An adaptive traffic coordination method and system based on big data analysis

Abstract

The application discloses a kind of self-adapting traffic coordination method and system based on big data analysis, comprising the following steps: using traffic volume empirical model to predict the maximum traffic volume of each traffic line in coordination area, and using maximum traffic volume to carry out threshold screening to select coordination line for traffic line, in turn, select replacement line for each coordination line in all traffic lines, and coordination line and all replacement lines of coordination line form a coordination whole;Set real-time traffic volume expected value and communication rate maximum value, based on the real-time traffic volume expected value and real-time traffic volume limit minimum value and real-time traffic volume, each coordination unit in each coordination whole is carried out real-time command scheduling of traffic vehicle.This application uses traffic volume empirical model to select coordination line in coordination area, and each coordination unit in each coordination whole is carried out real-time command scheduling of traffic vehicle, to realize self-adapting equalization of the traffic pressure of each coordination unit.

Description

Technical Field

[0001] This invention relates to the field of traffic management technology, specifically to an adaptive traffic management method and system based on big data analysis. Background Technology

[0002] In recent years, urban rail transit has developed rapidly, with passenger flow experiencing explosive growth. To ensure that urban rail transit system operation and management departments have real-time access to the dynamic distribution of passenger flow across multiple scales—network, lines, sections, stations, and trains—and to take proactive measures during peak hours, holidays, periods of high passenger volume at stations, and emergencies, real-time monitoring of passenger flow across the urban rail transit network, lines, sections, stations, and trains is necessary. However, due to limitations in basic data, real-time passenger flow monitoring relies on real-time estimation of passenger flow status. Real-time estimation of urban rail transit passenger flow status is the foundation for real-time passenger flow monitoring, passenger flow control, and train scheduling, playing a crucial role in the organization of urban rail transit transportation.

[0003] In existing technologies, traffic management mainly relies on human judgment, which is highly subjective and has poor timeliness. Summary of the Invention

[0004] The purpose of this invention is to provide an adaptive traffic management method and system based on big data analysis, so as to solve the technical problems of existing traffic management relying on human judgment, which is highly subjective and has poor timeliness.

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

[0006] An adaptive traffic management method based on big data analytics includes the following steps:

[0007] Step S1: Predict the maximum traffic volume of each traffic route in the auxiliary management area using the traffic volume experience model, and select auxiliary management routes by thresholding the traffic routes using the maximum traffic volume. Then, select replacement routes for each auxiliary management route from all traffic routes, and combine the auxiliary management route and all its replacement routes into an auxiliary management system. The replacement routes have the same traffic effect as the auxiliary management routes.

[0008] Step S2: Take the assisted lines and the real-time traffic volume on each assisted line as a single assisted line unit, and also take the replacement lines and the real-time traffic volume on the replacement lines as a single assisted line unit, and count the real-time traffic volume of each assisted line unit.

[0009] Step S3: Set the expected real-time traffic volume and the maximum communication rate. Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, conduct real-time command and dispatch of vehicles in each auxiliary management unit in each auxiliary management system to achieve adaptive balance of traffic pressure in each auxiliary management unit.

[0010] As a preferred embodiment of the present invention, the step of predicting the maximum traffic volume of each traffic route in the coordinated management area using a traffic volume empirical model includes:

[0011] The regional features of the assisted management area and the route features of the traffic route are extracted sequentially, and the regional features and the route features are output to the traffic volume empirical model. The traffic volume empirical model outputs the maximum traffic volume of the traffic route.

[0012] The construction of the traffic volume empirical model includes:

[0013] Multiple traffic routes were selected as sample routes in different areas, and the real-time traffic volume of the sample routes was monitored. The real-time traffic volume of the sample routes was averaged to obtain the average real-time traffic volume as the maximum traffic volume of the sample routes.

[0014] Extract the line features of the sample line and the regional features of the area to which the sample line belongs. Use the line features and regional features as input terms of the BP neural network, and use the maximum traffic volume of the sample line as the output term of the BP neural network.

[0015] By using a BP neural network to perform convolution training on the input and output terms, an empirical traffic volume model representing the mapping relationship between line features, regional features and maximum traffic volume is obtained.

