A traffic bearing capacity auxiliary analysis method and system based on CUBE data

Abstract

The application relates to a traffic bearing capacity auxiliary analysis method and system based on CUBE data; the method comprises the following steps: establishing a traffic flow prediction model; collecting traffic flow data on a network and performing traffic flow data preprocessing on the network; organizing and storing the collected and preprocessed traffic flow data; taking the processed traffic flow data as an input source of the traffic flow prediction model, predicting real-time traffic flow through the traffic flow prediction model, and warning and analyzing abnormal traffic flow data. The average driving speed, vehicle density, space occupancy and time occupancy in the road section are calculated, the traffic condition in the road section can be directly observed, and timely observation can be performed when an accident occurs in the road section.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of traffic carrying capacity simulation, in particular to a traffic carrying capacity auxiliary analysis method and system based on CUBE data. BACKGROUND

[0002] Traffic carrying capacity analysis is a prediction analysis of traffic flow of road sections, toll gates and road networks, which is of great significance for realizing intelligent traffic information systems, formulating reasonable traffic security strategies and traffic management and guidance measures, and relieving road traffic congestion. At present, technological innovation is becoming more and more intense, and intelligentization and digitization have become the direction of social progress. Especially with the development of electric vehicles, driverless vehicles and intelligent toll gates, a large amount of heterogeneous traffic flow information will be generated. This has higher requirements for the real-time performance and prediction accuracy of intelligent traffic systems. Therefore, it is becoming more and more important to study the real-time traffic information system for analyzing road traffic carrying capacity and planning real-time traffic routes for thousands of vehicles according to real-time traffic conditions.

[0003] Current traffic data collection equipment is quite mature, and various intelligent three-dimensional toll gate systems exist at road entrances and exits, toll stations, inter-provincial toll gates, etc. Road information collection systems can collect various information of vehicles such as image data and speed data, and can further analyze detailed information of vehicles such as image data according to the collected data. For example, the resolution of a high-resolution digital camera used in a toll gate is 2 million pixels, which can completely show the license plate, vehicle body and driver condition of a vehicle. Through analysis, various information such as vehicle type and vehicle color can be obtained.

[0004] However, the development of road information collection brings a large amount of available data, but a road network will generate TB-level traffic data every day, and the existing traffic carrying capacity analysis system cannot intuitively observe the traffic conditions in the road section, which increases the difficulty of analyzing and storing the road traffic carrying capacity. SUMMARY

[0005] In order to solve the above technical defects existing in the prior art, the present application provides a traffic carrying capacity auxiliary analysis method and system based on CUBE data, which can effectively solve the problems in the background art.

[0006] In order to solve the above technical problems, the technical scheme provided by the present application is as follows:

[0007] The embodiment of the present application discloses a traffic carrying capacity auxiliary analysis method based on CUBE data, which comprises the following steps:

[0008] Step 1: Establish a traffic flow prediction model;

[0009] Step two: collect traffic flow data on the network and preprocess the traffic flow data on the network;

[0010] Step three: organize and store the collected and preprocessed traffic flow data;

[0011] Step four: use the processed traffic flow data as the input source of the traffic flow prediction model, predict the real-time traffic flow through the traffic flow prediction model, and prewarn and analyze the abnormal traffic flow data.

[0012] In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula The traffic flow prediction model is calculated by the formula

[0013] In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula The traffic flow prediction model is calculated by the formula

[0014] In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula The traffic flow prediction model is calculated by the formula

[0015] In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula The traffic flow prediction model is calculated by the formula k The traffic flow prediction model is calculated by the formula i In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula

[0016] In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula The traffic flow prediction model is calculated by the formula s The traffic flow prediction model is calculated by the formula i In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula

[0017] In any of the above solutions, preferably, the traffic flow prediction model is calculated by the formula The traffic flow prediction model is calculated by the formula

[0018] In any of the above solutions, it is preferred that a warning is issued when the traffic flow prediction model predicts that the vehicle density, the space occupancy rate in the road section and the time occupancy rate in the road section are greater than a threshold value.

[0019] In any of the above solutions, it is preferred that the modeling step of the traffic flow prediction model comprises:

[0020] Step 1: stationary test is performed on the original flow sequence by using ADF unit root test method, if the original flow sequence is not stationary, then difference transformation is performed until the new sequence passes the stationary test;

[0021] Step 2: the order of the model is determined by using AIC criterion, i.e. the value of p in the AR(p) model is determined, and the least square method is used for parameter estimation;

[0022] Step 3: the determined model is used for traffic flow prediction;

[0023] Step 4: when new sampling data is generated, steps 1 to 3 are repeated to realize dynamic prediction of the flow.

