A method and system for estimating traffic status of large-scale highway network

Abstract

The present invention provides a method and system for estimating the traffic state of a large-scale highway road network, which relates to the technical field of traffic data analysis, including: obtaining an initial data set, the initial data set including the ETC gantry pair codes, pile numbers and distance information on each line of the entire road network; performing decomposition processing according to the initial data set to obtain a cellular division result; performing road network construction processing according to the cellular division result to obtain a static road network structure; performing real-time flow calculation according to the static road network structure to obtain an updated cellular state; generating the traffic situation of each type of interchange in the entire road network according to the updated cellular state and the static road network structure, and obtaining a hundred-meter-level traffic state estimation result of the entire road network. The present invention combines ETC gantry section data with video monitoring points and toll station data through multi-source data fusion, effectively making up for the estimation accuracy problem caused by sparse data.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic data analysis, and in particular to a method and system for estimating traffic status of a large-scale highway network. Background Art

[0002] With the continuous development of national infrastructure and the rapid expansion of the highway network, real-time understanding of the traffic status of the entire road network has become the key to traffic management and planning. At present, ETC gantries are widely used on highways to collect mainline flow data. In addition, toll stations, video surveillance points and other facilities are also used to monitor traffic conditions. However, existing traffic state estimation methods still have many shortcomings. Although the traditional situation analysis method based on vehicle trajectory data has high accuracy, it is difficult to obtain data and has limited coverage, which cannot meet the real-time monitoring needs of large-scale road networks. On the other hand, although the model-driven method based on sparse cross-sectional data has good interpretability and portability, it is easy to cause the estimation accuracy to decrease due to the sparse data source, especially when it comes to complex interchange structures (such as cross-shaped, T-shaped and toll station interchanges). Existing technologies are difficult to differentiate different interchange types and cannot accurately describe the flow dynamics between the main road and the ramp. In addition, most of the current traffic state estimation systems are difficult to expand to the entire highway network, and lack situation estimation solutions that can operate efficiently in large-scale, multi-type road network environments.

[0003] In response to the above problems, the present invention proposes a method and system for estimating the traffic status of a large-scale highway network based on the fusion of sparse cross-sectional data and multi-source data. The method utilizes the sparse cross-sectional data collected by the ETC gantry, combines multiple data sources such as toll stations and video surveillance points, and realizes the hundred-meter-level traffic status estimation of the entire highway network based on the CTM model. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for estimating the traffic state of a large-scale highway network to improve the above-mentioned problem. In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a method for estimating traffic status of a large-scale highway network, comprising:

[0006] Acquire an initial data set, wherein the initial data set includes ETC gantry pair codes, pile numbers, and distance information on each line of the entire road network;

[0007] Decomposing the initial data set, dividing the road section into cells using the upstream and downstream gantry information of the same route and the same direction, and optimizing the cell length and number to obtain a cell division result, wherein the cell division result includes main road cell information, ramp cell information, and corresponding connection information;

[0008] A road network construction process is performed according to the cell division result, by merging cells, gantry pairs and interchange structures step by step, connecting main road cells and ramp cells, and constructing a virtual gantry at a T-shaped interchange to obtain a static road network structure;

[0009] According to the static road network structure, combined with the real-time collected data of the diversion ratio and confluence ratio of the main road cell and the ramp cell, the real-time flow calculation is performed to obtain the flow distribution ratio of the main road and the ramp in the entire road network, and the real-time flow and density data of the cell are calculated according to the collected average speed of the vehicle in the cell to obtain the updated cell state;

[0010] According to the updated cellular state and the static road network structure, differentiated estimates are made for different interchange types, and state estimation is performed using the flow data of the gantry pairs and the speed data of the video points. The cellular flow and density are deduced through a preset CTM model to generate the traffic situation for each interchange type in the entire road network, and obtain the 100-meter-level traffic state estimation result for the entire road network.

[0011] In a second aspect, the present application also provides a highway large-scale road network traffic state estimation system, comprising:

[0012] An acquisition module is used to acquire an initial data set, wherein the initial data set includes ETC gantry pair codes, pile numbers and distance information on each line of the entire road network;

[0013] A decomposition module is used to perform decomposition processing according to the initial data set, divide the road section into cells using the upstream and downstream gantry information of the same route and the same direction, and optimize the cell length and number to obtain a cell division result, wherein the cell division result includes main road cell information, ramp cell information and corresponding connection information;

[0014] A construction module is used to perform road network construction processing according to the cell division result, connect the main road cells with the ramp cells by merging cells, gantry pairs and interchange structures step by step, and construct a virtual gantry at the T-shaped interchange to obtain a static road network structure;

[0015] An updating module is used to calculate the real-time flow rate according to the static road network structure and the real-time collected data of the flow distribution ratio and confluence ratio of the main road cell and the ramp cell to obtain the flow distribution ratio of the main road and the ramp in the whole road network, and calculate the real-time flow rate and density data of the cell according to the collected average speed of the vehicle in the cell to obtain the updated cell state;

[0016] The estimation module is used to perform differentiated estimation on different interchange types according to the updated cellular state and the static road network structure, perform state estimation using the flow data of the gantry pair and the speed data of the video points, and deduce the cellular flow and density through a preset CTM model to generate the traffic situation of each interchange type in the entire road network, and obtain the 100-meter-level traffic state estimation result of the entire road network.

[0017] The beneficial effects of the present invention are:

[0018] The present invention combines ETC gantry cross-section data with video surveillance points and toll station data through multi-source data fusion, effectively compensating for the estimation accuracy problem caused by sparse data. On the basis of the CTM model, the multi-source data is used to optimize the estimation process of flow and density, significantly improving the accuracy of the state estimation of the entire highway network. The present invention designs differentiated flow estimation methods for different types of interchanges, such as cross-type interchanges, T-type interchanges, and toll station interchanges, to ensure the accuracy of flow transfer between the main road and the ramp.

[0019] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A schematic diagram of a flow chart of a method for estimating a large-scale highway network traffic state according to an embodiment of the present invention;

[0022] Figure 2 A structural diagram of a highway large-scale road network traffic state estimation system according to an embodiment of the present invention;

[0023] Figure 3 Flowchart for cell partitioning.

[0024] Markings in the figure: 1. acquisition module; 2. decomposition module; 3. construction module; 4. update module; 5. estimation module; 21. first division unit; 22. first calculation unit; 23. first detection unit; 24. second division unit; 31. first processing unit; 32. first construction unit; 33. first generation unit; 34. second generation unit; 41. first acquisition unit; 42. first allocation unit; 43. second calculation unit; 44. first update unit; 51. first recognition unit; 52. first estimation unit; 53. second update unit; 54. third calculation unit; 541. fourth calculation unit; 542. fifth calculation unit; 543. sixth calculation unit; 544. seventh calculation unit. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0027] Embodiment 1:

[0028] This embodiment provides a method for estimating traffic status of a large-scale highway network.

[0029] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400 and step S500.

[0030] Step S100, obtaining an initial data set, the initial data set including ETC gantry pair codes, pile numbers and distance information on each line of the entire road network;

[0031] It is understood that an ETC gantry pair is a group of gantries arranged on a highway, usually with one located upstream of a section and the other located downstream, forming a "pair" for data collection. Each pair of gantry has its own unique code to identify their specific location in the entire road network. The code usually includes the route number, direction and gantry position number to ensure that the system can uniquely identify each gantry pair. The stake number is a marking system used to identify specific locations on highways, usually set at a certain distance interval, such as marking a stake number every 1 kilometer or 100 meters. The stake number information of the ETC gantry can accurately indicate the geographical location of the gantry and help the system accurately locate the specific location of each pair of gantry. The distance information refers to the physical distance between two ETC gantries, that is, the interval between the upstream and downstream gantries of the gantry pair. This distance determines the coverage of data collection and the accuracy of the system in monitoring traffic flow on different sections.