[0016] The functional expression of the traffic volume empirical model is:

[0017] T = BP(S, P);

[0018] In the formula, T is the character representing the maximum traffic volume, S is the character representing the line characteristics, P is the character representing the area characteristics, and BP is the character representing the BP neural network.

[0019] As a preferred embodiment of the present invention, the step of selecting the auxiliary traffic routes by threshold screening using the maximum traffic volume includes:

[0020] The maximum throughput of each traffic route is compared with the throughput threshold.

[0021] When the maximum traffic volume of a passage is greater than or equal to the traffic volume threshold, the passage will be marked as a regulated passage.

[0022] If the maximum traffic volume of a passage is less than the traffic volume threshold, the passage will be marked as a non-managed passage.

[0023] As a preferred embodiment of the present invention, the step of forming a unified auxiliary management system by comprising the auxiliary management line and all its replacement lines includes:

[0024] The assisted management lines and all their replacement lines are sequentially grouped into a single assisted management unit, and the inclusion relationships between all assisted management units are compared.

[0025] If there is an inclusion relationship between any two co-managed groups, then the co-managed group that is included in the two co-managed groups will be removed.

[0026] If there is no inclusion relationship between any two co-managed entities, then both co-managed entities will be retained.

[0027] As a preferred embodiment of the present invention, the real-time command and dispatch of vehicles in each auxiliary management unit within each auxiliary management system based on the expected value of real-time traffic volume, the minimum value of real-time traffic volume, and the real-time traffic volume includes:

[0028] The real-time coordination role of a coordination unit is determined based on its real-time traffic volume. This real-time coordination role includes both a real-time traffic volume increaser and a real-time traffic volume decreaser.

[0029] When the real-time throughput of the co-management unit is less than the minimum real-time throughput limit, the co-management unit is designated as the real-time throughput booster, and the real-time throughput booster determines the real-time boosting requirement for the real-time throughput mitigation unit to perform co-management monitoring.

[0030] When the real-time traffic volume of the auxiliary management unit is greater than or equal to the minimum real-time traffic volume, the auxiliary management unit is designated as the real-time traffic volume mitigation unit, and the real-time demand supply is determined by the real-time traffic volume mitigation unit to participate in the bidding for the increase demand of the real-time traffic volume enhancement unit.

[0031] The system matches the real-time increase demand of the real-time traffic volume increase side with the real-time demand supply of all the real-time traffic volume decrease side, and performs real-time traffic volume transaction updates on the real-time traffic volume of the real-time traffic volume decrease side and the real-time traffic volume increase side corresponding to the real-time demand supply that best matches the real-time increase demand, so as to achieve real-time balance of traffic volume of each traffic line. The real-time increase demand is a real-time description of the traffic volume to be adjusted in the real-time traffic volume increase side.

[0032] As a preferred embodiment of the present invention, the step of determining the real-time increase demand by the real-time traffic volume increase method includes:

[0033] The difference between the minimum real-time throughput and the real-time throughput at the real-time timestamp is taken as the minimum demand.

[0034] The difference between the expected real-time traffic volume and the real-time traffic volume at the real-time timestamp is taken as the maximum limit of demand.

[0035] The maximum and minimum demand values ​​are combined to form the real-time increase demand.

[0036] As a preferred embodiment of the present invention, the step of determining real-time demand supply by the real-time traffic volume mitigation method includes:

[0037] The difference between the real-time traffic volume at the real-time timestamp and the minimum real-time traffic volume is taken as the maximum supply limit, and then the maximum supply limit is taken as the real-time demand supply.

[0038] As a preferred embodiment of the present invention, the real-time increase demand of the real-time traffic volume increase party is matched with the real-time demand supply of all real-time traffic volume decrease parties, including:

[0039] Compare the maximum supply limit of real-time demand supply in each real-time traffic volume mitigation side with the maximum demand limit of real-time demand resources in each real-time traffic volume enhancement side, and take the real-time traffic volume mitigation side corresponding to the maximum difference as the best matching side of real-time traffic volume enhancement side.