[0024] In any of the above solutions, when the stationary test is performed on the original flow sequence by using ADF unit root test method, the formula is used to calculate the absolute percentage error of the prediction model; wherein x i is the predicted value, and x i is the actual value.

[0025] In any of the above solutions, when the stationary test is performed on the original flow sequence by using ADF unit root test method, the formula is used to calculate the mean absolute percentage error of the prediction model; wherein N is the number of predictions.

[0026] In any of the above solutions, when the traffic flow data is collected, the traffic flow data is collected by using one or more of the following technologies: coil detection technology, radar detection technology, infrared detection technology, image detection technology, floating car detection technology, vehicle recognition system, intermodulation station detection system and toll station detection system.

[0027] In any of the above schemes, preferably, the coil detection technology is mainly achieved by underground spaced induction coils. When a vehicle passes through the coil buried road section, the magnetic flux line is cut off, generating an electric current in the coil, and then the system analyzes the collected current signal to obtain vehicle speed and traffic flow data. This coil detection method can be widely applied, mainly because the technical principle is simple, and since it is buried underground, it is not easily disturbed by the outside world, and the accuracy is high. However, the underground buried method is not conducive to equipment maintenance once it is damaged.

[0028] In any of the above schemes, preferably, the radar detection technology is based on the principle that the detection device transmits radar waves to the road surface, and when there is a vehicle, the vehicle will block and reflect the radar waves. The detection device collects and calculates the reflected waves to obtain vehicle speed, traffic flow and other information.

[0029] In any of the above schemes, preferably, the infrared detection technology is achieved by an infrared detector, which is divided into active infrared detectors and passive infrared detectors. The active infrared detector uses a laser diode to actively emit laser light, which is detected by energy reflection to calculate vehicle speed and profile. The passive infrared detector uses a thermocouple sensor or a pyroelectric sensor to receive the heat radiation emitted by the vehicle engine and convert it into an electrical signal. According to multiple signals, vehicle speed information is calculated. The infrared detector has insufficient accuracy for high-speed vehicle speed detection, and compared to the radar detector, although the principle is similar, it is not suitable for rainy and snowy weather.

[0030] In any of the above schemes, preferably, the image detection technology is achieved by an image detector, which uses a camera to collect images of road traffic intersections or road sections and transmits the images to an image analysis device. Through image processing technology, the speed, shape and lane change information of the vehicle are obtained. Compared to other detection methods, the image detection method can obtain more information about the vehicle relying on image processing technology, but it has high technical bottlenecks, limited camera life, and is easily disturbed by the external environment.

[0031] In any of the above schemes, preferably, the floating car detection technology uses GPS-equipped city public vehicles to travel information, applies path estimation, map matching and other calculation models to match vehicles with road space and time, and calculates traffic condition information. The floating car technology is an important means of obtaining road traffic condition information in the current intelligent transportation system.

[0032] In any of the above schemes, preferably, when storing and organizing traffic flow data, the main road is used as the main node to organize traffic flow data.

[0033] The second aspect is a traffic carrying capacity auxiliary analysis system based on CUBE data, comprising:

[0034] A generation module is configured to establish a traffic flow prediction model.

[0035] A collection module is configured to collect traffic flow data on a network and perform traffic flow data preprocessing on the network.

[0036] A storage module is configured to organize and store the collected and preprocessed traffic flow data.

[0037] An analysis module is configured to use the preprocessed traffic flow data as an input source of the traffic flow prediction model, predict real-time traffic flow through the traffic flow prediction model, and perform early warning and analysis on abnormal traffic flow data.

[0038] Compared with the prior art, the present application has the following advantages:

[0039] 1. The present application divides the traffic road into different road sections, reduces the calculation workload of the traffic flow prediction model, and increases the prediction accuracy of the prediction model.

[0040] 2. The present application can intuitively observe the traffic conditions in the road section by calculating the average driving speed, vehicle density, space occupancy rate and time occupancy rate in the road section, and can timely observe when an accident occurs in the road section.

[0041] 3. The present application can collect accurate traffic flow data by using the combination of radar detection technology, image detection technology, intermodulation station detection technology and toll station detection technology, thereby increasing the prediction accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0042] The accompanying drawings are used to further understand the present application, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application.

[0043] Figure 1 is a flow chart of the traffic carrying capacity auxiliary analysis method based on CUBE data provided by the embodiments of the present application.

[0044] Figure 2 is a schematic diagram of the traffic carrying capacity auxiliary analysis system module based on CUBE data provided by the embodiments of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0046] It should be noted that when an element is referred to as being "fixed" or "set" on another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or indirectly connected to the other element.

[0047] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0048] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0049] In order to better understand the above technical solutions, the technical solutions of the present application will be described in detail below in conjunction with the drawings and specific embodiments of the specification.