[0032] Step S200: Decompose the initial data set, divide the road section into cells using the upstream and downstream gantry information of the same route and direction, and optimize the cell length and number to obtain the cell division result, which includes the main road cell information, ramp cell information and corresponding connection information;

[0033] It should be noted that the cell (Cell Transmission Model, CTM) is a basic unit based on the CTM model, which is used to represent a small section of the road. Cell division is to divide the road section into several small units, and each unit is regarded as an independent traffic state estimation area. Specifically, the length of each road section is first determined based on the pile number and distance information of the upstream and downstream gantries. Then, the length and number of cells are optimized in combination with the road characteristics and the accuracy requirements of traffic estimation to ensure that the length of each cell is kept consistent as much as possible while meeting the minimum division requirements. The results of cell division include main road cell information and ramp cell information. The main road cell is mainly used to represent the traffic status on the main highway, while the ramp cell is used to represent the traffic status of the ramp area. The connection information is the connection relationship between the main road cell and the ramp cell, which describes how the traffic flow is distributed and transferred between the main road and the ramp.

[0034] Step S300, performing road network construction processing according to the cell division result, by merging cells, gantry pairs and interchange structures step by step, connecting main road cells and ramp cells, and constructing virtual gantry at T-shaped interchanges, to obtain a static road network structure;

[0035] Specifically, this step connects the main road cells and ramp cells according to the actual topological structure of the road by merging each cell level by level. The cells of each section are connected in series through the upstream and downstream ETC gantry pair information to form a road network unit at the gantry pair level. Subsequently, each gantry pair is further combined to construct an interchange structure to ensure that the connection relationship between the main road cells and the ramp cells is accurately reflected. For conventional cross interchanges, ramp interchanges and other structures, the system can directly use gantry pairs to connect the main road and the ramp. However, in the T-shaped interchange structure, since some sections are not covered by actual physical gantries, virtual gantries need to be introduced. Virtual gantries are used to simulate the traffic flow of these special sections, thereby ensuring that these uncovered sections can also participate in the overall structure of the road network. Through this step, a complete static road network structure including the main road, ramps, interchange structures and virtual gantries is finally generated, providing an accurate network foundation for subsequent traffic state estimation.

[0036] Step S400: According to the static road network structure, combined with the real-time collected data of the flow splitting ratio and the flow confluence ratio of the main road cell and the ramp cell, the real-time flow calculation is performed to obtain the flow distribution ratio of the main road and the ramp in the whole road network, and the real-time flow and density data of the cell are calculated according to the collected average speed of the vehicle in the cell to obtain the updated cell state;

[0037] It can be understood that this step obtains key parameters such as the diversion ratio (the proportion of main road vehicles flowing to the ramp) and the confluence ratio (the proportion of ramp vehicles merging into the main road) of each cell through real-time acquisition equipment. These data reflect the real-time changes in traffic flow between the main road and the ramp. By using these ratio data, combined with the connection information of the main road and ramp cells in the static road network, the real-time flow distribution of each cell is calculated to determine the flow distribution ratio between the main road and the ramp. Then, the average speed information of the vehicles in each cell is obtained through further real-time data collection, and the speed and flow data are combined. The real-time traffic density in the cell (that is, the number of vehicles per unit distance) is calculated according to the traffic flow theory formula. These density and flow data jointly define the traffic state of each cell at the current time. Finally, the state of the cell is updated according to the calculation results, and the updated state data containing dynamic parameters such as the real-time flow, density and diversion and confluence ratio of each cell is generated.

[0038] Step S500: Differentiate the different interchange types according to the updated cellular state and static road network structure, use the flow data of the gantry pair and the speed data of the video point to estimate the state, and use the preset CTM model to deduce the cellular flow and density to generate the traffic situation of each interchange type in the entire road network, and obtain the 100-meter-level traffic state estimation result of the entire road network.

[0039] Specifically, this step first uses the gantry to collect traffic data and vehicle speed data at video points, combined with the characteristics of each interchange structure to make accurate state estimation. For cross-type interchanges, traffic estimation needs to deal with the complex traffic flow relationship of multiple divergences and confluences; in T-type interchanges, the system simulates the traffic status of uncovered sections by introducing virtual gantries, and estimates the flow transfer between ramps and main roads; in toll station interchanges, the flow input and output between the main road and the toll station are specially processed. Then, based on the preset CTM model, the flow and density data of the cells are deduced, and the flow and density changes during the divergence and confluence process are dynamically calculated by simulating the flow transfer between the main road and ramp cells, and the real-time traffic situation estimation results of various types of interchanges are generated. Through this differentiated processing, the 100-meter-level traffic situation estimation of each type of interchange in the entire road network can be achieved, providing managers with refined road network traffic conditions and improving the estimation accuracy in complex road network structures.

[0040] Further, step S200 includes step S210 to step S240.

[0041] Step S210: performing division processing according to the ramp structure information in the initial data set, converting each ramp into a corresponding ramp cell, and forming a data document to obtain ramp cell information;

[0042] Step S220, calculating the optimal number and length of cell divisions according to the gantry-to-road section length data in the initial data set, and obtaining the optimal division scheme;

[0043] Step S230, performing consistency detection based on the upstream and downstream gantry pile number information in the initial data set, and checking whether the upstream and downstream pile numbers are consistent by comparison. If there is inconsistency, performing pile number conversion processing based on a preset conversion rule to obtain converted pile number information;

[0044] Step S240: Perform main road cell division processing according to the converted pile number information to obtain main road cell information, and integrate the main road cell with the ramp cell, match their respective connection relationships, and obtain the connection information of the main road and the ramp.

[0045] It should be noted that in the early stage of work, it is necessary to collect the basic structural information such as the coding, pile number and distance (D) of the upstream and downstream gantry pairs on each line of the entire road network, and organize this information into a file of a certain format. This information is the basic framework for cellular division. After importing the above data information into the program, the road section is divided into cells based on the upstream and downstream gantry information of the same route and direction, and the cellular division of the entire road network is realized by traversing all the gantry pair information on the road network. The specific workflow is as follows: Figure 3 As shown:

[0046] The first step is to integrate the sorted ramp structure information, convert each ramp into a corresponding ramp cell, and output it as a data document.

[0047] The second step is to calculate the optimal number of cell divisions and cell length corresponding to the gantry on the road section. According to the estimation requirements, the consistency of the cell lengths divided on the road section should be ensured, and the cell length should be greater than the lower limit and as small as possible from the empirical upper limit. The lower limit is determined by the free flow speed and the simulation step size, and the relevant formula is as follows:

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] in, represents the optimal number of cell divisions; represents the cell length; Indicates the distance between the gantry and the road section; Indicates the lower limit of cell length; represents the simulation step length; represents the free flow speed; Represents the empirical upper limit of the cell length.

[0053] Preferably, in this embodiment, enumeration is used to find integers from 1 to 1000 that can divide the length of the road section, and the divisible numbers are used as candidate values ​​for the number of cell divisions. After obtaining the candidate values ​​for the number of cell divisions, the cell lengths corresponding to the candidate value solutions can be calculated by dividing them by the road section distance.

[0054] ;

[0055] ;

[0056] in, Indicates the cell length corresponding to the candidate value of the number of cell partitions; Represents the candidate value of the number of cell partitions; Represents a set of candidate values.

[0057] Next, the partitioning schemes that meet the upper and lower limits of the cell length are screened out. If there is no suitable candidate value between the upper and lower limits, the lower limit is cancelled. Finally, the maximum number of cell partitions that meet the conditions is taken as the optimal number of cell partitions.