[0040] The minimum demand value is used as the adjusted traffic volume. In the real-time traffic volume boosting side, vehicles waiting to enter the real-time traffic volume boosting side are selected according to the adjusted traffic volume and guided to the real-time traffic volume mitigation side, which is the best matching side. The real-time traffic volume in the real-time traffic volume boosting side and the real-time traffic volume mitigation side is balanced in real time in a decentralized manner to reduce the time spent on travel.

[0041] In a preferred embodiment of the present invention, the real-time traffic volume is output by a pre-established real-time traffic volume prediction model, wherein the establishment of the real-time traffic volume prediction model includes:

[0042] The system monitors the real-time traffic volume of each passageway in the coordinated management area and records a set of real-time traffic volume sequences. The real-time traffic volume sequences are then fed into an LSTM neural network for network training to obtain a real-time traffic volume prediction model that represents the mapping relationship between time sequence timestamps and real-time traffic volume.

[0043] As a preferred embodiment of the present invention, the present invention provides a traffic management system based on the adaptive traffic management method based on big data analysis, comprising:

[0044] The auxiliary management unit uses a traffic volume experience model to predict the maximum traffic volume of each traffic route in the auxiliary management area, and selects auxiliary management routes by using the maximum traffic volume as a threshold. Then, it selects replacement routes for each auxiliary management route from all traffic routes, and combines the auxiliary management route and all its replacement routes into an auxiliary management unit.

[0045] The traffic volume statistics unit takes the assisted lines and the real-time traffic volume on the assisted lines in each assisted management system as a single assisted management unit, and also takes the replacement lines and the real-time traffic volume on the replacement lines as a single assisted management unit, and counts the real-time traffic volume of each assisted management unit.

[0046] The adaptive scheduling unit sets the expected real-time traffic volume and the maximum communication rate. Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, it performs real-time command and scheduling of vehicles in each auxiliary management unit within each auxiliary management system to achieve adaptive balance of traffic pressure among the various auxiliary management units.

[0047] Compared with the prior art, the present invention has the following advantages:

[0048] This invention utilizes a traffic volume experience model to select auxiliary routes within the auxiliary management area. The auxiliary routes and all their alternative routes are combined into an auxiliary management system. A real-time traffic volume expectation value and a maximum communication rate are set. Based on the real-time traffic volume expectation value, the real-time traffic volume minimum value, and the real-time traffic volume, the invention performs real-time command and dispatch of vehicles in each auxiliary management unit within each auxiliary management system to achieve adaptive balance of traffic pressure in each auxiliary management unit. Attached Figure Description

[0049] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0050] Figure 1 This is a flowchart of the adaptive traffic management method provided in an embodiment of the present invention;

[0051] Figure 2 This is a block diagram of the co-management system provided in an embodiment of the present invention.

[0052] The labels in the diagram represent the following:

[0053] 1-Assisted management unit; 2-Traffic volume statistics unit; 3-Adaptive scheduling unit. Detailed Implementation

[0054] 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.

[0055] like Figure 1 As shown, this invention provides an adaptive traffic management method based on big data analysis, comprising the following steps:

[0056] Step S1: Predict the maximum traffic volume of each route in the auxiliary management area using the traffic volume experience model, and select auxiliary management routes by thresholding the routes based on the maximum traffic volume. Then, select replacement routes for each auxiliary management route from all routes. Combine the auxiliary management route and all its replacement routes into an auxiliary management system. The replacement routes have the same traffic effect as the auxiliary management routes, that is, the replacement routes can meet the travel expectations of travelers just like the auxiliary management routes.

[0057] Route features include route length, route width, route connecting areas, and route road surface layout. Regional features include regional resident population characteristics, regional transportation status characteristics, regional travel population characteristics, and regional public institution distribution characteristics. Features can be added or deleted in actual use.

[0058] The maximum traffic volume for each traffic route in the coordinated management area is predicted using an empirical traffic volume model, including:

[0059] The regional features of the assisted management area and the route features of the traffic routes are extracted sequentially, and the regional features and route features are output to the traffic volume empirical model. The traffic volume empirical model outputs the maximum traffic volume of the traffic routes.

