Intersection average traffic flow analysis method, system, and storage medium

Abstract

The present application belongs to the field of traffic technology, and particularly relates to a method and system for analyzing average traffic flow at an intersection and a storage medium. First, a traffic flow monitoring platform is constructed, and traffic flow monitoring devices are arranged at each intersection according to a preset device arrangement method. Then, the traffic flow monitoring devices are connected to vehicle terminals at each intersection, real-time data and recent data of the vehicle terminals at each intersection, as well as real-time flow data and recent flow data are obtained, and transmitted to the traffic flow monitoring platform. The obtained data is input into a traffic average flow model in the traffic flow monitoring platform for training and verification, and a trained traffic average flow model is output. Finally, the trained traffic average flow model is used to generate average flow data of the current period at the intersection. The present application can solve the problems of high cost and difficult maintenance in the analysis of average traffic flow at an intersection in the existing traffic management process.

Description

Technical Field

[0001] This invention belongs to the field of traffic technology, and in particular relates to a method, system and storage medium for analyzing average traffic flow at intersections. Background Technology

[0002] Intersection average traffic flow analysis is a crucial research area in traffic management and planning. It utilizes intersection traffic flow data to measure and study the characteristics of traffic flow and its relationship with traffic management decisions. Currently, its specific applications in traffic management and planning include intersection safety assessment, traffic management decision-making, and urban planning. Intersection safety assessment identifies potential safety risks at intersections by analyzing average traffic flow. Traffic management decision-making assesses traffic congestion by analyzing traffic flow at different time periods and intersections, thereby guiding traffic signal optimization, road improvements, and public transportation planning. Urban planning determines future urban traffic needs and infrastructure planning by analyzing traffic flow distribution and trends.

[0003] Current methods for analyzing average traffic flow at intersections mainly employ statistical analysis, machine learning, and simulation models. Statistical analysis calculates average traffic flow over different time periods by analyzing historical traffic flow data. Machine learning uses algorithms to build traffic flow prediction models to predict future traffic flow. Simulation models simulate intersection operations under different traffic flow scenarios to optimize decision-making.

[0004] While existing algorithms and models exist for analyzing average traffic flow at intersections, several problems remain. These include data quality issues, such as noise and errors in the collected data affecting the accuracy of the analysis; and a lack of comprehensive analysis, as current methods primarily focus on traffic flow at single intersections, limiting their ability to analyze the entire traffic segment. To overcome these problems, current measures include increasing the quantity and quality of data collection devices and improving the multi-processing capabilities of processing servers. However, these measures are costly and difficult to maintain. Therefore, there is a need to develop a low-cost, easily maintainable method for analyzing average traffic flow at intersections. Summary of the Invention

[0005] The technical problem solved by this invention is to provide a method, system and storage medium for analyzing average traffic flow at intersections, so as to solve the problems of high cost and difficulty in maintenance of existing technologies for analyzing average traffic flow at intersections in the process of traffic management.

[0006] The basic solution provided by this invention: an average traffic flow analysis method for intersections, including:

[0007] S1: Construct a traffic flow monitoring platform, deploy traffic flow monitoring equipment at each traffic intersection according to the preset equipment deployment method, and connect the traffic flow monitoring equipment to the traffic flow monitoring platform;

[0008] S2: Connect the traffic flow monitoring equipment to the vehicle terminals at each traffic intersection to obtain real-time and recent data from the vehicle terminals at each traffic intersection, as well as real-time and recent traffic flow data from the traffic intersection through the traffic flow monitoring equipment, and transmit the obtained real-time and recent data from the vehicle terminals, as well as the real-time and recent traffic flow data, to the traffic flow monitoring platform.

[0009] S3: In the traffic flow monitoring platform, a traffic average flow model is built. The recent traffic flow data of traffic intersections is input into the traffic average flow model and the traffic average flow model is trained. Then, the recent data of vehicle terminals is input into the trained traffic average flow model to verify the training results of the traffic average flow model until the verified traffic average flow model is output after training is completed.

[0010] S4: Input the real-time traffic flow data of the intersection and the real-time data of the vehicle terminal into the trained average traffic flow model to generate the average traffic flow data of the intersection for the current time period.

