A traffic intelligent operation management and control and cross-region linkage command system

Abstract

The application belongs to the technical field of intelligent traffic operation control, and particularly relates to a traffic intelligent operation management and control and cross-region linkage command system, which comprises a congestion pre-judgment module, an inflow evaluation module, an undertaking and shunting module, a recursive splitting module and a linkage command module. The congestion early warning value is calculated based on the speed difference of the boundary road section and the falling duration, the abnormal boundary road section is identified, the propagation timing relationship is determined in combination with the historical propagation time difference, and the potential congestion road section is locked in advance. The accumulated difference of the traffic flow within the preset duration is calculated, compared with the remaining storage capacity, and used to determine whether the congestion is intensified. When the shunting is triggered, the step-by-step distribution is performed in combination with the shunting priority value and the additional travel time, so as to avoid the overload of the alternative path or the insufficient shunting. Finally, the coordinated release period is determined according to the first-stage shunting and the step-by-step distribution result and in combination with the predicted arrival time, so that the dredging efficiency of the cross-region collaborative control is effectively exerted.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation operation control technology, specifically a traffic intelligent operation management and cross-regional linkage command system. Background Technology

[0002] As urban transportation networks evolve from single-area management to cross-area collaborative management, the coupling effects between boundary road sections, internal road sections, and intersections in adjacent areas are becoming increasingly prominent. Especially at expressway entrances and exits, cross-regional arterial road connections, and road sections around large hubs, congestion often does not form independently on a single road section, but rather gradually spreads from traffic anomalies on boundary road sections into the interior of the area, further affecting signal control and traffic order in adjacent areas.

[0003] Existing technologies typically employ congestion detection methods based on real-time speed or saturation of local road segments. Once queues increase or average vehicle speeds significantly decrease on the target road segment, measures are taken through nearby guidance or manual intervention. While this approach can respond to congested road segments to some extent, existing technologies have several limitations: 1. Existing solutions often use the congestion status of the target road segment itself as a trigger condition, lacking joint analysis of the relationship between anomalies at the target area boundaries and the internal propagation within the area. Control measures can only be triggered after congestion has entered the area, resulting in control measures always lagging behind the rate of congestion spread.

[0004] 2. Most existing diversion methods directly divert traffic from the nearest intersection or fixed alternative route, lacking a coordinated assessment of new merging traffic, the remaining capacity of the target road segment, and the remaining capacity of the alternative route. This makes it impossible to determine whether the new merging traffic will indeed further exacerbate potential congestion, or whether the alternative route has the capacity to continuously accommodate the traffic, which can easily lead to over-diversion or under-diversion.

[0005] 3. Existing cross-regional joint control is mostly limited to information sharing or fixed timing linkage. It does not uniformly calculate the target diversion traffic volume, expected arrival time, and coordinated release time at intersections. This causes diverted vehicles to queue again at intersections in adjacent areas due to signal phase incoordination, weakening the effect of cross-regional collaborative control. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, embodiments of the present invention provide a traffic intelligent operation management and cross-regional linkage command system, which can effectively solve the problems involved in the prior art.

[0007] The objective of this invention can be achieved through the following technical solution: a traffic intelligent operation management and cross-regional linkage command system, comprising: a congestion prediction module, an inflow assessment module, a diversion module, a decomposition module, and a linkage command module.

[0008] The congestion prediction module is connected to the inflow assessment module, the inflow assessment module is connected to the receiving and diversion module, the receiving and diversion module is connected to the recursive splitting module, and the recursive splitting module is connected to the linkage command module.

[0009] The congestion prediction module determines the congestion warning value based on the speed difference between the historical average traffic speed and the real-time traffic speed of the boundary road segment of the target area, as well as the duration of the speed decrease. It also identifies abnormal boundary road segments by combining the road network topology and determines potential congested road segments based on the congestion propagation time sequence.

[0010] The merging assessment module constructs a sequence of new merging traffic flows heading towards potentially congested road sections within a preset time period, calculates the cumulative difference between the new merging traffic flow sequence and the expected outflow traffic volume, and determines whether the congestion of the potentially congested road sections is aggravated based on the comparison between the maximum value of the cumulative difference and the remaining vehicle capacity.

[0011] The diversion module determines the primary diversion junctions by following the reverse path of subsequent vehicles heading towards the potentially congested road section when it determines that the congestion will worsen. It then executes the primary diversion based on the remaining capacity of each candidate diversion path and the additional travel time.

[0012] The recursive splitting module allocates traffic flow level by level along the reverse path in the order of upstream intersections when there are still unallocated traffic flows after the initial splitting.

[0013] The joint command module implements cross-regional traffic guidance, signal coordination, and traffic dispatch based on the results of hierarchical allocation.

