscopic signal control units based on breadth search, characterized in that: The following steps are involved: Obtain multiple intersections and their traffic information in the road network; Calculating an optimal signal timing plan for each intersection based on the traffic information; Calculate the queue length of each intersection's downstream section and adjust the release time of the intersection based on the remaining downstream capacity; Calculate the coordination benefit of each turn and select turns with benefits greater than zero as coordination candidates, including: For each turn, if its required flow rate is greater than the actual flow rate, the turn is considered to be in an overflow state and does not participate in green wave coordination. Overflow turns are eliminated from all turns, and the remaining non-overflowing turns form the coordination candidate set. For each candidate turn in the coordination candidate set, calculate its coordination benefit: Where FR(T) is the turning capacity in the optimal single intersection solution, W coor (T) is the proportion of coordinated traffic flow in turning, FR(I) is the total traffic capacity of the intersection in the optimal single intersection solution, N ph is the total number of signal stages at the intersection; The turns with coordination benefits greater than zero are retained and enter the final coordination candidate set; Expand the coordination path based on the breadth search algorithm to form a signal coordination trunk, which includes: Step 1: Initialize the Shinkansen, including: (1) In the candidate turn set TurnSet valid In the process, choose the turn with the largest coordination benefit max As the starting point of the trunk line, add the turn set to be expanded TurnSet expand ; (2) Assign a unique identifier Corr to the current trunk id ; (3) Set the intersection relation field INTER of the trunk line to null, indicating that the current trunk line does not intersect with other trunk lines; Step 2: Trunk expansion Step 2.1, from TurnSet expand Select any one of the turns current ; Step 2.2, Turn current Perform coordination access verification. If the following conditions are met, the Turn is discarded. current And return to step 2.1: Current Main Line Corr id It has intersected with other main lines; and Turn current Intersection with other main lines; step2.3, Turn current Join the current trunk line and mark it as the trunk line coordinated turn coor ; step2.4, if Turn current Corr′ with other trunk lines id If they intersect, update the current trunk intersection relationship attributes: INTER=Corr′ id step2.5, turn current Upstream Turn up and downstream turn down To expand: Turn up and Turn down Add to the set to be expanded TurnSet expand ; Step 2.6: Turn Set from Candidate valid Remove Turn current All turns in the current stage; step 2.7, repeat steps 2.1 to 2.7 until TurnSet expand If it is empty, the current trunk generation is completed; Step 3: Generate the trunk line Corr id Stored in the trunk set, if the candidate turn set TurnSet valid If it is not empty, return to step 1 and generate the next trunk line; otherwise, terminate the process; Divide the signal control unit according to the trunk line topology relationship; The signal phase difference is calculated based on the main coordination direction of each trunk line to optimize the signal timing.

2. The method for dividing urban mesoscopic signal control units based on breadth search according to claim 1 is characterized in that: The step of obtaining multiple intersections in the road network and traffic information thereof includes: Obtain intersection traffic demand, intersection upstream and downstream topology, intersection signal control configuration, and historical record data; Data fusion is performed with turns as the minimum granularity to generate fields including intersection number, turn number, number of lanes, lane length, signal control stage number, actual intersection cycle, turn green-signal ratio, actual turn flow rate and demand flow rate, upstream and downstream turns, and corresponding traffic flow ratios.

3. The method for dividing urban mesoscopic signal control units based on breadth search according to claim 1, characterized in that: The step of calculating the optimal signal timing scheme for each intersection based on the traffic information includes: Calculate the optimal signal cycle based on the number of intersection stages, stage switching loss time and saturation: Among them, C(I) is the optimal signal period, N ph is the number of intersection stages, T loss is the single-stage switching loss time, X(I) is the total saturation of the intersection; For each signal control stage, calculate its saturation: Among them, X ph is the stage saturation, DR(T rep ) is the demand flow rate of the representative lane in the stage, FR(T rep ) is the capacity of the representative lane at the stage, that is, the upper limit of the traffic volume that the lane can pass under the current signal timing scheme, that is, the lane saturation flow rate multiplied by the lane green signal ratio; Based on the stage saturation and the total saturation of the intersection, the green-signal ratio of each stage is allocated according to the principle of balance: Among them, G(T) is the green-to-trust ratio of the stage; Calculate the green light time in each stage based on the green-to-signal ratio and intersection cycle: T ph =G(T)×(C(I)-N ph ×T loss ) Among them, T ph Green light time for the stage.