[0011] Furthermore, in step S1, the deployment of traffic flow monitoring equipment at each intersection according to a preset equipment deployment method specifically includes:

[0012] Obtain historical traffic data stored in the traffic flow monitoring platform, and extract traffic flow, vehicle type, and traffic congestion information from the historical traffic data;

[0013] Extract the map stored in the traffic flow monitoring platform, and display the extracted traffic flow, vehicle type and traffic congestion in the map in three dimensions;

[0014] Based on historical traffic data, traffic intersections are marked on the map.

[0015] Based on traffic flow data from historical traffic data, the free-flow speed at marked traffic intersections is calculated at preset time intervals.

[0016] Based on traffic flow and vehicle type in historical traffic data, target vehicles in the vehicle type are identified, and the trajectory length and time difference of the identified target vehicles in different cycles before and after the marked traffic intersection are obtained. The average speed of the marked traffic intersection is then calculated.

[0017] The average driving speed is compared with the free flow speed to generate a traffic operation status level at each time interval, and traffic flow monitoring equipment is deployed according to the traffic operation status level.

[0018] Furthermore, S2 includes:

[0019] S2-1: Connect vehicle terminals within a preset range to the traffic flow monitoring equipment, receive data from the vehicle terminals at the current traffic intersection, and transmit it to the traffic flow monitoring platform;

[0020] S2-2: The traffic flow monitoring platform obtains the vehicle model corresponding to the vehicle terminal based on the transmitted vehicle terminal data and generates the vehicle conversion coefficient.

[0021] S2-3: The traffic flow monitoring platform generates a data collection period based on the traffic operation status level of the traffic intersection, and extracts recent data and current real-time data from the data transmitted by vehicle terminals during the data collection period.

[0022] S2-4: Traffic flow monitoring equipment monitors the current traffic intersection, acquires recent traffic flow data and current real-time traffic flow data for the data collection period based on the current traffic operation status level, and transmits them to the traffic flow monitoring platform.

[0023] Furthermore, S3 includes:

[0024] S3-1: Receives real-time and recent data from vehicle terminals, as well as real-time and recent traffic flow data from traffic intersections;

[0025] S3-2: Based on the recent traffic flow data received from traffic intersections, generate variables for traffic flow, traffic condition level, vehicle speed, and vehicle type according to traffic tags;

[0026] S3-3: Filter variables based on their missing values, missing rate, and IV value;

[0027] S3-4: The selected variables are used as a sample set, which is then divided into a training set and a test set. The optimal hyperparameter combination is selected using a grid search method. For each hyperparameter combination, a model is trained using the training set to obtain a model. The data values ​​of the model are calculated in the test set. The hyperparameter combination of the model with the best data values ​​in the test set is used as the final hyperparameter value. The variables used are used as the final input variables. The training set and the test set are then merged. Based on the final hyperparameter values, the final input variables, and the vehicle conversion coefficient, the model is fitted to generate the average traffic flow model.

[0028] S3-5: Collect recent data from vehicle terminals, extract vehicle speed and vehicle model from the collected recent data, generate verification traffic flow data, process the verification traffic flow data according to steps S3-2 to S3-3, input it into the average traffic flow model, evaluate the output results, and generate the final average traffic flow model.

[0029] Furthermore, S4 includes:

[0030] S4-1: Input the real-time traffic flow data of the traffic intersection into the final average traffic flow model and output the real-time average traffic flow result;

[0031] S4-2: Input the real-time data from the vehicle terminal into the final average traffic flow model and output the verification average traffic flow results;

[0032] S4-3: Compare the real-time average traffic flow result with the verification average traffic flow result. If the comparison result meets the requirements, the real-time average traffic flow result shall be used as the final average traffic flow analysis result of the traffic intersection.

[0033] Furthermore, the vehicle conversion factors are as follows: 1.0 for passenger cars, 2.0 for large passenger cars, 2.5 for large freight cars, and 3.0 for articulated vehicles.

[0034] Furthermore, it also includes S5: obtaining historical traffic flow data at traffic intersections and determining the traffic flow level at traffic intersections; dividing the average traffic flow model into multiple levels based on the traffic flow level at traffic intersections, with each level corresponding to a traffic flow level at a traffic intersection;

[0035] S6: Construct a level prediction model to obtain preliminary traffic flow information at traffic intersections. Determine the traffic flow level at the intersections using the level prediction model, and call the average traffic flow model corresponding to the traffic flow level when executing S4.