[0014] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention constructs a directed road network topology map, combines the speed difference to accumulate and calculate the congestion warning value, identifies the abnormal boundary road segment when the speed of the boundary road segment continues to drop abnormally but has not yet formed substantial congestion in the internal road segment, and forms a quantitative propagation time sequence relationship based on the historical propagation time difference, thereby locking the potential congested road segment in advance and moving the control opportunity forward to before the congestion is formed.

[0015] (2) The present invention constructs a sequence of newly added merging traffic flow and a sequence of expected outflow traffic flow, calculates the cumulative difference of traffic flow within a preset time period, and if the cumulative difference of traffic flow is greater than the remaining storage capacity of the potentially congested road section, it is determined that the newly added merging traffic flow will aggravate congestion and trigger diversion; when determining the diversion path, the diversion priority value and the additional travel time are combined to perform step-by-step allocation to avoid overloading of alternative paths or insufficient diversion.

[0016] (3) Based on the results of the first-level diversion and the step-by-step allocation, the present invention converts the target diversion traffic flow of each upstream intersection into the signal timing adjustment amount of the corresponding intersection; and determines the coordinated release period based on the expected arrival time, forming a continuous release zone on the cross-regional diversion path, eliminating the phenomenon of vehicles queuing again at the intersection of adjacent areas due to signal incoordination, and effectively exerting the diversion efficiency of cross-regional collaborative control. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the module connection of the present invention;

[0019] Figure 2 This is a flowchart illustrating the determination of the congestion propagation timing relationship in this invention.

[0020] Figure 3 This is a flowchart illustrating the logical judgment process of the present invention for determining whether newly added merging traffic will exacerbate potential congestion on road sections. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Reference Figure 1 As shown, the present invention provides a traffic intelligent operation management and cross-regional linkage command system, including: a congestion prediction module, an inflow assessment module, a diversion module, a decomposition module, and a linkage command module.

[0023] The congestion prediction module is connected to the inflow assessment module, the inflow assessment module is connected to the receiving and diversion module, the receiving and diversion module is connected to the recursive splitting module, and the recursive splitting module is connected to the linkage command module.

[0024] The congestion prediction module determines the congestion warning value based on the speed difference between the historical average traffic speed and the real-time traffic speed of the boundary road segment of the target area, as well as the duration of the speed decrease. It also identifies abnormal boundary road segments by combining the road network topology and determines potential congested road segments based on the congestion propagation time sequence.

[0025] The target area specifically refers to the traffic management area where operational efficiency needs to be prioritized; the adjacent area specifically refers to the surrounding area that has a direct coupling relationship with the target area in terms of road boundaries, traffic flow direction, or intersection control; the boundary road segment specifically refers to the input or output road segment connecting the target area and the adjacent area; and the internal road segment specifically refers to the road segment located inside the target area that receives further traffic from the boundary.

[0026] The road network topology relationship is as follows: First, obtain the road geometry data, road connection data, and road traffic direction data of the electronic map of the target area and adjacent areas.

[0027] Subsequently, the continuous road is divided using the intersection stop line, the ramp merging and diverging point, and the location where the turning mode can change as the dividing points. Road segments with the same traffic direction and located between two adjacent dividing points are identified as a road segment unit, and each road segment unit is assigned a unique identifier.

[0028] Then, each road segment unit is set as a road segment node, and a directed topological edge is established between the corresponding road segment nodes according to the actual allowed travel direction of vehicles between adjacent road segment units.

[0029] Next, the traffic control platform pre-sets the target area boundary range based on the actual management scope of the area to be controlled, and determines the connection attributes based on the positional relationship of each road segment unit relative to the target area boundary range and the directed connectivity relationship with adjacent road segment units. Among them, road segment units that are connected to the outside of the target area at one end and to the inside of the target area at the other end are determined as boundary road segment nodes, road segment units located inside the target area and having a continuous passage relationship with boundary road segment nodes are determined as internal road segment nodes, and connection positions with diversion or merging relationships are determined as intersection association nodes.

[0030] Finally, the length, number of lanes, average travel time, and upstream and downstream connections of each directed topological edge are stored as edge attributes to obtain a directed road network topology graph.

[0031] It should be noted that after obtaining the directed road network topology graph, directional consistency verification and connectivity verification can also be performed on it. The directional consistency verification specifically involves: verifying the starting and ending road segment nodes of each directed topology edge according to the road traffic direction data; if the direction of the directed topology edge is consistent with the actual allowed traffic direction of the vehicle, the directed topology edge is retained; otherwise, the directed topology edge is deleted or its direction is corrected.