[0060] The construction of the traffic volume empirical model includes:

[0061] Multiple traffic routes were selected as sample routes in different areas, and the real-time traffic volume of the sample routes was monitored. The real-time traffic volume of the sample routes was averaged to obtain the average real-time traffic volume as the maximum traffic volume of the sample routes.

[0062] Extract the line features of the sample line and the regional features of the area to which the sample line belongs. Use the line features and regional features as input terms of the BP neural network, and use the maximum traffic volume of the sample line as the output term of the BP neural network.

[0063] By using a BP neural network to perform convolution training on the input and output terms, an empirical traffic volume model representing the mapping relationship between line features, regional features and maximum traffic volume is obtained.

[0064] The functional expression of the traffic volume empirical model is:

[0065] T = BP(S, P);

[0066] In the formula, T is the character representing the maximum traffic volume, S is the character representing the line characteristics, P is the character representing the area characteristics, and BP is the character representing the BP neural network.

[0067] The routes to be managed by the co-management system are selected by threshold filtering based on the maximum traffic volume, including:

[0068] The maximum throughput of each traffic route is compared with the throughput threshold.

[0069] When the maximum traffic volume of a passage is greater than or equal to the traffic volume threshold, the passage will be marked as a regulated passage.

[0070] If the maximum traffic volume of a passage is less than the traffic volume threshold, the passage will be marked as a non-managed passage.

[0071] By using a traffic volume experience model to predict the traffic volume of passing vehicles, it is possible to identify the routes in the auxiliary management area that require auxiliary management intervention. If the real-time traffic volume of a route is higher, the route is more prone to traffic congestion and requires auxiliary management intervention to ensure smooth traffic. Therefore, routes with real-time traffic volume greater than or equal to the traffic volume threshold are designated as auxiliary management routes, that is, routes prone to congestion are designated as auxiliary management routes for auxiliary management intervention.

[0072] The managed lines and all their replacement lines are combined into a single managed unit, including:

[0073] The assisted management lines and all their replacement lines are sequentially grouped into a single assisted management unit, and the inclusion relationships between all assisted management units are compared.

[0074] If there is an inclusion relationship between any two co-managed groups, then the co-managed group that is included in the two co-managed groups will be removed.

[0075] If there is no inclusion relationship between any two co-managed entities, then both co-managed entities will be retained.

[0076] Step S2: Take the assisted lines and the real-time traffic volume on each assisted line as a single assisted line unit, and also take the replacement lines and the real-time traffic volume on the replacement lines as a single assisted line unit, and count the real-time traffic volume of each assisted line unit.

[0077] Step S3: Set the expected real-time traffic volume and the maximum communication rate. Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, conduct real-time command and dispatch of vehicles in each auxiliary management unit in each auxiliary management system to achieve adaptive balance of traffic pressure in each auxiliary management unit.

[0078] Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, real-time command and dispatch of vehicles is carried out for each auxiliary management unit in each auxiliary management system, including:

[0079] The real-time coordination role of a coordination unit is determined based on its real-time traffic volume. This role includes both a real-time traffic volume increaser and a real-time traffic volume decreaser.

[0080] When the real-time throughput of the co-management unit is less than the minimum real-time throughput limit, the co-management unit is designated as the real-time throughput booster, and the real-time throughput booster determines the real-time boosting requirement for the real-time throughput mitigation unit to perform co-management monitoring.

[0081] When the real-time traffic volume of the auxiliary management unit is greater than or equal to the minimum real-time traffic volume limit, the auxiliary management unit is regarded as the real-time traffic volume mitigation party, and the real-time demand supply is determined by the real-time traffic volume mitigation party to participate in the bidding for the increase demand of the real-time traffic volume increase party.

[0082] The system matches the real-time increase demand of the real-time traffic volume increase side with the real-time demand supply of all the real-time traffic volume decrease side, and performs real-time traffic volume transaction updates on the real-time traffic volume decrease side and the real-time traffic volume increase side corresponding to the real-time demand supply that best matches the real-time increase demand, so as to achieve real-time balance of traffic volume of each traffic line. The real-time increase demand is a real-time description of the traffic volume to be adjusted in the real-time traffic volume increase side.