4. The method for dividing urban mesoscopic signal control units based on breadth search according to claim 1, characterized in that: The step of calculating the length of the detention queue at the downstream section of each intersection and adjusting the release time of the intersection based on the downstream remaining traffic capacity includes: Calculate the downstream lane queue length based on the downstream turn flow difference and signal cycle: Among them, L queue (D) is the queue length of the downstream lane, DR(D) is the demand flow rate of the downstream turn, FR(D) is the actual flow rate of the downstream turn, C(D) is the signal cycle of the downstream intersection, L std The standard queuing vehicle spacing; Calculate the capacity of the downstream free lane based on the total length of the downstream road section, the length of the delayed queue, and the safety distance: L free (D)=L c (D)-L queue (D)-S safe Among them, L free (D) is the capacity of the downstream free lane, L c (D) is the lane length of the downstream section, S safe is the safe distance to the exit; if the following conditions are met, it is determined that there is an overflow risk downstream: L free (D)<0 or L queue (D)≥L c (D)-S safe Based on the length of the downstream free lane, calculate the downstream maximum green light capacity, that is, the maximum green light time allowed for release at this intersection: Among them, G max (D) is the maximum green light capacity of the downstream, T std is the intersection saturation flow rate headway; For downstream turns with overflow risks, adjust the green light time at the corresponding stage of the intersection: T′ ph =min(G max (D),T ph -S)+S Among them, T′ ph is the green light time of the corresponding stage at this intersection, T ph is the green light time of the stage, and S is the reserved buffer time; Recalculate the total intersection cycle based on the adjusted green light time: C′(I)=∑T′ ph Wherein, C′(I) is the adjusted intersection signal cycle.

5. The method for dividing urban mesoscopic signal control units based on breadth search according to claim 1, characterized in that: The step of dividing the signal control units according to the trunk line topology relationship includes: The signal control unit is divided according to the trunk topology structure, and the signal control unit includes: Single-point signal control unit: an independent intersection, not part of any trunk line; Trunk signal control unit: a single trunk line without crossing; Regional signal control unit: a control area consisting of multiple cross-links; Merge trunk lines based on the intersection situation to form regional signal control units.

6. The method for dividing urban mesoscopic signal control units based on breadth search according to claim 1, characterized in that: The step of calculating the signal phase difference based on the main coordination direction of each trunk line includes: For the intersections within the signal control unit, the coordinated turn at the trunk intersection is selected as the regional coordinated turn; The signal control phase where the coordinated turn occurs is defined as the coordination phase of the intersection; Based on the travel time and queue length of the coordinated turn, the phase difference is calculated: Among them, L turn is the length of the turning section, V is the average speed of the section, T ph Green light time for the stage, W coor (T) is the proportion of coordinated traffic that is turned.

7. The method for dividing urban mesoscopic signal control units based on breadth search according to claim 6, characterized in that: The step of selecting the coordinated turning at the trunk intersection as the regional coordinated turning for the intersection within the signal control unit includes: If the intersection belongs to two or more intersecting trunk roads, the main coordinated turn is determined by the priority of the trunk road generation order.

8. A device for dividing city mesoscopic signal control units based on breadth search, used to execute the method for dividing city mesoscopic signal control units based on breadth search according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, used to obtain intersection traffic information and calculate relevant flow data; Single intersection signal optimization module, used to calculate the optimal signal timing plan for each intersection; Downstream overflow control module, used to adjust the release time of the intersection based on the downstream traffic capacity; Coordination benefit calculation module, used to calculate the coordination benefit of each turn and screen candidate turns; A trunk generation module, configured to expand the coordination path using breadth search to form a signal coordination trunk; A signal control unit division module is used to divide the signal control unit; The coordination direction and phase difference calculation module is used to calculate the coordination direction and signal phase difference to achieve green wave coordinated control.

Images