[0032] The connectivity check specifically involves: starting with the boundary road segment node, traversing along the directed topological edge into the target area, and determining whether each internal road segment node has a directed path reachable from the boundary road segment node; if there is an unreachable internal road segment node or an interrupted directed path, then missing connecting edges are added or incorrect connecting edges are deleted according to the road connection relationship data, and the connectivity check is re-executed until a continuous directed passage link is formed between the boundary road segment node and the internal road segment node.

[0033] It should also be noted that the directed topological edges are established according to the actual permitted travel direction of vehicles. For two-way roads, two directed topological edges in opposite directions are established respectively. For connection relationships restricted by traffic organization, only the directed topological edges in the permitted direction are retained.

[0034] In a preferred embodiment of the present invention, the specific process of determining potential congested road sections is as follows: First, read the current system time of the traffic control platform, determine the start and end times of the current running period, and mark the current running day as a day type according to whether it is a working day or a non-working day.

[0035] Multiple historical dates that match the day type and have the same start and end times as the current operating segment are selected from the historical traffic database as historical sample dates.

[0036] For each boundary road segment, speed records for the corresponding time period within each historical sample date are extracted, and missing and outlier values ​​are removed to obtain the effective speed samples for that boundary road segment under each historical sample date.

[0037] The average speed of the effective speed samples under each historical sample date is calculated to obtain the historical average speed of the boundary road segment under each historical sample date. Finally, the arithmetic average of the historical average speeds corresponding to the multiple historical sample dates is calculated to obtain the average traffic speed of the boundary road segment under the current operating period.

[0038] Next, real-time speed records of each boundary road segment are received according to the preset sampling period, and the moving average value of the real-time speed records of multiple consecutive sampling periods before the current time is calculated. The result is used as the real-time traffic speed of the corresponding boundary road segment, and the difference between the real-time traffic speed and the average traffic speed of the corresponding boundary road segment is determined as the speed difference.

[0039] If the real-time traffic speed is continuously lower than the average traffic speed, the timer is accumulated from the sampling time when it first falls below the average traffic speed, and the accumulated time is determined as the duration of the speed drop; conversely, if the real-time traffic speed recovers to a level not lower than the average traffic speed, the duration of the speed drop is reset to zero and the timer is restarted.

[0040] The product of the absolute value of the speed difference and the duration of the speed decrease is used as the congestion warning value, and the boundary road segments that exceed the preset abnormal judgment value are identified as abnormal boundary road segments.

[0041] The preset anomaly determination value is obtained through experimental calibration. Specifically, it involves obtaining boundary road segment speed samples from non-congested historical dates that are consistent with the current operating day type, calculating the congestion warning value sequence for each sample date, and taking the median of the congestion warning value sequence as the preset anomaly determination value.

[0042] Finally, by combining the congestion propagation time sequence relationship between the abnormal boundary road segment and each associated road segment, potential congested road segments are identified from each associated road segment. Specifically, the road segment node corresponding to the abnormal boundary road segment is taken as the starting node, and the directed topological edge from the boundary road segment to the target area is traversed layer by layer in the directed road network topology graph. All internal road segment nodes that can be reached from the starting node and are still located inside the target area are extracted, and the corresponding road segments are identified as associated road segments.

[0043] For each historical sample date, read the congestion warning value sequence of the abnormal boundary road segment and each associated road segment at each sampling time in the corresponding historical period. For each associated road segment, determine the sampling time when the congestion warning value of the abnormal boundary road segment and the associated road segment first exceeds the preset abnormal judgment value in the same historical sample date, and record it as the boundary threshold crossing time and the associated threshold crossing time.

[0044] The difference between the threshold crossing time of the associated road segment and the threshold crossing time of the boundary is determined as the propagation time difference under the historical sample date. Historical sample dates with a propagation time difference greater than zero are retained as valid propagation events. The number of occurrences of valid propagation events corresponding to each associated road segment is counted. The propagation time difference of the associated road segment under each valid propagation event is arithmetically averaged to obtain the representative propagation time difference of the associated road segment relative to the abnormal boundary road segment.

[0045] Retain the associated road segments whose representative propagation time difference is within the preset propagation time range and whose effective propagation event occurrences are greater than the preset frequency threshold, and sort them in ascending order of representative propagation time difference; when the representative propagation time differences are the same, sort them in descending order of effective propagation event occurrences, and use the sorting results as the congestion propagation time sequence relationship, and identify the associated road segment that ranks first in the sorting results as the potential congested road segment.

[0046] Reference Figure 2 As shown, the congestion propagation time sequence relationship specifically refers to the sorting relationship formed by the order of congestion warning values ​​exceeding the threshold of abnormal boundary road segments and the associated road segments under each historical sample date, the propagation time difference, and the number of effective propagation events for each associated road segment.