[0083] The real-time traffic volume improvement provider determines the real-time improvement requirements, including:

[0084] The difference between the minimum real-time throughput and the real-time throughput at the real-time timestamp is taken as the minimum demand.

[0085] The difference between the expected real-time traffic volume and the real-time traffic volume at the real-time timestamp is taken as the maximum limit of demand.

[0086] The maximum and minimum demand values ​​are combined to form the real-time demand increase.

[0087] Real-time demand and supply are determined by the real-time traffic volume reduction factor, including:

[0088] The difference between the real-time traffic volume at the real-time timestamp and the minimum real-time traffic volume is taken as the maximum supply limit, and then the maximum supply limit is taken as the real-time demand supply.

[0089] The real-time increase demand of the traffic volume increase side is matched with the real-time demand supply of all the traffic volume decrease sides, including:

[0090] Compare the maximum supply limit of real-time demand supply in each real-time traffic volume mitigation side with the maximum demand limit of real-time demand resources in each real-time traffic volume enhancement side, and take the real-time traffic volume mitigation side corresponding to the maximum difference as the best matching side of real-time traffic volume enhancement side.

[0091] The minimum demand value is used as the adjusted traffic volume. In the real-time traffic volume boosting side, vehicles waiting to enter the real-time traffic volume boosting side are selected according to the adjusted traffic volume and guided to the real-time traffic volume mitigation side, which is the best matching side. The real-time traffic volume in the real-time traffic volume boosting side and the real-time traffic volume mitigation side is balanced in real time in a decentralized manner to reduce the time spent on travel.

[0092] The decentralized real-time traffic volume management intervention provided in this embodiment quantifies the real-time traffic pressure of a traffic line using its real-time traffic volume. Based on the real-time traffic pressure of multiple traffic lines within a single management system, it guides and transfers passing vehicles. Traffic lines with high real-time traffic pressure experience pressure relief; that is, vehicles passing through a high-volume traffic line are allocated to a low-volume traffic line with the same traffic effect. This balances the real-time traffic volume across multiple traffic lines with the same effect, and avoids the problem of traffic centralization, preventing centralized phenomena that cause congestion on a particular traffic line due to vehicle clustering, thus achieving traffic coordination management.

[0093] Real-time traffic volume is output by a pre-established real-time traffic volume prediction model. The establishment of the real-time traffic volume prediction model includes:

[0094] The system monitors the real-time traffic volume of each route in the coordinated management area and records a set of real-time traffic volume sequences. These sequences are then fed into an LSTM neural network for training to obtain a real-time traffic volume prediction model that represents the mapping relationship between time-series timestamps and real-time traffic volume. This allows for advance prediction and pre-emptive vehicle scheduling, ensuring smooth operation of the routes under the intervention of the coordinated management system.

[0095] like Figure 2 As shown, based on the above adaptive traffic management method, this invention provides a traffic management system, including:

[0096] The auxiliary management unit 1 uses a traffic volume experience model to predict the maximum traffic volume of each traffic route in the auxiliary management area, and selects auxiliary management routes by using the maximum traffic volume as a threshold. Then, it selects replacement routes for each auxiliary management route from all traffic routes, and combines the auxiliary management route and all its replacement routes into an auxiliary management whole.

[0097] Traffic volume statistics unit 2 takes the traffic flow of the auxiliary lines and the real-time traffic flow on the auxiliary lines in each auxiliary management system as a single auxiliary management unit, and also takes the replacement lines and the real-time traffic flow on the replacement lines as a single auxiliary management unit, and counts the real-time traffic flow of each auxiliary management unit.

[0098] The adaptive scheduling unit 3 sets the expected real-time traffic volume and the maximum communication rate. Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, it performs real-time command and scheduling of vehicles in each auxiliary management unit in each auxiliary management system to achieve adaptive balance of traffic pressure in each auxiliary management unit.