[0036] The intersection average traffic flow analysis system, applied to the above-mentioned intersection average traffic flow analysis method, includes a traffic flow monitoring platform and traffic flow monitoring equipment. The traffic flow monitoring platform and traffic flow monitoring equipment are communicatively connected. The traffic flow monitoring equipment is deployed according to a preset equipment deployment method and is networked with vehicle terminals within a preset range at each traffic intersection.

[0037] The traffic flow monitoring device is used to collect real-time and recent traffic flow data at traffic intersections and transmit them to the traffic flow monitoring platform; the traffic flow monitoring device is also used to acquire real-time and recent data from vehicle terminals at each traffic intersection and transmit them to the traffic flow monitoring platform.

[0038] The traffic flow monitoring platform includes a model building module, a model verification module, and a data processing module. The model building module is used to build a traffic average flow model and input recent traffic flow data from traffic intersections into the traffic average flow model to train the traffic average flow model. The model verification module is used to input recent vehicle terminal data into the traffic average flow model to verify the training results of the traffic average flow model until the verification is successful and the trained traffic average flow model is output. The data processing module is used to input real-time traffic flow data from traffic intersections and real-time vehicle terminal data into the trained traffic average flow model to generate the traffic intersection average flow data for the current time period.

[0039] A storage medium storing a program or instructions that causes a computer to execute the intersection average traffic flow analysis method as described above.

[0040] The principle and advantages of this invention are as follows: In this application, a traffic flow monitoring platform is first constructed. As the average flow calculation terminal at traffic intersections, the traffic flow monitoring platform can process the received data. As for the real-time performance and validity of the received data, this application improves the performance by deploying traffic flow monitoring equipment. First, the traffic flow monitoring equipment collects real-time and recent flow data at traffic intersections. Second, it connects with vehicle terminals within a preset area of ​​the traffic intersection to receive real-time and recent data from the vehicle terminals. The acquisition of real-time and real-time flow data ensures the real-time performance of the data source, and the acquisition of recent and recent flow data ensures the validity of the data after processing.

[0041] Meanwhile, for the calculation of average traffic flow, this application considers different vehicle types, such as passenger cars and large trucks. Different vehicle types have different data on road travel. By configuring the vehicle conversion coefficient, the calculation results are more accurate. Furthermore, when constructing the average traffic flow model, the optimal hyperparameter combination and input parameters of the model are determined through model fitting, reducing the model calculation difficulty and computational load, and enabling more timely feedback.

[0042] Finally, based on the verification of the calculation results, on the one hand, it is possible to discover the error in the result, and then track the problem to avoid equipment failure or model error. When building the entire intersection average traffic flow system, the equipment installation is set by reasonable and effective judgment of traffic condition level, reducing the number of unnecessary equipment installations and reducing costs. At the same time, the variables considered in the model in the platform can be set as needed, reducing the amount of calculation. The results are fed back in time and the maintenance is relatively simple. Attached Figure Description

[0043] Figure 1 This is a flowchart of an embodiment of the present invention;

[0044] Figure 2 This is a functional block diagram of an embodiment of the present invention. Detailed Implementation

[0045] The following detailed description illustrates the specific implementation method:

[0046] The basic implementation examples are as follows: Figure 1 As shown: the method for analyzing average traffic flow at intersections includes:

[0047] S1: Construct a traffic flow monitoring platform, deploy traffic flow monitoring equipment at each traffic intersection according to the preset equipment deployment method, and connect the traffic flow monitoring equipment to the traffic flow monitoring platform;

[0048] In this embodiment, in order to better monitor traffic flow data at traffic intersections, a traffic flow monitoring platform is first built that can receive, process, and visualize data. The service architecture of the traffic flow monitoring platform includes a front-end device layer, a platform service layer, and an application layer. The front-end device layer is used to connect to the traffic flow monitoring equipment installed at the traffic intersection and receive the data transmitted by the traffic flow monitoring equipment. The platform service layer is used to process and analyze the received data. The application layer is used to visualize or store the processed and analyzed data.