[0058] The third step is to check whether the pile numbers of the upstream and downstream gantries are in the same pile number coding framework. In the actual road network system, two connected sections may have different pile number coding methods for the same route and direction due to various problems. Therefore, before subsequent operations, it is necessary to check whether the pile numbers of the upstream and downstream gantries are in the same pile number coding framework. If they are in the same pile number coding framework, no further processing is required. If they are not in the same pile number coding framework, corresponding processing is required to convert the pile numbers so that the pile numbers of the upstream and downstream gantries are in the same coding framework. In the actual operation process, when the pile numbers of the upstream and downstream gantries are not in the same pile number coding framework, there are four corresponding situations. In the first case, the upstream gantry increases along the road direction, and the downstream gantry decreases along the road direction. In the second case, the upstream gantry increases along the road direction, and the downstream gantry increases along the road direction. Corresponding to these two situations, the processing method is: the pile number of the upstream gantry remains unchanged, and the pile number of the downstream gantry is converted to the sum of the pile number of the upstream gantry and the distance between the two gantry. The specific expression is as follows:

[0059] ;

[0060] in, Indicates the upstream gantry pile number; Indicates the downstream gantry pile number.

[0061] In the third case, the upstream gantry decreases along the road direction, and the downstream gantry decreases along the road direction. The processing method is: the pile number of the downstream gantry remains unchanged, and the pile number of the upstream gantry is converted to the sum of the pile number of the downstream gantry and the distance between the two gantries. The specific expression is as follows:

[0062] ;

[0063] In the fourth case, the upstream gantry decreases along the road direction, and the downstream gantry increases along the road direction. The processing method is: the pile number of the downstream gantry becomes 0, and the pile number of the upstream gantry is converted to the sum of the distances between the two gantry. The specific expression is as follows:

[0064] ;

[0065] In addition, if there is a junction where the main road and the ramp meet on this road section, the pile number also needs to be converted so that it is generally in the same pile number coding framework as the upstream and downstream gantries.

[0066] In the fourth step, the starting and ending pile numbers of each cell are calculated by combining the optimal cell length in the second step and the pile number processing and conversion work in the third step. After obtaining the starting and ending pile numbers of each main road cell, the main road cell and the ramp cell connected to it can be matched, and the corresponding connection information can be stored in the attributes of the main road cell to facilitate the subsequent call of the road network initialization module.

[0067] Step 5: Check whether the gantry pile number has been converted in step 2. If the conversion has been implemented, the model will perform a reverse operation to restore the cellular pile number to the real pile number to ensure the accuracy of the front-end road network rendering.

[0068] In the sixth step, the module automatically processes and stores all cellular information, organizes it into a data document for output, and provides detailed cellular pile number information and ramp connection information for subsequent estimation work.

[0069] Further, step S300 includes step S310 to step S340.

[0070] Step S310: performing connection processing of the main road and ramp cells according to the cell division result, by analyzing the upstream and downstream position relationship between the main road and the ramp, matching adjacent main road cells and ramp cells one by one, and obtaining preliminary connection information;

[0071] Step S320: Perform road network construction processing at the gantry pair level according to the preliminary connection information, by classifying and grouping the connected cells to form gantry pairs, and obtaining gantry pair information including main road and ramp cells;

[0072] Step S330: generate and process the intercommunication structure according to the gantry pair information, determine the intercommunication type of each gantry pair by analyzing the correlation between different gantry pairs, form multiple intercommunication structures, and process the connection relationship between the main road and the ramp in the intercommunication to obtain the intercommunication structure data;

[0073] Step S340, perform virtual gantry generation processing according to the uncovered sections in the T-shaped interchange in the interchange structure data, and introduce virtual gantries in the uncovered main sections to ensure the completeness of the diversion and merging logic of the main road and the ramp, so as to obtain a static road network structure including virtual gantries.

[0074] Specifically, in this process, the divided road network is connected by merging the slave cell-gantry pair-interconnection-road network step by step. The main road cell is defined as , the ramp cell is defined as , the main path cell set is defined as , the ramp cell set is defined as , is the number of main path cells within the gantry, is the number of ramp cells in the gantry pair. The gantry pair is defined as , intercommunication is defined as , the road network is defined as Specifically, the structure of the entire road network is: ; ; , Corresponding to the main path cell set, Corresponding to the ramp cell set. is the number of interconnections in the road network, The number of portal pairs in the intercommunication is the number of main road and ramp cell sets contained in the gantry pair.

[0075] The main road module contains attributes such as the cell number, the main road cell numbers connected to it, and the ramp cell numbers connected to it; the ramp module contains the ramp cell number attribute. The road structure data obtained by the cell division module is assigned to the actual data information of the main road and ramp, and then the ramp numbers involved in the main road cells are matched with the actual ramp cells in the road network connection module to connect the main road cells with the ramp cells. That is, the construction from the cell level to the gantry level is completed. Definition The next node of this node, The previous node of this node, This node is the subsidiary ramp node. The connection formula between the main road cell and the ramp cell is as follows:

[0076] ;

[0077] in, Indicates the serial number of the main road cell; Indicates Main road cells; Indicates +1 main road cell; represents the basic cells in the main road, that is, cells that are not connected to the ramp; Represents the cells in the main road that are connected to the ramp and cause diversion; Represents the cells in the main road that are connected to the ramp and merge into it.

[0078] The road network construction from the gantry pair level to the interchange level requires the cell division module to provide corresponding data. Different gantry pairs may belong to the same interchange, and the main road cells under different gantry pairs are connected through ramps. Under the same interchange, the connection formula is as follows:

[0079] ;

[0080] in, represents a ramp cell; represents a main road cell under the p-th gantry pair; represents a main road cell under the qth gantry pair; p is One of the portal pairs in the interchange, q is Connecting China and Africa One of the gantry pairs. The main path cell has no subscript, which indicates an uncertain situation and needs to be determined based on the cell division module.

[0081] When constructing the interchange-level structure, the T-shaped interchange cannot be completely composed of the existing gantry pairs. In order to better and more accurately estimate the status of the main road and the ramp, it is necessary to propose a virtual gantry and a real gantry to form a gantry pair for the T-shaped interchange. This can completely cover the roads that are not included in the gantry pair on the main highway, so that its status can be estimated more accurately in the future. After the virtual gantry is constructed, the method of dividing and merging the main highway and the ramp changes. The specific formula is as follows:

[0082] ;

[0083] ;

[0084] in, Indicates the last cell number of the main road cell in this gantry pair; represents the virtual upstream gantry; represents the virtual downstream gantry; Indicates that there is a virtual gantry, then this gantry pair can be used as a virtual gantry pair. Modify this method to connect the main road and the ramp through the virtual gantry, and then connect the inside of the interchange. The road network is composed of several interconnections, and finally the entire road network is constructed. At this time, each cell in the entire road network constructed inside the system has multiple attributes, which provides convenience for the subsequent real-time estimation module.

[0085] Further, step S400 includes step S410 to step S440.

[0086] Step S410: According to the static road network structure, real-time data collection and processing of the main road cell and the ramp cell are performed, and the data of the diversion ratio and confluence ratio of the main road and the ramp are obtained regularly through the collection equipment to obtain the real-time flow distribution data between the main road and the ramp cells;

[0087] Step S420: Perform real-time traffic distribution calculation processing according to the real-time traffic distribution data, perform traffic distribution calculation between the main road and the ramp by analyzing the traffic ratio of each main road cell and the ramp cell, and obtain the real-time traffic distribution ratio between the main road cell and the ramp cell;

[0088] Step S430: Calculate the average speed of vehicles in the cell according to the real-time traffic distribution ratio, calculate the average speed of vehicles in each cell based on the relationship between the traffic and density in the cell, and obtain the vehicle speed data of each cell;

[0089] Step S440: update the cell state according to the vehicle speed data, and update the traffic state of each cell by using the flow and density data in the cell to obtain an updated cell state.