[0099] This invention utilizes a traffic volume experience model to select auxiliary routes within the auxiliary management area. The auxiliary routes and all their alternative routes are combined into an auxiliary management system. A real-time traffic volume expectation value and a maximum communication rate are set. Based on the real-time traffic volume expectation value, the real-time traffic volume minimum value, and the real-time traffic volume, the invention performs real-time command and dispatch of vehicles in each auxiliary management unit within each auxiliary management system to achieve adaptive balance of traffic pressure in each auxiliary management unit.

[0100] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. An adaptive traffic management method based on big data analysis, characterized in that, Includes the following steps: Step S1: Predict the maximum traffic volume of each traffic route in the auxiliary management area using the traffic volume experience model, and select auxiliary management routes by thresholding the traffic routes using the maximum traffic volume. Then, select replacement routes for each auxiliary management route from all traffic routes, and combine the auxiliary management route and all its replacement routes into an auxiliary management system. The replacement routes have the same traffic effect as the auxiliary management routes. Step S2: Take the assisted lines and the real-time traffic volume on each assisted line as a single assisted line unit, and also take the replacement lines and the real-time traffic volume on the replacement lines as a single assisted line unit, and count the real-time traffic volume of each assisted line unit. Step S3: Set the expected value of real-time traffic volume and the maximum value of communication rate. Based on the expected value of real-time traffic volume, the minimum value of real-time traffic volume, and the real-time traffic volume, conduct real-time command and dispatch of traffic vehicles in each auxiliary management unit in each auxiliary management system to achieve adaptive balance of traffic pressure in each auxiliary management unit. Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, real-time command and dispatch of vehicles is carried out for each auxiliary management unit in each auxiliary management system, including: The real-time coordination role of a coordination unit is determined based on its real-time traffic volume. This role includes both a real-time traffic volume increaser and a real-time traffic volume decreaser. When the real-time throughput of the co-management unit is less than the minimum real-time throughput limit, the co-management unit is designated as the real-time throughput booster, and the real-time throughput booster determines the real-time boosting requirement for the real-time throughput mitigation unit to perform co-management monitoring. When the real-time traffic volume of the auxiliary management unit is greater than or equal to the minimum real-time traffic volume limit, the auxiliary management unit is regarded as the real-time traffic volume mitigation party, and the real-time demand supply is determined by the real-time traffic volume mitigation party to participate in the bidding for the increase demand of the real-time traffic volume increase party. The system matches the real-time increase demand of the real-time traffic volume increase side with the real-time demand supply of all the real-time traffic volume decrease side, and performs real-time traffic volume transaction updates on the real-time traffic volume decrease side and the real-time traffic volume increase side corresponding to the real-time demand supply that best matches the real-time increase demand, so as to achieve real-time balance of traffic volume of each traffic line. The real-time increase demand is a real-time description of the traffic volume to be adjusted in the real-time traffic volume increase side. The real-time traffic volume improvement provider determines the real-time improvement requirements, including: The difference between the minimum real-time throughput and the real-time throughput at the real-time timestamp is taken as the minimum demand. The difference between the expected real-time traffic volume and the real-time traffic volume at the real-time timestamp is taken as the maximum limit of demand. The maximum and minimum demand values ​​are combined to form real-time demand increases; The real-time increase demand of the traffic volume increase side is matched with the real-time demand supply of all the traffic volume decrease sides, including: Compare the maximum supply limit of real-time demand supply in each real-time traffic volume mitigation side with the maximum demand limit of real-time demand resources in each real-time traffic volume enhancement side, and take the real-time traffic volume mitigation side corresponding to the maximum difference as the best matching side of real-time traffic volume enhancement side. The minimum demand value is used as the adjusted traffic volume. In the real-time traffic volume boosting side, vehicles waiting to enter the real-time traffic volume boosting side are selected according to the adjusted traffic volume and guided to the real-time traffic volume mitigation side, which is the best matching side. The real-time traffic volume in the real-time traffic volume boosting side and the real-time traffic volume mitigation side is balanced in real time in a decentralized manner to reduce the time spent on travel.