[0047] The propagation time difference specifically refers to the difference between the time when the congestion warning value first exceeds the threshold of the associated road segment and the time when the congestion warning value first exceeds the threshold of the abnormal boundary road segment. If the propagation time difference is less than or equal to zero, it indicates that the associated road segment was not affected after the abnormal boundary road segment, or that the two do not have a stable upstream propagation relationship, and therefore it is not considered a valid propagation event.

[0048] In this embodiment, associated road segments whose propagation time difference is within a preset propagation time range and whose number of effective propagation events is greater than a preset frequency threshold are considered as valid candidates. The preset frequency threshold is preferably set to more than 60% of the number of events in the current historical samples of the same type.

[0049] It should be noted that the preset sampling period and preset propagation time range are determined by statistical analysis of the time distribution of historical samples of the same type. Specifically, multiple historical sample dates that are consistent with the current running segment are selected, and the adjacent collection time intervals of the detection records of each boundary road segment and the historical propagation time difference from the abnormal boundary road segment to each associated road segment are extracted.

[0050] The intervals between adjacent sampling times and the historical propagation time differences are statistically analyzed. The interval between adjacent sampling times that occurs most frequently is taken as the preset sampling period, and the maximum value among the historical propagation time differences is taken as the preset propagation time range.

[0051] It should also be noted that, based on the temporal relationship of congestion propagation, and in order of increasing time difference and decreasing frequency of congestion events, the associated road segments at the top of the list are prioritized as potential congested road segments. This is because the smaller the time difference, the faster the spread of the impact of congestion on the abnormal boundary road segment on the associated road segment, and the more events, the higher the historical confidence of the propagation pattern. This allows for the accurate identification of the internal road segments most likely to experience congestion spread.

[0052] The merging assessment module constructs a sequence of new merging traffic flows heading towards potentially congested road sections within a preset time period, calculates the cumulative difference between the new merging traffic flow sequence and the expected outflow traffic volume, and determines whether the congestion of the potentially congested road sections is aggravated based on the comparison between the maximum value of the cumulative difference and the remaining vehicle capacity.

[0053] Considering that when new merging traffic continues to flow in, if the outflow capacity of potentially congested road sections is limited, the cumulative difference in traffic flow on potentially congested road sections caused by the new merging traffic will exceed the remaining storage capacity. In this case, the road section will trigger overflow due to exceeding its carrying capacity. This will not only exacerbate its own congestion but will also spread to upstream intersections, forcing more subsequent vehicles into queues, thereby affecting the traffic efficiency of the entire road network and the feasibility of diversion scheduling.

[0054] Reference Figure 3As shown, based on this, the process of determining whether newly added merging traffic will exacerbate congestion on potentially congested road sections is as follows: read the vehicle detection records and path identification results of each upstream entrance lane, connecting road section and adjacent intersection of the potentially congested road section.

[0055] The system filters out subsequent vehicles that are expected to pass through potentially congested road sections within a preset time period, and determines the current road section of each subsequent vehicle, the merging direction towards the potentially congested road section, and the expected arrival time.

[0056] According to the preset statistical time period, subsequent vehicles with the same merging direction and expected arrival time falling within the same statistical time period are aggregated to obtain the new merging traffic flow corresponding to each merging direction in each statistical time period. The new merging traffic flow of each statistical time period is then arranged in chronological order to form a new merging traffic flow sequence.

[0057] Read the current real-time traffic speed of potentially congested road sections, and extract the average traffic speed and average traffic capacity for the same time period under the corresponding historical sample dates.

[0058] The preset duration is divided into multiple consecutive statistical periods, and the historical average traffic capacity corresponding to each statistical period is determined.

[0059] The ratio of the current real-time traffic speed of a potentially congested road segment to the average traffic speed of the corresponding historical period is multiplied by the historical average traffic capacity of each statistical period. The result is used as the expected outflow traffic volume for each statistical period, and an expected outflow traffic volume sequence is constructed.

[0060] The theoretical maximum vehicle storage capacity is calculated based on the length of the potentially congested road segment and the number of lanes. The formula for calculating the theoretical maximum vehicle storage capacity is as follows: .

[0061] in Indicates the length of a potentially congested road segment; This indicates the number of lanes, that is, the number of lanes in one direction or two directions on this road segment; This indicates the equivalent length occupied by a single vehicle, that is, the longitudinal length occupied by a vehicle when it is stationary or moving at low speed. It is usually taken as the standard vehicle length, such as 4.5m. This indicates the safety distance, which is the minimum safe distance that vehicles should maintain. It is determined based on the road section's design speed or historical headway. For example, it is 2m when the design speed is ≤60km / h, and 5m otherwise. This represents the theoretical maximum vehicle capacity, which is the maximum number of vehicles that can be accommodated in a potentially congested road section while maintaining safe distances.