[0090] Specifically, the main road cells in the road network are connected to the ramp cells. From a macro perspective, the diversion ratio and confluence ratio between the main road and the ramp change in real time, so the diversion ratio and confluence ratio data between the main road and the ramp need to be refreshed periodically. It is defined as the confluence ratio of the main road cell and the ramp cell. It is defined as the diversion ratio between the main road cell and the ramp cell, and is expressed as follows:

[0091] ;

[0092] ;

[0093] in, It indicates the ratio of the flow rate delivered from the previous main road cell to this main road cell to the flow rate delivered from the previous main road cell and all ramps to this main road cell; Expressed as The ratio of the flow delivered to this main road cell by a ramp to the flow delivered to this main road cell by the previous main road cell and all ramps; It indicates the ratio of the flow rate transmitted from this main road cell to the next main road cell to the flow rate sent from this main road cell to the next main road cell and all ramps; This main cell transmits The ratio of the flow of a ramp to the flow sent by this main road cell to the next main road cell and all ramp flows; Indicates the proportion of the flow rate of the first ramp merging into the downstream main road in the case of merging; Indicates The proportion of traffic from each ramp merging into the downstream main road; Indicates the proportion of flow that is diverted from the main road to the first downstream ramp; It indicates the proportion of the flow rate exported from the main road to the downstream ramp in the case of diversion. It indicates the proportion of the flow rate exported from the main road to the downstream ramp in the case of diversion. These proportions will be refreshed regularly after the data are collected by the collection equipment.

[0094] Furthermore, the above steps also provide a state correction function for state deviation, and the specific formula is:

[0095] ;

[0096] in, Represents the vehicle density in the cell at the current moment; represents the average speed of vehicles in the observed cell; represents the congestion density of the cell; The density representing the maximum capacity of the cell; represents the free flow speed. This formula allows the cell state to be refreshed based on the observed speed.

[0097] Further, step S500 includes step S510 to step S540.

[0098] Step S510: Perform interchange type identification processing according to the updated cell state and static road network structure, determine the specific structure type of the interchange by analyzing the connection between the main road cell and the ramp cell in the interchange, and obtain the identification result of the interchange, which includes a cross interchange, a T-shaped interchange, and a toll booth interchange;

[0099] Step S520: perform flow data estimation processing according to the identification result and the upstream and downstream flow data of the gantry pair, calculate the real input flow by comparing the flow recorded by the upstream gantry with the maximum acceptable amount of the cell, and perform weighted calculation in combination with the downstream gantry flow and the maximum transmittable amount of the cell to obtain the flow data of the cells in the gantry pair, the flow data including the actual input and output flows;

[0100] Step S530: Perform cell state update processing according to the traffic data and the vehicle speed data collected at the video points, convert the cell speed data into vehicle density and traffic information, and calculate the number of acceptable vehicles and the number of sendable vehicles for each cell in combination with the basic graph formula to obtain the real-time state of the cell;

[0101] Step S540: According to the real-time status of the cells, the CTM model is used to deduce the flow and density. Based on the flow transfer relationship between cells, the real-time status of the main road cells and ramp cells in each type of interchange is deduced, and the state transfer of each cell is calculated according to the diversion and merging rules to obtain the 100-meter-level traffic situation estimation results of each interchange type in the entire road network.

[0102] Specifically, the above estimation process considers the heterogeneity of the interchange structure and proposes different estimation methods, which can adapt to the interchange types of all situations in the entire highway network, thereby realizing the state estimation of the hundred-meter level for the entire road network of nearly 10,000 kilometers. This process will jointly estimate the cell state in combination with the road network refresh process, and use the constructed entire road network with multiple attributes to design differentiated estimation methods considering different interchange types of highway main roads and ramps, and use the gantry pair as a closed area to estimate the internal cells. The highway network is complex, but through the analysis of the road network structure, it can be found that the interchange types of the entire road network are mainly covered by cross interchanges, T-shaped interchanges, and general interchanges consisting of main roads, ramps and toll stations. More complex interchange types are composed of these three interchanges. Corresponding estimation methods are constructed for these three interchanges, and on this basis, the estimation methods under the combined interchanges are compatible, which can be applied to any road network, so it is a highly portable system.

[0103] The sparse cross-sectional data of the gantry pair is used as the main source data for estimation. Indicates The mast is aligned The flow value recorded in the upstream gantry at ; Indicates The mast is aligned The flow value recorded in the downstream gantry at that time.

[0104] Secondly, the speed data such as the video points observed inside the gantry are used to assist in improving the estimation accuracy of the cell. Indicates that the door frame is on the inner There are observed speed data for each cell. The upstream gantry flow in the gantry pair is taken smaller than the maximum acceptable amount of the first cell to obtain the actual input flow in the gantry pair, and the downstream gantry flow is taken weighted fusion with the maximum sendable amount of the last cell to obtain the actual output flow in the gantry pair. The number of vehicles acceptable to the cell and the number of vehicles sent are obtained by the basic graph formula:

[0105] ;

[0106] ;

[0107] in, represents the free flow velocity in this cell; represents the density of the maximum traffic capacity in this cell; represents the congestion density within this cell; represents the density within this cell; Indicates The number of vehicles a cell can accept from the previous cell; Indicates The number of vehicles that a cell can send to the next cell for acceptance.

[0108] The main data comes from the traffic recorded by the gantry. When using the gantry traffic, the state of the gantry to the boundary cells must be considered. Therefore, the data input and output formulas are as follows:

[0109] ;

[0110] ;

[0111] in, Represents the number of vehicles finally input to the first cell; Indicates the number of vehicles finally sent out from the last cell; It represents the maximum number of vehicles that the first main road cell in the i-th gantry pair can accept; represents the maximum number of vehicles emitted by the last cell in the i-th gantry pair; and Represents the hyperparameter values ​​involved in weighting; Indicates The mast is aligned The flow value recorded in the upstream gantry at ; Indicates The mast is aligned The flow value recorded in the downstream gantry at that time.

[0112] Each main road cell within the gantry uses the CTM model to adapt the deduction rules of the basic, diversion and confluence situations, assigning the final input flow value to the first cell and the final output flow value to the last cell. The internal status will be updated in real time according to the deduction rules.

[0113] The basic situation deduction rule formula is as follows:

[0114] ;

[0115] The formula for deducing the diversion rules is as follows:

[0116] ;

[0117] The formula for deducing the rules for the confluence situation is as follows:

[0118] ;

[0119] in, Representation Cell exist The number of vehicles actually dispatched during the time period; Representation Cell The actual number of vehicles accepted; Represents ramp cell The actual number of vehicles accepted; Indicates the ramp cell number ; Indicates The maximum number of vehicles that a main road cell can accept; Representation Cell The actual number of vehicles accepted; Indicates The maximum number of vehicles that a ramp cell can accept; Indicates the proportion of traffic sent from the main path cell to the next main path cell in the case of traffic diversion; In the case of splitting, the main cell sends The flow ratio of ramp cells; Indicates The number of vehicles that can be sent by a main road cell; Indicates the proportion of traffic sent from the main path cell to the next main path cell in the case of confluence; In the case of confluence, The proportion of traffic sent from a ramp cell to the next main road cell.

[0120] After obtaining the number of vehicles actually accepted by each cell, the state transfer equation formula of each cell based on the CTM model can be calculated as follows:

[0121] ;

[0122] The state representation formula obtained from the observed data is as follows:

[0123] ;

[0124] in, Indicates the first The current number of vehicles in a cell; Indicates the first The number of vehicles in the last cell; Indicates The number of vehicles actually accepted by the cell at the last moment; Represents the number of cell vehicles converted from observed speed data; Indicates The length of a cell.

[0125] The final state is obtained by combining the state derived from the CTM model with the state converted from the observed data. The formula is as follows:

[0126] ;

[0127] in, Represents the weighted value of the number of cellular vehicles deduced by the CTM model and the number of cellular vehicles obtained by converting the observed data; Indicates the number of vehicles after weighted fusion is used to convert the calculated The average speed of the cells in this step; and is the weighting coefficient.