2. The adaptive traffic management method based on big data analysis according to claim 1, characterized in that: The method of predicting the maximum traffic volume for each traffic route in the coordinated management area using an empirical traffic volume model includes: The regional features of the assisted management area and the route features of the traffic route are extracted sequentially, and the regional features and the route features are output to the traffic volume empirical model. The traffic volume empirical model outputs the maximum traffic volume of the traffic route. The construction of the traffic volume empirical model includes: Multiple traffic routes were selected as sample routes in different areas, and the real-time traffic volume of the sample routes was monitored. The real-time traffic volume of the sample routes was averaged to obtain the average real-time traffic volume as the maximum traffic volume of the sample routes. Extract the line features of the sample line and the regional features of the area to which the sample line belongs. Use the line features and regional features as input terms of the BP neural network, and use the maximum traffic volume of the sample line as the output term of the BP neural network. By using a BP neural network to perform convolution training on the input and output terms, an empirical traffic volume model representing the mapping relationship between line features, regional features and maximum traffic volume is obtained. The functional expression of the traffic volume empirical model is: T=BP(S,P); In the formula, T is the character representing the maximum traffic volume, S is the character representing the line characteristics, P is the character representing the area characteristics, and BP is the character representing the BP neural network.

3. The adaptive traffic management method based on big data analysis according to claim 2, characterized in that: The step of selecting auxiliary traffic routes by using the maximum traffic volume threshold filtering method includes: The maximum throughput of each traffic route is compared with the throughput threshold. When the maximum traffic volume of a passage is greater than or equal to the traffic volume threshold, the passage will be marked as a regulated passage. If the maximum traffic volume of a passage is less than the traffic volume threshold, the passage will be marked as a non-managed passage.

4. The adaptive traffic management method based on big data analysis according to claim 3, characterized in that: The process of forming a single managed system, comprising the managed lines and all their replacement lines, includes: The assisted management lines and all their replacement lines are sequentially grouped into a single assisted management unit, and the inclusion relationships between all assisted management units are compared. If there is an inclusion relationship between any two co-managed groups, then the co-managed group that is included in the two co-managed groups will be removed. If there is no inclusion relationship between any two co-managed entities, then both co-managed entities will be retained.

5. The adaptive traffic management method based on big data analysis according to claim 4, characterized in that, The determination of real-time demand supply by the real-time traffic volume mitigation method includes: The difference between the real-time traffic volume at the real-time timestamp and the minimum real-time traffic volume is taken as the maximum supply limit, and then the maximum supply limit is taken as the real-time demand supply.

6. The adaptive traffic management method based on big data analysis according to claim 5, characterized in that, The real-time traffic volume is output by a pre-established real-time traffic volume prediction model, and the establishment of the real-time traffic volume prediction model includes: The system monitors the real-time traffic volume of each passageway in the coordinated management area and records a set of real-time traffic volume sequences. The real-time traffic volume sequences are then fed into an LSTM neural network for network training to obtain a real-time traffic volume prediction model that represents the mapping relationship between time sequence timestamps and real-time traffic volume.

7. A traffic management system based on big data analysis and an adaptive traffic management method according to any one of claims 1-6, characterized in that, include: The auxiliary management unit uses a traffic volume experience model to predict the maximum traffic volume of each traffic route in the auxiliary management area, and selects auxiliary management routes by using the maximum traffic volume as a threshold. Then, it selects replacement routes for each auxiliary management route from all traffic routes, and combines the auxiliary management route and all its replacement routes into an auxiliary management unit. The traffic volume statistics unit takes the assisted lines and the real-time traffic volume on the assisted lines in each assisted management system as a single assisted management unit, and also takes the replacement lines and the real-time traffic volume on the replacement lines as a single assisted management unit, and counts the real-time traffic volume of each assisted management unit. The adaptive scheduling unit sets the expected real-time traffic volume and the maximum communication rate. Based on the expected real-time traffic volume, the minimum real-time traffic volume, and the real-time traffic volume, it performs real-time command and scheduling of vehicles in each auxiliary management unit within each auxiliary management system to achieve adaptive balance of traffic pressure among the various auxiliary management units.

Images