[0062] In this formula, This represents the total lane length resource of a road segment, i.e., the total longitudinal space available for all lanes; This represents the average longitudinal length occupied by each vehicle; dividing the two, i.e., dividing the total space by the space occupied by a single vehicle, gives the theoretical maximum number of vehicles that the road segment can accommodate.

[0063] The current occupied parking space is determined based on the current traffic flow status. This current occupied parking space can be determined based on the number of vehicles detected in the current road segment, and is denoted as [missing information]. Then, the theoretical maximum storage capacity... With the current number of vehicles in storage The difference between them is taken as the remaining number of vehicles in storage.

[0064] The preset time period is divided into multiple consecutive statistical periods. The newly added inflow traffic and the expected outflow traffic within each period are counted separately, and the traffic flow difference between the two is calculated.

[0065] The traffic flow differences for each statistical period are summed up in chronological order to obtain the cumulative difference for each statistical period.

[0066] The maximum value of the cumulative difference for each statistical period is taken as the cumulative traffic flow difference, and it is compared with the remaining vehicle storage capacity. When the maximum value of the cumulative traffic flow difference is greater than the remaining vehicle storage capacity, it is determined that the newly added merging traffic will aggravate the congestion of the potential congestion section; otherwise, it is determined that the newly added merging traffic will not aggravate the congestion of the potential congestion section.

[0067] It should be noted that the preset duration is determined based on the historical congestion duration distribution of the potential congested road segment. Specifically, the duration from the first time the congestion warning value exceeds the preset anomaly judgment value to the dissipation of congestion is extracted from multiple historical sample dates for the road segment, and the 90th percentile of these durations is taken as the preset duration. The preset statistical period is consistent with the real-time speed sampling period of the road segment.

[0068] The diversion module determines the primary diversion junctions by following the reverse path of subsequent vehicles heading towards the potentially congested road section when it determines that the congestion will worsen. It then executes the primary diversion based on the remaining capacity of each candidate diversion path and the additional travel time.

[0069] The process of performing primary diversion based on the remaining capacity and additional travel time of each candidate diversion path is as follows: taking the road segment node corresponding to the potential congested road segment as the starting point of the reverse search, tracing back the upstream road segment node segment by segment along the opposite direction of the vehicle's journey to the potential congested road segment, and taking the upstream connecting path formed by the backtracking as the reverse path. The reverse path is not an arbitrary search path that extends upstream, but a reverse connecting path with the potential congested road segment as the endpoint and the direction in which the vehicle actually travels to the potential congested road segment as a constraint.

[0070] In sequence, determine whether the connection position corresponding to each upstream road segment node is connected to other passable paths in addition to the original passable path.

[0071] When the first connection location that meets the above conditions is found, the connection location is determined as the primary branching intersection, and the search upstream is stopped.

[0072] Starting from the primary diversion junction, all alternative routes that can bypass potentially congested road sections and still point to the original vehicle travel direction are extracted as candidate basic routes for primary diversion.

[0073] For each alternative route, the real-time traffic flow within a preset time period is first counted by the cross-section detector, and the average traffic flow for the same time period under the corresponding historical sample date is extracted. The average traffic flow is then converted into the capacity of vehicles within the preset time period. For example, if the average traffic flow is 600 vehicles per hour and the preset time period is 5 minutes, then the capacity of vehicles is 600 × (5 / 60) = 50 vehicles.

[0074] The difference between the available traffic flow and the current real-time traffic flow is determined as the remaining capacity of the alternative travel route.

[0075] The estimated travel time of the alternative route is calculated based on the length of each segment of the alternative route and the real-time traffic speed. The estimated travel time of the original route is calculated in the same way, and the difference between the two is determined as the additional travel time.

[0076] The diversion priority value is calculated based on the remaining capacity and additional travel time of each alternative travel route. The specific calculation process is as follows: For each alternative travel route, the remaining capacity and additional travel time of the alternative travel route are determined, and 1 is added to the value of the additional travel time to obtain the corresponding modified additional travel time.

[0077] Divide the remaining capacity of the alternative travel path by the modified additional travel time to obtain the diversion priority value corresponding to the alternative travel path.

[0078] The diversion priority values ​​of each alternative travel path are compared. The higher the diversion priority value, the higher the diversion priority value, indicating that the alternative travel path has a high capacity while having a low additional travel cost.

[0079] The newly added traffic flow is allocated according to the diversion priority value from large to small until the corresponding alternative traffic path reaches the upper limit of the remaining capacity.

[0080] In a preferred embodiment of the present invention, the primary diversion process further includes: removing candidate diversion paths that still pass through potentially congested road sections from the candidate diversion paths, and using the remaining paths as updated candidate diversion paths.