[0128] Further, step S540 includes step S541 to step S544.

[0129] Step S541: according to the real-time status of the cells, flow calculation processing is performed for the cross hub interchange and the toll station interchange. By analyzing multiple gantry pairs in the interchange, the flow of the upstream gantry is gradually allocated to the first cell, and the flow of the downstream gantry is gradually allocated to the last cell. In accordance with the cell numbering sequence, the flow is gradually transferred and calculated in the main road and the ramp to obtain the first flow estimation result in the cross-toll station interchange.

[0130] Step S542: perform virtual gantry flow estimation processing for the T-shaped interchange according to the real-time status of the cell, estimate the flow transferred from the ramp to the main road through the virtual gantry, and transfer the flow between the virtual gantry and the actual gantry in combination with the diversion ratio, and calculate and obtain a second flow estimation result in the T-shaped interchange;

[0131] Step S543: Analyze different ramp connection conditions for combined interconnection according to the real-time status of the cell, perform weighted allocation of the ramp input or output flow and the main road cell flow based on different ramp types, and update the state according to the input and output flow of each cell to obtain the third flow estimation result in the complex combined interconnection;

[0132] Step S544: Perform cell state update processing based on the first flow estimation result, the second flow estimation result, and the third flow estimation result. Combined with the diversion and merging rules of each interchange type, the flow transfer process of each cell is inferred, and the state transfer of the main road cell and the ramp cell is calculated. Finally, the 100-meter-level traffic situation estimation result of each interchange type in the entire road network is obtained.

[0133] In addition, the estimation of main road cells and ramp cells will vary depending on the interchange type.

[0134] Estimation method for cross-type hub interchange and toll booth-type general interchange:

[0135] The cross-type hub interchange is composed of four gantry pairs, and the internal ramps connect different highway main roads; while the toll station type general interchange is composed of two gantry pairs, and the internal ramps connect the highway main road and the toll station. After traversing the gantry pairs in the interchange, assigning the final gantry input flow to the first cell, and the final gantry output flow to the last cell, the cell state is updated in the order of the cell numbering in the gantry pair. In the process of deduction, the corresponding state propagation rules are called according to the structural properties of the cell. In addition, an external observation data set is input to transform the observed cell state, so as to update the final state. After updating the state of the main road cells in the gantry pair, and then updating the state of the connected ramps, it is necessary to distinguish the type of ramp. If the ramp is connected to the main road at both ends, the state transfer equation of the ramp is directly called to estimate the state of the ramp; if one end of the ramp is connected to the main road and the other end is connected to the toll station, the real-time flow of the toll station is required to be used as the input or output of one end of the ramp, and the toll station is defined as , whose flow rate is Then call the ramp state update equation to estimate the ramp state. The above formula has written the implementation logic of this method. The difference is that when one end of the ramp is connected to the main road and the other end is connected to the toll station, its state update equation changes. The formula is as follows:

[0136] ;

[0137] in, Indicates The real-time flow value of the toll station after the ramp is connected to the toll station; Indicates the number of vehicles actually received by the downstream main road from the ramp cell when the ramp is connected to the main road downstream; represents the cell connected to the upstream of the ramp; Indicates a toll booth; represents the first The number of vehicles in the ramp cell at the current moment; represents the first The number of vehicles in the ramp cell at the last moment; It indicates the number of vehicles actually accepted from the upstream main road when the ramp is connected to the toll station downstream and the main road upstream; Represents the cells connected to the ramp cell downstream.

[0138] T-shaped hub interchange estimation method:

[0139] The T-shaped interchange consists of three gantry pairs. Due to the T-shaped structure, there will be a section of the main road that is not included in the gantry pair. In this case, it is difficult to divide the cells and estimate the state. In order to solve this problem, the concept of virtual gantry is proposed. A virtual gantry is built at the end of this section of the main road. The virtual gantry and the real gantry are combined into a gantry pair for cell division and state estimation. Therefore, there are two virtual gantries in the T-shaped interchange. The virtual gantry does not have collection equipment to record traffic flow and other data, so the estimation method needs to be different from the cross-type interchange and needs to be adjusted. First, the number of vehicles sent from the ramp downstream through the virtual gantry connected to the main road cell is calculated, which is defined as , the sum of the number of vehicles sent from all ramps is used as the gantry input source of the next gantry pair, and the other main road cells are updated through the deduction rules. Then the number of vehicles sent by the last cell is sent to the connected ramps according to the diversion ratio. The specific formula is expressed as follows:

[0140] ;

[0141] in, Indicates that the main line sends The flow rate proportion of the ramps; Afterwards, the ramp state is updated through the state update equation.

[0142] By following this approach, the estimation of the T-shaped interchange can be completed.

[0143] Complex combination intercommunication estimation method:

[0144] Complex combined interchanges are generally composed of a combination of cross-type hub interchanges, T-type hub interchanges and general interchanges at toll stations, which are mainly reflected in different ramp types. In a cross-type interchange, both ends of the ramp are connected to the main road cells; in a toll station interchange, one end of the ramp is connected to the main road cell and the other end is connected to the toll station; in a T-type interchange, one end of the ramp is connected to the main road cell and the other end is connected to the main road cell through a virtual gantry. In a complex combination of interchanges, there will be a situation where one end of the ramp is connected to the toll station and the other end is connected to the main road cell through a virtual gantry. As long as this type of situation is applicable, the estimation of complex combined interchanges can be compatible. In order to facilitate the expression of the formula, we simplify the situation where the ramp upstream receives the toll station and the downstream is connected to the virtual gantry in the formula to the ramp upstream receives the toll station, and vice versa. The specific formula is as follows:

[0145] ;

[0146] When the estimation under multiple types of interchanges can be completed, in order to reduce the problem of inaccurate precision caused by sparse cross-sectional data sources, other data are integrated to assist in improving the estimation accuracy when using gantry data for estimation. Other data include speed data of video points, etc. The speed data recorded at the video point is converted into the number of vehicles through the basic traffic map, and then weighted fused with the number of vehicles of this cell obtained by deduction. The fused number of vehicles is then converted through the basic traffic map to obtain the speed index. The number of vehicles that each cell finally sends to the next cell in real time during the deduction process is used as the cross-sectional flow of this step. Finally, the state estimation of cell-gantry pair-interchange-road network can be completed.

[0147] Embodiment 2:

[0148] like Figure 2 As shown, this embodiment provides a system for estimating traffic status of a large-scale highway network, and the system includes:

[0149] Acquisition module 1, used to acquire an initial data set, which includes ETC gantry pair codes, pile numbers and distance information on each line of the entire road network;

[0150] Decomposition module 2 is used to perform decomposition processing according to the initial data set, divide the road section into cells using the upstream and downstream gantry information of the same route and the same direction, and optimize the cell length and number to obtain the cell division result, which includes the main road cell information, ramp cell information and corresponding connection information;

[0151] Construction module 3 is used to construct the road network according to the cell division result, by merging cells, gantry pairs and interchange structures step by step, connecting the main road cells and ramp cells, and constructing a virtual gantry at the T-shaped interchange to obtain a static road network structure;

[0152] Update module 4 is used to calculate the real-time flow rate according to the static road network structure and the real-time collected flow distribution ratio and confluence ratio data of the main road cell and the ramp cell to obtain the flow distribution ratio of the main road and the ramp in the whole road network, and calculate the real-time flow rate and density data of the cell according to the collected average speed of the vehicle in the cell to obtain the updated cell state;

[0153] Estimation module 5 is used to make differentiated estimates of different interchange types based on the updated cellular state and static road network structure, use the flow data of the gantry pair and the speed data of the video points to make state estimates, and use the preset CTM model to deduce the cellular flow and density to generate the traffic situation of each interchange type in the entire road network, and obtain the 100-meter-level traffic state estimation results of the entire road network.