[0049] In this application, to address the issues of high equipment installation costs and difficult maintenance associated with existing intersection average traffic flow analysis technologies, a pre-defined equipment deployment method is adopted for the deployment of traffic flow monitoring equipment, specifically as follows:

[0050] Historical traffic data is extracted from the traffic flow monitoring platform. The main parameters considered in this data are traffic flow, vehicle type, and traffic congestion. Specifically, traffic flow refers to the historical traffic flow data for that intersection. Vehicle type refers to whether the vehicles traveling at the intersection are passenger cars, large buses, large trucks, or articulated vehicles, as the proportion of traffic flow varies depending on the vehicle type. Traffic congestion includes the duration and type of congestion at the intersection. After extracting these three parameters from the historical traffic data, the traffic flow monitoring platform also deploys map information. The map displays the traffic conditions of each intersection at different times of the day, providing a highly intuitive and effective representation of the traffic situation.

[0051] In order to effectively obtain the installation status of monitoring equipment at different traffic intersections, this application first marks the traffic intersections on the map according to the traffic congestion situation. Specifically, when a traffic intersection is congested at a certain time, it is marked on the map by means of an electronic fence. The diameter of the electronic fence is set according to the traffic operation status level of the traffic intersection.

[0052] Regarding the determination of traffic condition levels, this application uses historical traffic data to calculate the free-flow speed at marked intersections at preset time intervals based on traffic flow. Specifically:

[0053] 1. Obtain traffic data from 6:00 to 24:00 every day for 30 days, sample every 15 minutes, calculate the arithmetic mean of the average travel speed, and obtain the average travel speed samples for each time interval of the day.

[0054] 2. Sort the samples from highest to lowest based on the average driving speed monitored over the entire 30-day period, and select the top 1 / 9 of the samples;

[0055] 3. Calculate the arithmetic mean of the selected samples as the free-flow velocity. If the free-flow velocity is greater than the road speed limit, then the road speed limit shall be used.

[0056] To obtain the average driving speed, based on traffic flow and vehicle type data from historical traffic data, target vehicles within each vehicle type are identified. The length and time difference of the trajectory of the identified target vehicles at different cycles before and after the marked intersection are obtained, and the average driving speed at the marked intersection is calculated. Specifically:

[0057] 1. By randomly selecting vehicle types at traffic intersections, the selected target vehicles are identified and the same parameters sampled during free-flow velocity calculation are used to obtain the target vehicles' driving speed, driving trajectory, and location information.

[0058] 2. Based on the target vehicle's trajectory, filter and monitor vehicles passing through the traffic intersection section, and calculate the vehicle trajectory length and time difference;

[0059] 3. Calculate the average driving speed of the road segment. Average driving speed = cumulative length of trajectory of each target vehicle / cumulative travel time of each target vehicle.

[0060] Based on the free-flow velocity and average driving speed obtained above, a comparison can be made to obtain the traffic operation status level for each time interval. Specifically:

[0061] Smooth flow: Average travel speed > 70% of free-flow speed;

[0062] Basic smooth flow: 70% > average travel speed > free flow speed 50%;

[0063] Mild congestion: 50% > average travel speed > free-flow speed 40%;

[0064] Moderate congestion: 40% > average travel speed > free-flow speed 30%;

[0065] Severe congestion: 30% > average travel speed;

[0066] The following table, Table 1, provides a detailed explanation of the traffic operation status levels:

[0067] Table 1 Classification of Traffic Operation Status

[0068]

[0069] In Table 1 above, V kj V represents the average driving speed. f This indicates the free-flow velocity.

[0070] Therefore, the diameter of the electronic fences deployed at traffic intersections is determined according to the traffic operation status classification table, based on the traffic congestion situation. The diameter of the electronic fence at a smooth intersection is 0m, the diameter at a basically smooth intersection is 10m, the diameter at a slightly congested intersection is 15m, the diameter at a moderately congested intersection is 18m, and the diameter at a severely congested intersection is 20m.

[0071] Similarly, the deployment of traffic flow monitoring equipment at intersections is determined based on the obtained traffic condition level. Specifically, if more than 80% of the sampled data indicates smooth traffic flow, then one monitoring device for real-time traffic imaging and one monitoring device connected to vehicle terminals will be deployed at each intersection, depending on the intersection's shape. If more than 80% of the sampled data indicates basically smooth traffic flow, then one primary imaging monitoring device, one auxiliary imaging monitoring device, and one monitoring device connected to vehicle terminals will be deployed at each intersection, depending on the intersection's shape. If more than 80% of the sampled data indicates mild congestion, then... At each intersection, 1-2 primary monitoring devices, 2-3 secondary monitoring devices, and 2-3 monitoring devices connected to vehicle terminals will be deployed. When the traffic congestion at the intersection accounts for more than 80% of the sampled data, then, depending on the shape of the intersection, 3-4 primary monitoring devices, 2-3 secondary monitoring devices, and 3-4 monitoring devices connected to vehicle terminals will be deployed at each intersection. When the traffic congestion at the intersection accounts for more than 80% of the sampled data, then, depending on the shape of the intersection, more than 4 primary monitoring devices, more than 3 secondary monitoring devices, and more than 4 monitoring devices connected to vehicle terminals will be deployed at each intersection.