[0081] The candidate routing paths are sorted from smallest to largest based on their additional travel time after the update. If the additional travel time is the same, they are sorted from largest to smallest based on their remaining capacity. The primary routing is then executed based on the sorting results.

[0082] It should be noted that if there are no candidate diversion paths with a remaining capacity greater than 0 at the primary diversion junction, the path will be directly reversed upstream to the next junction as the primary diversion junction.

[0083] It should also be noted that the sorting process is used to establish a clearer execution order for traffic diversion priorities. When the diversion priorities of several alternative routes are close, the route with the shorter additional travel time is selected first, which increases the probability of vehicles accepting guidance. When the additional travel time is the same, the route with the larger remaining capacity is selected, which further reduces the risk of candidate routes being quickly filled. Through this execution order, the primary diversion can maintain both technical feasibility and route accessibility and capacity stability.

[0084] The recursive splitting module allocates traffic flow level by level along the reverse path in the order of upstream intersections when there are still unallocated traffic flows after the initial splitting.

[0085] The process of allocating traffic flow in the reverse path according to the upstream branch intersection sequence is as follows: newly added merging traffic flow that has not been allocated after the first-level diversion is determined as the current traffic flow to be allocated.

[0086] Along the reverse path, upstream intersections are selected sequentially in the direction away from potential congestion sections, and candidate diversion path extraction, remaining capacity calculation, and diversion priority value calculation are repeated for each upstream intersection.

[0087] The allocation amount of traffic to be allocated to each candidate diversion path at the current upstream intersection is determined based on the diversion priority value ratio of each candidate diversion path, and the current traffic to be allocated is allocated using the corresponding remaining capacity as the upper limit of the allocation amount.

[0088] After the current upstream intersection completes its allocation, the remaining traffic flow to be allocated is passed to the next upstream intersection for further processing. Once all upstream intersections along the reverse path have been processed, the final allocation result of the diversion path corresponding to each upstream intersection is obtained.

[0089] It should be noted that the step-by-step allocation does not push all the traffic to be allocated to the distant alternative path at once, but rather prioritizes reducing the risky traffic that is about to merge at the intersections closest to the potentially congested road sections; only when the initial diversion capacity is insufficient will it be pushed to the upstream intersections.

[0090] The joint command module implements cross-regional traffic guidance, signal coordination, and traffic dispatch based on the results of hierarchical allocation.

[0091] The specific process of implementing cross-regional traffic guidance, signal coordination and traffic scheduling based on the hierarchical allocation results is as follows: First, based on the primary diversion results and the hierarchical allocation results, the target diversion traffic flow of each upstream intersection on each diversion path is summarized.

[0092] Subsequently, for each diversion path, route guidance information including the target intersection, recommended driving direction, and diversion time period is generated, and traffic guidance instructions are issued to vehicles located at the corresponding upstream intersection and its upstream road segment.

[0093] Then, based on the target diversion traffic flow, current real-time traffic flow, and real-time traffic status of the boundary road sections corresponding to each diversion path, the green light ratio adjustment and phase duration adjustment of each release direction at the corresponding intersection are calculated and written into the signal timing parameters of the corresponding intersection.

[0094] The process of converting the target diversion traffic volume into signal timing parameters is as follows: based on the saturation constraints and lane saturation flow rates of each diversion path, the basic green light ratio for each release direction is calculated using the equal saturation principle; based on the target diversion traffic volume, the lane demand for the corresponding release direction of the diversion path is incrementally corrected, and the increase in green light duration is determined according to the ratio of incremental traffic volume to saturation flow rate; and the phase difference between adjacent intersections is calculated in conjunction with the road segment travel time to form a green wave coordination control scheme.

[0095] For example, when a certain diversion path is expected to generate additional diversion traffic of... The saturation flow rate of the critical lanes along this path is The original green light duration was If the signal cycle duration is C, then the adjusted green light duration is... .

[0096] Finally, the traffic control units of the target area and adjacent areas implement release control synchronously according to the updated signal timing parameters to guide the newly merging traffic to cross the area according to the predetermined diversion path.

[0097] It should also be noted that if, after traversing along the reverse path to the upstream boundary entrance intersection of the target area, the sum of the remaining capacity of each candidate diversion path is still insufficient to fully accommodate the current unassigned traffic flow, then further searching along the reverse path will cease. The remaining unassigned traffic flow will be used as the mandatory control flow at the entrance of the target area, generating detour guidance information for vehicles upstream of the upstream intersection, and adjusting the signal timing scheme of the relevant intersections upstream of that entrance to restrict vehicles from continuing to enter the target area until the potentially congested road sections return to normal traffic levels.

[0098] In a preferred embodiment of the present invention, the process of implementing cross-regional traffic guidance, signal coordination and traffic scheduling further includes: calculating the expected arrival time of the diverted vehicles at each intersection based on the target diversion traffic volume corresponding to each diversion path and the expected travel time of each road segment along the path.