[0154] In a specific embodiment disclosed in the present application, the decomposition module 2 includes:

[0155] The first division unit 21 is used to perform division processing according to the ramp structure information in the initial data set, convert each ramp into a corresponding ramp cell, and form a data document to obtain ramp cell information;

[0156] The first calculation unit 22 is used to calculate the optimal number and length of cell divisions according to the gantry-to-road section length data in the initial data set to obtain the optimal division scheme;

[0157] The first detection unit 23 is used to perform consistency detection based on the upstream and downstream gantry pile number information in the initial data set, and check whether the upstream and downstream pile numbers are consistent by comparison. If there is inconsistency, the pile number conversion processing is performed based on a preset conversion rule to obtain the converted pile number information;

[0158] The second division unit 24 is used to perform main road cell division processing according to the converted pile number information to obtain main road cell information, and integrate the main road cell with the ramp cell to match their respective connection relationships to obtain the connection information between the main road and the ramp.

[0159] In a specific implementation disclosed in the present application, building block 3 includes:

[0160] The first processing unit 31 is used to perform connection processing of the main road and ramp cells according to the cell division result, and to obtain preliminary connection information by matching adjacent main road cells and ramp cells one by one by analyzing the upstream and downstream position relationship between the main road and the ramp;

[0161] A first construction unit 32 is used to perform road network construction processing at the gantry pair level according to the preliminary connection information, by classifying and grouping the connected cells to form gantry pairs, thereby obtaining gantry pair information including main road and ramp cells;

[0162] The first generating unit 33 is used to generate and process the interconnection structure according to the gantry pair information, determine the interconnection type of each gantry pair by analyzing the correlation between different gantry pairs, form multiple interconnection structures, and process the connection relationship between the main road and the ramp in the interchange to obtain the interconnection structure data;

[0163] The second generation unit 34 is used to perform virtual gantry generation processing according to the uncovered sections in the T-shaped interchange in the interchange structure data, and introduce virtual gantries in the uncovered main sections to ensure the completeness of the diversion and merging logic of the main road and the ramp, so as to obtain a static road network structure including virtual gantries.

[0164] In a specific implementation disclosed in the present application, the update module 4 includes:

[0165] The first acquisition unit 41 is used to perform real-time data acquisition and processing of the main road cell and the ramp cell according to the static road network structure, and obtain the data of the diversion ratio and confluence ratio of the main road and the ramp through the acquisition device at regular intervals to obtain the real-time flow distribution data between the main road and the ramp cell;

[0166] A first allocation unit 42 is used to perform real-time traffic allocation calculation processing according to the real-time traffic allocation data, and calculate the traffic allocation between the main road and the ramp by analyzing the traffic ratio of each main road cell and the ramp cell, so as to obtain the real-time traffic allocation ratio between the main road cell and the ramp cell;

[0167] The second calculation unit 43 is used to calculate the average speed of vehicles in the cell according to the real-time flow distribution ratio, and calculate the average speed of vehicles in each cell based on the relationship between the flow and density in the cell to obtain the vehicle speed data of each cell;

[0168] The first updating unit 44 is used to update the cell state according to the vehicle speed data, and to update the traffic state of each cell by using the flow and density data in the cell to obtain an updated cell state.

[0169] In a specific implementation disclosed in the present application, the estimation module 5 includes:

[0170] The first identification unit 51 is used to perform interchange type identification processing according to the updated cell state and the static road network structure, determine the specific structure type of the interchange by analyzing the connection between the main road cell and the ramp cell in the interchange, and obtain the identification result of the interchange, which includes a cross interchange, a T-shaped interchange and a toll station interchange;

[0171] The first estimation unit 52 is used to perform flow data estimation processing according to the identification result and the upstream and downstream flow data of the gantry pair, calculate the real input flow by comparing the flow recorded by the upstream gantry with the maximum acceptable amount of the cell, and perform weighted calculation in combination with the downstream gantry flow and the maximum transmittable amount of the cell to obtain the flow data of the cells in the gantry pair, the flow data including the actual input and output flows;

[0172] The second updating unit 53 is used to perform cell state update processing according to the traffic data and the vehicle speed data collected at the video point, by converting the cell speed data into vehicle density and traffic information, and calculating the number of acceptable vehicles and the number of sendable vehicles for each cell in combination with the basic graph formula, to obtain the real-time state of the cell;

[0173] The third calculation unit 54 is used to deduce the flow and density according to the real-time state of the cells using the CTM model, deduce the real-time state of the main road cells and ramp cells in each type of interchange based on the flow transfer relationship between the cells, and calculate the state transfer of each cell according to the diversion and merging rules to obtain the 100-meter-level traffic situation estimation results of each interchange type in the entire road network.

[0174] In a specific implementation disclosed in the present application, the third calculation unit 54 includes:

[0175] The fourth calculation unit 541 is used to perform flow calculation processing for the cross hub interchange and the toll station interchange according to the real-time status of the cells, and gradually allocate the flow of the upstream gantry to the first cell and the flow of the downstream gantry to the last cell by analyzing the multiple gantry pairs in the interchange, and gradually transfer and calculate the flow in the main road and the ramp according to the cell numbering sequence to obtain the first flow estimation result in the cross-toll station interchange;

[0176] A fifth calculation unit 542 is used to perform virtual gantry flow estimation processing for the T-shaped interchange according to the real-time state of the cell, estimate the flow transferred from the ramp to the main road through the virtual gantry, and transfer the flow between the virtual gantry and the actual gantry in combination with the diversion ratio, so as to calculate and obtain a second flow estimation result in the T-shaped interchange;

[0177] The sixth calculation unit 543 is used to analyze different ramp connection conditions for the combined interchange according to the real-time state of the cell, perform weighted allocation of the ramp input or output flow and the main road cell flow based on different ramp types, and update the state according to the input and output flow of each cell to obtain a third flow estimation result in the complex combined interchange;

[0178] The seventh calculation unit 544 is used to perform cell state update processing according to the first flow estimation result, the second flow estimation result and the third flow estimation result, and to infer the flow transfer process of each cell in combination with the diversion and merging rules of each interchange type, and to calculate the state transfer of the main road cell and the ramp cell, and finally obtain the 100-meter-level traffic situation estimation results of each interchange type in the entire road network.