[0072] In the above-mentioned deployment of traffic flow monitoring equipment, the determination of traffic operation status is based on the ratio of the number of days with traffic operation status to the sampled data. When a traffic intersection experiences different congestion conditions at different times within a day, the required traffic flow monitoring equipment can be activated according to the operation status of that traffic intersection. The deployment and activation of traffic flow monitoring equipment in this application are flexible and can be adjusted in real time according to traffic operation conditions, which greatly improves the effectiveness and accuracy.

[0073] S2: Connecting the traffic flow monitoring equipment to vehicle terminals at each traffic intersection to obtain real-time and recent data from these terminals, as well as real-time and recent traffic flow data from the intersections, and transmitting the obtained real-time and recent vehicle terminal data, as well as the real-time and recent traffic flow data, to the traffic flow monitoring platform; S2 includes:

[0074] S2-1: Connect vehicle terminals within a preset range to the traffic flow monitoring equipment, receive data from the vehicle terminals at the current traffic intersection, and transmit it to the traffic flow monitoring platform;

[0075] S2-2: The traffic flow monitoring platform obtains the vehicle model corresponding to the vehicle terminal based on the transmitted vehicle terminal data and generates a vehicle conversion factor; the vehicle conversion factor is: 1.0 for passenger cars, 2.0 for large passenger cars, 2.5 for large trucks, and 3.0 for articulated vehicles.

[0076] S2-3: The traffic flow monitoring platform generates a data collection period based on the traffic operation status level of the traffic intersection, and extracts recent data and current real-time data from the data transmitted by vehicle terminals during the data collection period.

[0077] S2-4: Traffic flow monitoring equipment monitors the current traffic intersection, acquires recent traffic flow data and current real-time traffic flow data for the data collection period based on the current traffic operation status level, and transmits them to the traffic flow monitoring platform.

[0078] In this embodiment, the data collected by the deployed traffic flow monitoring equipment includes data obtained from vehicle terminal network connection and traffic condition data captured by camera. Specifically, the data obtained from vehicle terminal network connection includes real-time data and recent data from the connected vehicle terminal. The real-time data includes the vehicle model, vehicle trajectory, and vehicle speed of the currently connected vehicle terminal. The recent data includes some data of the currently connected vehicle terminal during recent driving, such as the driving section and speed in the 15 minutes before connecting to the traffic flow monitoring equipment. Through the recent data, the driving trend of vehicles flowing towards the traffic intersection can be known, and thus the existing traffic flow trend of the intersection during that time period can be obtained. The recent data is historical data, which can be transmitted to the traffic flow monitoring platform and used as a test set for the traffic average flow prediction model. The real-time data of the vehicle terminal can be used as a validation set for the traffic average flow prediction model.

[0079] Traffic data captured by traffic flow monitoring equipment includes real-time traffic data and recent traffic data. Recent traffic data represents the traffic flow at the intersection over a recent period, such as the traffic flow in the last 15 minutes. Recent traffic data is mainly used as a sample set for building a traffic average flow prediction model. During the training process, the completed traffic average flow prediction model is tested using recent data collected from vehicle terminals. When the test results meet the requirements, real-time traffic data is input into the traffic average flow prediction model for output. The output results are then evaluated based on real-time data obtained from the vehicle terminal network, thus effectively and accurately obtaining the current traffic average flow data for the intersection.