[0099] Centered on the expected arrival time, and combined with the current cycle length and phase structure of the corresponding intersection, the coordinated release period for the corresponding release direction of each intersection is determined.

[0100] The start time, duration, and phase difference of the green light at each intersection on the same diversion path are adjusted in a coordinated manner according to the coordinated release period. The release ratio of vehicles heading towards potentially congested sections to those heading towards diversion paths is determined based on the real-time traffic flow at the boundary sections.

[0101] It should be noted that the coordinated release period specifically refers to the time window during which the corresponding release direction of the intersection should be able to pass continuously when diverted vehicles are expected to arrive at a certain intersection. Based on the target diversion traffic volume and the expected arrival time of vehicles for each diversion path, the release requirements for each diversion path are first determined, and then the start time, duration and phase difference of the green light are uniformly calculated for multiple intersections on the diversion path, thereby forming a continuous release zone that propagates along the path.

[0102] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A traffic intelligent operation control and cross-regional joint command system, characterized in that, include: The congestion prediction module determines the congestion warning value based on the speed difference between the historical average traffic speed and the real-time traffic speed of the boundary road segment of the target area, as well as the duration of the speed decrease. It also identifies abnormal boundary road segments based on the road network topology and potential congested road segments based on the congestion propagation time sequence. The merging assessment module constructs a sequence of new merging traffic flows heading towards potentially congested road sections within a preset time period, calculates the cumulative difference between the new merging traffic flow sequence and the expected outflow traffic volume, and determines whether the congestion of the potentially congested road sections is aggravated based on the comparison between the maximum value of the cumulative difference and the remaining parking capacity. The diversion module determines the primary diversion intersections by following the reverse path of subsequent vehicles heading towards the potentially congested road section when it determines that the congestion will worsen. It then executes the primary diversion based on the remaining capacity of each candidate diversion path and the additional travel time. The recursive splitting module allocates traffic flow level by level along the reverse path in the order of the upstream intersections when there are still unallocated traffic flows after the initial splitting. The joint command module implements cross-regional traffic guidance, signal coordination, and traffic dispatch based on the results of hierarchical allocation.

2. The intelligent traffic operation management and cross-regional joint command system according to claim 1, characterized in that, The specific road network topology is as follows: Using the target area and adjacent areas as road segments as road segment nodes, and the traffic connection relationships between adjacent road segments as topological edges, a directed road network topology graph containing boundary road segments, internal road segments, and intersection connections is constructed.

3. The intelligent traffic operation control and cross-regional joint command system according to claim 2, characterized in that, The specific process for identifying potentially congested road sections is as follows: Obtain historical traffic data corresponding to the current runtime, and extract the average traffic speed of each boundary road segment in the target area within the corresponding historical time period; Collect the real-time speed of each boundary road segment in the target area, and calculate the speed difference between the real-time speed of each boundary road segment and the average traffic speed, as well as the duration of the speed decrease. The product of the absolute value of the speed difference and the duration of the speed decrease is used as the congestion warning value, and the boundary road segment with the congestion warning value greater than the preset abnormal judgment value is determined as the abnormal boundary road segment. Based on the directed connection relationship between abnormal boundary road segments and internal road segments in the directed road network topology graph, the associated road segments connected to the abnormal boundary road segments are extracted along the travel direction towards the target area. By combining the congestion propagation time sequence relationship between abnormal boundary road segments and each associated road segment, potential congested road segments are identified from each associated road segment.

4. The intelligent traffic operation control and cross-regional joint command system according to claim 3, characterized in that, The specific temporal relationship of the congestion propagation is as follows: The moment when the congestion warning value of the abnormal boundary road segment and each related road segment first exceeds the preset abnormal judgment value in each historical time period is determined respectively. Calculate the time difference between each associated road segment and the abnormal boundary road segment, and retain the associated road segments with a time difference greater than zero; The relevant road segments are sorted in ascending order of time difference. When the time differences are the same, they are sorted in descending order of the frequency of the corresponding congestion events in the historical period. The sorting results are used as the congestion propagation time sequence.

5. The intelligent traffic operation control and cross-regional joint command system according to claim 1, characterized in that, The process for determining whether congestion on potentially congested road sections is exacerbated is as follows: Within a preset time period, subsequent vehicles that will enter potentially congested sections of the traffic path are selected, and a new merging traffic sequence is constructed according to the merging direction and the expected arrival time. Based on the real-time traffic speed of potentially congested road sections and the corresponding historical traffic data for the corresponding time periods, the estimated outflow of traffic from the potentially congested road sections within a preset time period is determined. The remaining vehicle capacity for potential congestion sections is determined based on the section length, number of lanes, and current traffic flow. The cumulative difference in traffic flow for potentially congested road sections is obtained by subtracting the newly added inflow traffic sequence from the expected outflow traffic flow at the same time point. When the maximum value of the cumulative difference in traffic flow is greater than the remaining number of vehicles, it is determined that the newly added merging traffic will exacerbate the congestion of the potentially congested road segment; otherwise, it is determined that the newly added merging traffic will not exacerbate the congestion of the potentially congested road segment.