[0179] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for estimating traffic status of a large-scale highway network, characterized in that: include: Acquire an initial data set, wherein the initial data set includes ETC gantry pair codes, pile numbers, and distance information on each line of the entire road network; Decomposing the initial data set, dividing the road section into cells using the upstream and downstream gantry information of the same route and the same direction, and optimizing the cell length and number to obtain a cell division result, wherein the cell division result includes main road cell information, ramp cell information, and corresponding connection information; A road network construction process is performed according to the cell division result, by merging cells, gantry pairs and interchange structures step by step, connecting main road cells and ramp cells, and constructing a virtual gantry at a T-shaped interchange to obtain a static road network structure; According to the static road network structure, combined with the real-time collected data of the diversion ratio and confluence ratio of the main road cell and the ramp cell, the real-time flow calculation is performed to obtain the flow distribution ratio of the main road and the ramp in the entire road network, and the real-time flow and density data of the cell are calculated according to the collected average speed of the vehicle in the cell to obtain the updated cell state; According to the updated cellular state and the static road network structure, different interchange types are estimated differentially, the flow data of the gantry pair and the speed data of the video point are used to estimate the state, and the cellular flow and density are deduced through the preset CTM model to generate the traffic situation of each interchange type in the whole road network, and obtain the 100-meter-level traffic state estimation result of the whole road network; The road network construction process is performed according to the cell division result, by merging cells, gantry pairs and interchange structures step by step, connecting the main road cells and ramp cells, and constructing a virtual gantry at the T-shaped interchange to obtain a static road network structure, including: According to the cell division result, the main road and ramp cells are connected, and the upstream and downstream positional relationship between the main road and the ramp is analyzed, and the adjacent main road cells and ramp cells are matched one by one to obtain preliminary connection information; Performing road network construction processing at the gantry pair level according to the preliminary connection information, by classifying and grouping the connected cells to form gantry pairs, thereby obtaining gantry pair information including main road and ramp cells; Generate and process the interconnection structure according to the gantry pair information, determine the interconnection type of each gantry pair by analyzing the correlation between different gantry pairs, form multiple interconnection structures, and process the connection relationship between the main road and the ramp in the interchange to obtain the interconnection structure data; Performing virtual gantry generation processing according to the uncovered road sections in the T-shaped interchange in the interchange structure data, introducing virtual gantry in the uncovered main road sections, ensuring the completeness of the divergence and merging logic between the main road and the ramp, and obtaining a static road network structure including virtual gantry; The divided road network is connected by merging the slave cell-gantry pair-interconnection-road network step by step; the main road cell is defined as C m , the ramp cell is defined as C r , the main path cell set is defined as The ramp cell set is defined as n is the number of main road cells in the gantry pair, m is the number of ramp cells in the gantry pair; the gantry pair is defined as G, the interchange is defined as I, and the road network is defined as N; the structure of the entire road network: N = {I1, I2, …, I z }; I = {G1, G2, …, G n }; G = {M, R}, M corresponds to the main road cell set, and R corresponds to the ramp cell set; where z is the number of interchanges in the road network, n is the number of gantry pairs in the interchange, and i is the number of main road and ramp cell sets contained in the gantry pair; The main road module contains the cell number, the main road cell numbers connected to it, and the ramp cell number attributes connected to it; the ramp module contains the ramp cell number attribute; the road structure data obtained by the cell division module is assigned to the actual data information of the main road and the ramp, and then the ramp number involved in the main road cell is matched with the actual ramp cell in the road network connection module to connect the main road cell with the ramp cell; that is, the construction from the cell level to the gantry level is completed; Next(·) is defined as the next node of this node, Pre(·) is defined as the previous node of this node, and Ramp(·) is defined as the attached ramp node of this node; the formula for connecting the main road cell and the ramp cell is as follows: Among them, i represents the serial number of the main path cell; represents the i-th main road cell; represents the i+1th main road cell; baseCell represents the basic cell in the main road, that is, the cell that is not connected to the ramp; divergeCell represents the cell in the main road that is connected to the ramp and produces the divergence phenomenon; mergeCell represents the cell in the main road that is connected to the ramp and produces the confluence phenomenon; Under the same interconnection, the connection formula is as follows: Among them, C r represents a ramp cell; represents a main road cell under the p-th gantry pair; represents a main road cell under the qth gantry pair; p is one of the gantry pairs in the I interchange, and q is one of the non-p gantry pairs in the I interchange; the main road cell has no subscript, which represents an uncertain situation and needs to be determined according to the cell division module; For the T-shaped interchange, a virtual gantry and a real gantry are proposed to form a gantry pair. After the virtual gantry is constructed, the methods of dividing and merging the main road and the ramp of the expressway are changed. The specific formula is as follows: Among them, max represents the last cell number of the main road cell in this gantry pair; VirtualG upper VirtualG represents the virtual upstream gantry; lower Indicates a virtual downstream gantry; VirtualG indicates the existence of a virtual gantry, and this gantry pair can be regarded as a virtual gantry pair.

2. The method for estimating traffic status of a large-scale highway network according to claim 1, characterized in that: Decomposition processing is performed according to the initial data set, and the road section is cell-divided using the upstream and downstream gantry information of the same route and the same direction, and the cell length and number are optimized to obtain the cell division result, including: Performing division processing according to the ramp structure information in the initial data set, converting each ramp into a corresponding ramp cell, and forming a data document to obtain ramp cell information; Calculate the optimal number and length of cell divisions according to the gantry-to-road section length data in the initial data set to obtain the optimal division scheme; Perform consistency detection based on the upstream and downstream gantry pile number information in the initial data set, and check whether the upstream and downstream pile numbers are consistent by comparison. If there is inconsistency, perform pile number conversion processing based on a preset conversion rule to obtain converted pile number information; The main road cell division process is performed according to the converted pile number information to obtain the main road cell information, and the main road cell and the ramp cell are integrated to match their respective connection relationships to obtain the connection information of the main road and the ramp.

3. The method for estimating traffic status of a large-scale highway network according to claim 1, characterized in that: According to the static road network structure, combined with the real-time collected data of the diversion ratio and confluence ratio of the main road cell and the ramp cell, the real-time flow calculation is performed to obtain the flow distribution ratio of the main road and the ramp in the entire road network, and the real-time flow and density data of the cell are calculated according to the collected average speed of the vehicle in the cell to obtain the updated cell state, including: According to the static road network structure, real-time data collection and processing of the main road cell and the ramp cell are performed, and the data of the diversion ratio and confluence ratio of the main road and the ramp are obtained regularly through the collection equipment to obtain the real-time flow distribution data between the main road and the ramp cells; Performing real-time traffic distribution calculation processing according to the real-time traffic distribution data, performing traffic distribution calculation between the main road and the ramp by analyzing the traffic ratio of each main road cell and the ramp cell, and obtaining the real-time traffic distribution ratio between the main road cell and the ramp cell; Calculating and processing the average speed of vehicles in the cell according to the real-time traffic distribution ratio, and calculating the average speed of vehicles in each cell based on the relationship between the traffic and density in the cell to obtain vehicle speed data for each cell; The cell state is updated according to the vehicle speed data, and the traffic state of each cell is updated by using the flow and density data in the cell to obtain an updated cell state.

4. The method for estimating traffic status of a large-scale highway network according to claim 1, characterized in that: According to the updated cellular state and the static road network structure, different interchange types are estimated differentially, and the state is estimated using the flow data of the gantry pair and the speed data of the video point. The cellular flow and density are deduced through the preset CTM model to generate the traffic situation of each interchange type in the entire road network, and the 100-meter-level traffic state estimation results of the entire road network are obtained, including: According to the updated cell state and the static road network structure, an interchange type identification process is performed, and a specific structural type of the interchange is determined by analyzing the connection between the main road cell and the ramp cell in the interchange, and an identification result of the interchange is obtained, wherein the identification result includes a cross interchange, a T-shaped interchange, and a toll station interchange; Performing flow data estimation processing according to the identification result and the upstream and downstream flow data of the gantry pair, calculating the real input flow by comparing the flow recorded by the upstream gantry with the maximum acceptable amount of the cell, and performing weighted calculation in combination with the downstream gantry flow and the maximum transmittable amount of the cell, to obtain the flow data of the cells in the gantry pair, wherein the flow data includes the actual input and output flows; Performing cell state update processing according to the traffic data and vehicle speed data collected at the video points, converting the cell speed data into vehicle density and traffic information, and calculating the number of acceptable vehicles and the number of sendable vehicles for each cell in combination with the basic graph formula, to obtain the real-time state of the cell; According to the real-time status of the cells, the CTM model is used to deduce the flow and density. Based on the flow transfer relationship between cells, the real-time status of the main road cells and ramp cells in each type of interchange is deduced, and the state transfer of each cell is calculated according to the diversion and merging rules to obtain the 100-meter-level traffic situation estimation results of each interchange type in the entire road network.