[0080] Therefore, it also includes S3: Constructing an average traffic flow model in the traffic flow monitoring platform, inputting recent traffic flow data from intersections into the average traffic flow model, training the average traffic flow model, then inputting recent vehicle terminal data into the trained average traffic flow model to verify the training results, until the verified average traffic flow model is output; wherein, S3 includes:

[0081] S3-1: Receives real-time and recent data from vehicle terminals, as well as real-time and recent traffic flow data from traffic intersections;

[0082] S3-2: Based on the recent traffic flow data received from traffic intersections, generate variables for traffic flow, traffic condition level, vehicle speed, and vehicle type according to traffic tags;

[0083] S3-3: Filter variables based on their missing values, missing rate, and IV value;

[0084] S3-4: The selected variables are used as a sample set, which is then divided into a training set and a test set. The optimal hyperparameter combination is selected using a grid search method. For each hyperparameter combination, a model is trained using the training set to obtain a model. The data values ​​of the model are calculated in the test set. The hyperparameter combination of the model with the best data values ​​in the test set is used as the final hyperparameter value. The variables used are used as the final input variables. The training set and the test set are then merged. Based on the final hyperparameter values, the final input variables, and the vehicle conversion coefficient, the model is fitted to generate the average traffic flow model.

[0085] S3-5: Collect recent data from vehicle terminals, extract vehicle speed and vehicle model from the collected recent data, generate verification traffic flow data, process the verification traffic flow data according to steps S3-2 to S3-3, input it into the average traffic flow model, evaluate the output results, and generate the final average traffic flow model.

[0086] S4: Input the real-time traffic flow data and vehicle terminal real-time data from the intersection into the trained average traffic flow model to generate the average traffic flow data for the current time period. S4 includes:

[0087] S4-1: Input the real-time traffic flow data of the traffic intersection into the final average traffic flow model and output the real-time average traffic flow result;

[0088] S4-2: Input the real-time data from the vehicle terminal into the final average traffic flow model and output the verification average traffic flow results;

[0089] S4-3: Compare the real-time average traffic flow result with the verification average traffic flow result. If the comparison result meets the requirements, the real-time average traffic flow result shall be used as the final average traffic flow analysis result of the traffic intersection.

[0090] In this embodiment, the construction, testing, and evaluation of the model can ensure the accuracy of the model's output results. On the other hand, by inputting different datasets, the generalization ability of the model can be expanded, so that the model can effectively output for different input parameters, thus ensuring the model's operational capability.

[0091] In summary, this application first constructs a traffic flow monitoring platform, which serves as the average traffic flow calculation terminal at traffic intersections. This platform can process the received data. Regarding the real-time performance and validity of the received data, this application improves upon this by deploying traffic flow monitoring equipment. This equipment collects real-time and recent traffic flow data from the traffic intersections and connects to vehicle terminals within a pre-defined area of ​​the intersection to receive real-time and recent data from these terminals. The acquisition of real-time and real-time traffic flow data ensures the real-time nature of the data source, while the acquisition of recent data and recent traffic flow data ensures the validity of the processed data.

[0092] Meanwhile, for the calculation of average traffic flow, this application considers different vehicle types, such as passenger cars and large trucks. Different vehicle types have different data on road travel. By configuring the vehicle conversion coefficient, the calculation results are more accurate. Furthermore, when constructing the average traffic flow model, the optimal hyperparameter combination and input parameters of the model are determined through model fitting, reducing the model calculation difficulty and computational load, and enabling more timely feedback.

[0093] Finally, based on the verification of the calculation results, on the one hand, it is possible to discover the error in the result, and then track the problem to avoid equipment failure or model error. When building the entire intersection average traffic flow system, the equipment installation is set by reasonable and effective judgment of traffic condition level, reducing the number of unnecessary equipment installations and reducing costs. At the same time, the variables considered in the model in the platform can be set as needed, reducing the amount of calculation. The results are fed back in time and the maintenance is relatively simple.

[0094] like Figure 2 As shown, in another embodiment of this embodiment, an intersection average traffic flow analysis system is also included, which is applied to the above-mentioned intersection average traffic flow analysis method. The system includes a traffic flow monitoring platform and a traffic flow monitoring device. The traffic flow monitoring platform and the traffic flow monitoring device are communicatively connected. The traffic flow monitoring device is deployed according to a preset device deployment method and is connected to vehicle terminals within a preset range at each traffic intersection.

[0095] The traffic flow monitoring device is used to collect real-time and recent traffic flow data at traffic intersections and transmit them to the traffic flow monitoring platform; the traffic flow monitoring device is also used to acquire real-time and recent data from vehicle terminals at each traffic intersection and transmit them to the traffic flow monitoring platform.