6. The intelligent traffic operation control and cross-regional joint command system according to claim 2, characterized in that, The process of performing primary traffic diversion based on the remaining capacity and additional travel time of each candidate diversion path is as follows: In a directed road network topology graph, starting from a potentially congested road segment, the upstream connecting path formed by tracing back the direction of vehicles traveling towards the potentially congested road segment in reverse is used as the reverse path. Following the reverse path of subsequent vehicles heading towards the potential congestion section, identify the intersection closest to the potential congestion section that connects to other routes besides the original route as the primary diversion intersection, and use the other routes as alternative routes. The system obtains real-time traffic flow and corresponding historical traffic data for each alternative route, determines the capacity of the alternative route within a preset time period, and uses the difference between the capacity and the current traffic flow as the remaining capacity. Alternative routes with a remaining capacity greater than zero are selected as candidate diversion routes. Determine the additional travel time of each candidate diversion path relative to the original travel path, and calculate the diversion priority value of each candidate diversion path in combination with the remaining capacity. Newly arriving traffic is diverted first-level according to the diversion priority value from large to small, with the corresponding remaining capacity as the upper limit.

7. A traffic intelligent operation control and cross-regional linkage command system according to claim 6, characterized in that, The primary-level diversion process also includes: Among the candidate diversion paths, those that still pass through potentially congested road sections are removed, and the remaining paths are used as the updated candidate diversion paths. The candidate routing paths are sorted from smallest to largest based on their additional travel time after the update. If the additional travel time is the same, they are sorted from largest to smallest based on their remaining capacity. The primary routing is then executed based on the sorting results.

8. The intelligent traffic operation control and cross-regional joint command system according to claim 1, characterized in that, The process of allocating resources step-by-step along the reverse path according to the order of the upstream branch intersections is as follows: Newly arriving traffic that has not entered a candidate diversion path after the primary diversion is treated as traffic to be allocated. Along the reverse path, upstream intersections are selected sequentially in the direction away from potential congestion sections. The real-time traffic flow, capacity, and additional travel time of each candidate diversion path at the current upstream intersection are obtained to determine the remaining capacity and diversion priority value of each candidate diversion path. Based on the proportion of the diversion priority value of each candidate diversion path to the sum of the diversion priority values ​​of all candidate diversion paths at the current upstream intersection, the traffic flow to be allocated for each candidate diversion path is determined, and the remaining capacity of each candidate diversion path is used as the upper limit of the corresponding traffic flow to be allocated. The remaining traffic flow to be allocated after the current upstream intersection has been allocated is used as the traffic flow to be allocated at the next upstream intersection, until the hierarchical allocation along the reverse path is completed.

9. A traffic intelligent operation control and cross-regional linkage command system according to claim 1, characterized in that, The specific process of implementing cross-regional traffic guidance, signal coordination, and traffic scheduling based on the hierarchical allocation results is as follows: Based on the primary diversion results and the hierarchical allocation results, the target diversion traffic volume for each upstream intersection corresponding to the diversion path is determined; Based on the traffic flow of each target diversion route, generate corresponding diversion path guidance information and issue traffic guidance instructions to the vehicles corresponding to the target diversion traffic flow. Adjust the signal timing parameters of the intersections corresponding to each diversion path based on the target diversion traffic volume and real-time traffic volume of each diversion path; Traffic scheduling is implemented in the area where the potential congestion section is located and adjacent areas based on the adjusted signal timing parameters, so as to guide newly merging traffic to be diverted to the corresponding diversion path according to the hierarchical allocation results.

10. A traffic intelligent operation control and cross-regional linkage command system according to claim 1, characterized in that, The process of implementing cross-regional traffic guidance, signal coordination, and traffic scheduling also includes: Based on the target diversion traffic volume and expected arrival time of each diversion path, determine the coordinated release time period at the intersections on each diversion path; According to the coordinated release period, the start time, duration and phase difference of the green light at each intersection on the same diversion path are adjusted in a coordinated manner so that diverted vehicles can pass continuously along the diversion path; Based on the real-time traffic flow of the area where the potential congestion section is located and the boundary sections of adjacent areas, the release ratio of vehicles heading towards the potential congestion section and those heading towards diversion routes is dynamically adjusted.

Images