5. A highway large-scale road network traffic state estimation system, characterized in that: include: An acquisition module is used to acquire an initial data set, wherein the initial data set includes ETC gantry pair codes, pile numbers and distance information on each line of the entire road network; A decomposition module is used to perform decomposition processing according to the initial data set, divide the road section into cells using the upstream and downstream gantry information of the same route and the same direction, and optimize the cell length and number to obtain a cell division result, wherein the cell division result includes main road cell information, ramp cell information and corresponding connection information; A construction module is used to perform road network construction processing according to the cell division result, connect the main road cells with the ramp cells by merging cells, gantry pairs and interchange structures step by step, and construct a virtual gantry at the T-shaped interchange to obtain a static road network structure; An updating module is used to calculate the real-time flow rate according to the static road network structure and the real-time collected data of the flow distribution ratio and confluence ratio of the main road cell and the ramp cell to obtain the flow distribution ratio of the main road and the ramp in the whole road network, and calculate the real-time flow rate and density data of the cell according to the collected average speed of the vehicle in the cell to obtain the updated cell state; An estimation module is used to perform differentiated estimation on different interchange types according to the updated cellular state and the static road network structure, perform state estimation using the flow data of the gantry pair and the speed data of the video point, and deduce the cellular flow and density through a preset CTM model to generate the traffic situation of each interchange type in the entire road network, and obtain the 100-meter-level traffic state estimation result of the entire road network; Wherein, the building blocks include: A first processing unit is used to perform connection processing of the main road and ramp cells according to the cell division result, and to obtain preliminary connection information by matching adjacent main road cells and ramp cells one by one by analyzing the upstream and downstream position relationship between the main road and the ramp; A first construction unit is used to perform road network construction processing at the gantry pair level according to the preliminary connection information, by classifying and grouping the connected cells to form gantry pairs, thereby obtaining gantry pair information including main road and ramp cells; A first generating unit is used to generate an intercommunication structure according to the gantry pair information, determine the intercommunication type of each gantry pair by analyzing the correlation between different gantry pairs, form multiple intercommunication structures, and process the main road and ramp connection relationship in the intercommunication to obtain intercommunication structure data; A second generation unit is used to perform virtual gantry generation processing according to the uncovered road section in the T-shaped interchange in the interchange structure data, and introduce a virtual gantry in the uncovered main road section to ensure the completeness of the divergence and merging logic of the main road and the ramp, so as to obtain a static road network structure including the virtual gantry; The divided road network is connected by merging the slave cell-gantry pair-interconnection-road network step by step; the main road cell is defined as C m , the ramp cell is defined as C r , the main path cell set is defined as The ramp cell set is defined as n is the number of main road cells in the gantry pair, m is the number of ramp cells in the gantry pair; the gantry pair is defined as G, the interchange is defined as I, and the road network is defined as N; the structure of the entire road network: N = {I1, I2, …, I z }; I = {G1, G2, …, G n }; G = {M, R}, M corresponds to the main road cell set, and R corresponds to the ramp cell set; where z is the number of interchanges in the road network, n is the number of gantry pairs in the interchange, and i is the number of main road and ramp cell sets contained in the gantry pair; The main road module contains the cell number, the main road cell numbers connected to it, and the ramp cell number attributes connected to it; the ramp module contains the ramp cell number attribute; the road structure data obtained by the cell division module is assigned to the actual data information of the main road and the ramp, and then the ramp number involved in the main road cell is matched with the actual ramp cell in the road network connection module to connect the main road cell with the ramp cell; that is, the construction from the cell level to the gantry level is completed; Next(·) is defined as the next node of this node, Pre(·) is defined as the previous node of this node, and Ramp(·) is defined as the attached ramp node of this node; the formula for connecting the main road cell and the ramp cell is as follows: Among them, i represents the serial number of the main path cell; represents the i-th main road cell; represents the i+1th main road cell; baseCell represents the basic cell in the main road, that is, the cell that is not connected to the ramp; divergeCell represents the cell in the main road that is connected to the ramp and produces the divergence phenomenon; mergeCell represents the cell in the main road that is connected to the ramp and produces the confluence phenomenon; Under the same interconnection, the connection formula is as follows: Among them, C r represents a ramp cell; represents a main road cell under the p-th gantry pair; represents a main road cell under the qth gantry pair; p is one of the gantry pairs in the I interchange, and q is one of the non-p gantry pairs in the I interchange; the main road cell has no subscript, which represents an uncertain situation and needs to be determined according to the cell division module; For the T-shaped interchange, a virtual gantry and a real gantry are proposed to form a gantry pair. After the virtual gantry is constructed, the methods of dividing and merging the main road and the ramp of the expressway are changed. The specific formula is as follows: Among them, max represents the last cell number of the main road cell in this gantry pair; VirtualG upper VirtualG represents the virtual upstream gantry; lower Indicates a virtual downstream gantry; VirtualG indicates the existence of a virtual gantry, and this gantry pair can be regarded as a virtual gantry pair.

6. The highway large-scale road network traffic state estimation system according to claim 5, characterized in that: The decomposition module comprises: A first division unit is used to perform division processing according to the ramp structure information in the initial data set, convert each ramp into a corresponding ramp cell, and form a data document to obtain ramp cell information; A first calculation unit is used to calculate the optimal number and length of cell divisions according to the gantry-to-road section length data in the initial data set to obtain an optimal division scheme; A first detection unit is used to perform consistency detection based on the upstream and downstream gantry pile number information in the initial data set, and check whether the upstream and downstream pile numbers are consistent by comparison. If there is inconsistency, the pile number conversion processing is performed based on a preset conversion rule to obtain the converted pile number information; The second division unit is used to perform main road cell division processing according to the converted pile number information to obtain main road cell information, and integrate the main road cell with the ramp cell, match their respective connection relationships, and obtain the connection information between the main road and the ramp.

7. The highway large-scale road network traffic state estimation system according to claim 5, characterized in that: The update module includes: A first collection unit is used to collect and process real-time data of the main road cell and the ramp cell according to the static road network structure, and obtain the split ratio and confluence ratio data of the main road and the ramp through the collection device at regular intervals to obtain the real-time flow distribution data between the main road and the ramp cell; A first allocation unit is used to perform real-time traffic allocation calculation processing according to the real-time traffic allocation data, and perform traffic allocation calculation between the main road and the ramp by analyzing the traffic ratio of each main road cell and the ramp cell to obtain the real-time traffic allocation ratio between the main road cell and the ramp cell; A second calculation unit is used to calculate the average speed of vehicles in the cell according to the real-time traffic distribution ratio, and calculate the average speed of vehicles in each cell based on the relationship between the traffic and density in the cell to obtain vehicle speed data for each cell; The first updating unit is used to update the cell state according to the vehicle speed data, and update the traffic state of each cell by using the flow and density data in the cell to obtain an updated cell state.

8. The highway large-scale road network traffic state estimation system according to claim 5, characterized in that: The estimation module comprises: A first identification unit is used to perform interchange type identification processing according to the updated cell state and the static road network structure, determine the specific structure type of the interchange by analyzing the connection between the main road cell and the ramp cell in the interchange, and obtain an identification result of the interchange, wherein the identification result includes a cross interchange, a T-shaped interchange, and a toll booth interchange; a first estimation unit, configured to perform flow data estimation processing according to the identification result and the upstream and downstream flow data of the gantry pair, calculate the real input flow by comparing the flow recorded by the upstream gantry with the maximum acceptable amount of the cell, and perform weighted calculation in combination with the downstream gantry flow and the maximum transmittable amount of the cell, so as to obtain the flow data of the cells in the gantry pair, wherein the flow data includes the actual input and output flows; A second updating unit is used to perform a cell state update process according to the traffic data and the vehicle speed data collected at the video point, by converting the cell speed data into vehicle density and traffic information, and calculating the number of acceptable vehicles and the number of sendable vehicles for each cell in combination with the basic graph formula, to obtain the real-time state of the cell; The third calculation unit is used to deduce the flow and density using the CTM model according to the real-time state of the cell, perform real-time state deduction of the main road cells and ramp cells in each type of interchange based on the flow transfer relationship between cells, and calculate the state transfer of each cell according to the diversion and merging rules to obtain the 100-meter-level traffic situation estimation results of each interchange type in the entire road network.

Images