[0096] The traffic flow monitoring platform includes a model building module, a model verification module, and a data processing module. The model building module is used to build a traffic average flow model and input recent traffic flow data from traffic intersections into the traffic average flow model to train the traffic average flow model. The model verification module is used to input recent vehicle terminal data into the traffic average flow model to verify the training results of the traffic average flow model until the verification is successful and the trained traffic average flow model is output. The data processing module is used to input real-time traffic flow data from traffic intersections and real-time vehicle terminal data into the trained traffic average flow model to generate the traffic intersection average flow data for the current time period.

[0097] In addition to the methods and systems described above, embodiments of the present invention may also be computer program products, wherein the computer program products store programs or instructions that cause a computer to execute the intersection average traffic flow analysis method as described above.

[0098] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of the present invention. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0099] Furthermore, embodiments of the present invention may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps of the intersection average traffic flow analysis method provided in any embodiment of the present invention.

[0100] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0101] Example 2:

[0102] The difference between Example 2 and Example 1 is that Example 2 further includes: S5: obtaining historical traffic flow data of traffic intersections and determining the traffic flow level of traffic intersections; dividing the average traffic flow model into multiple levels based on the traffic flow level of traffic intersections, with each level corresponding to a traffic flow level of traffic intersections;

[0103] S6: Construct a level prediction model to obtain preliminary traffic flow information at traffic intersections. Determine the traffic flow level at the intersections using the level prediction model, and call the average traffic flow model corresponding to the traffic flow level when executing S4.

[0104] In this embodiment, during the training process of the traffic average flow model, outputting all types of traffic flow data through the traffic average flow model results in a long judgment time and slow processing efficiency. Therefore, this embodiment two sets up a multi-level traffic average flow model, for example, levels 1-4, 5-7, and 8-10. By using a multi-level traffic average flow model and setting up a level prediction model, after obtaining the preliminary traffic flow situation at the intersection, the level of the traffic flow is determined, and the appropriate traffic average flow model is called for analysis, thereby reducing the processing time of the model and improving efficiency.

[0105] The above are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for analyzing average traffic flow at intersections, characterized by: include: S1: Construct a traffic flow monitoring platform, deploy traffic flow monitoring equipment at each traffic intersection according to the preset equipment deployment method, and connect the traffic flow monitoring equipment to the traffic flow monitoring platform; S2: Connect the traffic flow monitoring equipment to the vehicle terminals at each traffic intersection to obtain real-time and recent data from the vehicle terminals at each traffic intersection, as well as real-time and recent traffic flow data from the traffic intersection through the traffic flow monitoring equipment, and transmit the obtained real-time and recent data from the vehicle terminals, as well as the real-time and recent traffic flow data, to the traffic flow monitoring platform. S3: A traffic flow average model is constructed in the traffic flow monitoring platform. Recent traffic flow data from intersections is input into the model, and the model is trained. Then, recent vehicle terminal data is input into the trained model to verify its training results, until the trained model is successfully verified and output. S3 includes: S3-1: Receives real-time and recent data from vehicle terminals, as well as real-time and recent traffic flow data from traffic intersections; S3-2: Based on the recent traffic flow data received from traffic intersections, generate variables for traffic flow, traffic operation status level, vehicle speed, and vehicle type according to traffic tags; S3-3: Filter variables based on their missing values, missing rate, and IV value; S3-4: The selected variables are used as a sample set, which is then divided into a training set and a test set. The optimal hyperparameter combination is selected using a grid search method. For each hyperparameter combination, a model is trained using the training set to obtain a model. The data values ​​of the model are calculated in the test set. The hyperparameter combination of the model with the best data values ​​in the test set is used as the final hyperparameter value. The variables used are used as the final input variables. The training set and the test set are then merged. Based on the final hyperparameter values, the final input variables, and the vehicle conversion coefficient, the model is fitted to generate the average traffic flow model. S3-5: Collect recent data from vehicle terminals, extract vehicle speed and vehicle model from the collected recent data, generate verification traffic flow data, process the verification traffic flow data according to steps S3-2 to S3-3, input it into the traffic average flow model, evaluate the output results, and generate the final traffic average flow model. S4: Input the real-time traffic flow data of the traffic intersection and the real-time data of the vehicle terminal into the trained average traffic flow model to generate the average traffic flow data of the traffic intersection for the current time period. It also includes S5: obtaining historical traffic flow data at traffic intersections and determining the traffic flow level at traffic intersections; dividing the average traffic flow model into multiple levels based on the traffic flow level at traffic intersections, with each level corresponding to a traffic flow level at a traffic intersection; S6: Construct a level prediction model to obtain preliminary traffic flow information at traffic intersections. Determine the traffic flow level at the traffic intersections through the level prediction model, and call the average traffic flow model corresponding to the traffic flow level when executing S4.

2. The intersection average traffic flow analysis method according to claim 1, characterized in that: In step S1, the traffic flow monitoring equipment is deployed at each traffic intersection according to a preset equipment deployment method, specifically as follows: Obtain historical traffic data stored in the traffic flow monitoring platform, and extract traffic flow, vehicle type, and traffic congestion information from the historical traffic data; Extract the map stored in the traffic flow monitoring platform, and display the extracted traffic flow, vehicle type and traffic congestion in the map in three dimensions; Based on historical traffic data, traffic intersections are marked on the map. Based on traffic flow data from historical traffic data, the free-flow speed at marked traffic intersections is calculated at preset time intervals. Based on traffic flow and vehicle type in historical traffic data, target vehicles in the vehicle type are identified, and the trajectory length and time difference of the identified target vehicles in different cycles before and after the marked traffic intersection are obtained. The average speed of the marked traffic intersection is then calculated. The average driving speed is compared with the free flow speed to generate a traffic operation status level at each time interval, and traffic flow monitoring equipment is deployed according to the traffic operation status level.

3. The intersection average traffic flow analysis method according to claim 2, characterized in that: S2 includes: S2-1: Connect vehicle terminals within a preset range to the traffic flow monitoring equipment, receive data from the vehicle terminals at the current traffic intersection, and transmit it to the traffic flow monitoring platform; S2-2: The traffic flow monitoring platform obtains the vehicle model corresponding to the vehicle terminal based on the transmitted vehicle terminal data and generates the vehicle conversion coefficient. S2-3: The traffic flow monitoring platform generates a data collection period based on the traffic operation status level of the traffic intersection, and extracts recent data and current real-time data from the data transmitted by vehicle terminals during the data collection period. S2-4: Traffic flow monitoring equipment monitors the current traffic intersection, acquires recent traffic flow data and current real-time traffic flow data for the data collection period based on the current traffic operation status level, and transmits them to the traffic flow monitoring platform.

4. The intersection average traffic flow analysis method according to claim 3, characterized in that: S4 includes: S4-1: Input the real-time traffic flow data of the traffic intersection into the final average traffic flow model and output the real-time average traffic flow result; S4-2: Input the real-time data from the vehicle terminal into the final average traffic flow model and output the verification average traffic flow results; S4-3: Compare the real-time average traffic flow result with the verification average traffic flow result. If the comparison result meets the requirements, the real-time average traffic flow result shall be used as the final average traffic flow analysis result of the traffic intersection.

5. The intersection average traffic flow analysis method according to claim 4, characterized in that: The vehicle conversion factors are as follows: 1.0 for passenger cars, 2.0 for large passenger cars, 2.5 for large trucks, and 3.0 for articulated vehicles.

6. An intersection average traffic flow analysis system, applied to the intersection average traffic flow analysis method according to any one of claims 1-5, characterized in that: It includes a traffic flow monitoring platform and traffic flow monitoring equipment, which are communicatively connected. The traffic flow monitoring equipment is deployed according to a preset equipment deployment method and is networked with vehicle terminals at each traffic intersection within a preset range. The traffic flow monitoring device is used to collect real-time and recent traffic flow data at traffic intersections and transmit them to the traffic flow monitoring platform; the traffic flow monitoring device is also used to acquire real-time and recent data from vehicle terminals at each traffic intersection and transmit them to the traffic flow monitoring platform. The traffic flow monitoring platform includes a model building module, a model verification module, and a data processing module. The model building module is used to build a traffic average flow model and input recent traffic flow data from traffic intersections into the traffic average flow model to train the traffic average flow model. The model verification module is used to input recent vehicle terminal data into the traffic average flow model to verify the training results of the traffic average flow model until the verification is successful and the trained traffic average flow model is output. The data processing module is used to input real-time traffic flow data from traffic intersections and real-time vehicle terminal data into the trained traffic average flow model to generate the traffic intersection average flow data for the current time period.

7. A storage medium, characterized in that: The storage medium stores a program or instructions that cause a computer to execute the intersection average traffic flow analysis method according to any one of claims 1-5.

Images