[0050] As shown in the figure, a traffic carrying capacity auxiliary analysis method based on CUBE data includes the following steps: Figure 1 Step 1: Establish a traffic flow prediction model;

[0051] Step 2: Collect traffic flow data on the network and preprocess traffic flow data on the network;

[0052] Step 3: Organize and store the collected and preprocessed traffic flow data;

[0053] Step 4: Use the processed traffic flow data as the input source of the traffic flow prediction model, predict the real-time traffic flow through the traffic flow prediction model, and analyze the abnormal traffic flow data.

[0054] Specifically, the traffic flow prediction model predicts the traffic volume of different lanes in the statistical road section through the formula

[0055] ; wherein q is the traffic volume, i is the lane number; T is the collection period, N is the number of cars counted in T collection period.​

[0056] Preferably, the traffic prediction model is derived from the formula... Calculate the average driving speed within the observation period; where V is the average speed within the period, M is the number of vehicles within the period, and n is the vehicle number within the period.

[0057] Preferably, the traffic prediction model is derived from the formula... Calculate the vehicle density within the calculation period; where N is the number of vehicles counted within the T collection period, and L is the road length within the statistical road interval.

[0058] Preferably, the traffic prediction model is derived from the formula... Calculate the space occupancy rate within the statistical road interval; where Z k To calculate the space occupancy rate within a road section, L represents the road length within that section, N represents the number of vehicles counted during the T collection period, and l i Let be the length of the i-th vehicle.

[0059] Preferably, the traffic prediction model is derived from the formula... Calculate the time occupancy rate within the statistical road interval; where Z s To calculate the time occupancy rate within a road segment, T is the length of the data collection period, N is the number of vehicles counted within the T data collection period, and t i Let be the time taken for the i-th vehicle to pass through the monitored road segment.

[0060] Preferably, the traffic prediction model statistically analyzes the probability of vehicles entering a road section within an adjacent road section based on traffic flow data, and defines it as the coherence coefficient of the road section.

[0061] Preferably, the traffic flow prediction model estimates the number of vehicles leaving the road section by measuring the vehicle density and speed of the road section, and estimates the number of vehicles entering the road section by measuring the vehicle data leaving the road section from adjacent road sections using the coherence coefficient of the road section, and then predicts the vehicle density.

[0062] Preferably, when the traffic flow prediction model predicts that the vehicle density, the spatial occupancy rate within the road section, and the time occupancy rate within the road section are greater than the threshold, an early warning is issued.

[0063] The modeling of the traffic flow prediction model includes the following steps:

[0064] Step 1: Use the ADF unit root test method to test the stationarity of the original flow series. If the original flow series is not stationary, perform a difference transformation until the new series passes the stationarity test.

[0065] Step two: use AIC criterion to determine the order of the model, that is, determine the value of p in AR(p) model, and use least square method to estimate parameters;

[0066] Step three: use the determined model to predict traffic flow;

[0067] Step four: when new sampling data is generated, repeat steps one to three to realize dynamic prediction of traffic flow.

[0068] As preferred, when using ADF unit root test method to test the stationarity of the original flow sequence, the formula The absolute percentage error of the prediction model is calculated; wherein, x i is the predicted value, and x i is the actual value.

[0069] As preferred, when using ADF unit root test method to test the stationarity of the original flow sequence, the formula The mean absolute percentage error of the prediction model is calculated; wherein, N is the number of predictions.

[0070] As preferred, when collecting traffic flow data, one or more of coil detection technology, radar detection technology, infrared detection technology, image detection technology, floating car detection technology, vehicle recognition system, intermodulation station detection system and toll station detection system is used to collect traffic flow data.

[0071] As preferred, the coil detection technology is mainly realized by underground interval buried inductive coils. When a vehicle passes through the coil buried road section, it will cut off the magnetic flux line, so that an electric current is generated in the coil. Then, the system analyzes the collected current signal to obtain vehicle speed and traffic flow data. This coil detection method can be widely applied, mainly because the technical principle is simple, and since it is buried underground, it is not easy to be disturbed by the outside world, and the accuracy is high. However, the underground buried method is not conducive to equipment maintenance once damaged.

[0072] As preferred, the principle of radar detection technology is that the detection device transmits radar waves to the road surface. When there is a vehicle, the vehicle will block and reflect the radar waves. The detection device collects the reflected waves and calculates the vehicle speed, traffic flow and other information.

[0073] As preferred, the infrared detection technology is realized by an infrared detector, which is divided into an active infrared detector and a passive infrared detector, the active infrared detector uses a laser diode to actively emit laser, calculates the vehicle speed and profile through the detection of energy reflection, the passive infrared detector uses a thermocouple sensor or a pyroelectric sensor to receive the heat radiation emitted by the engine of the vehicle and converts it into an electric signal, calculates the vehicle speed information according to multiple signals, the infrared detector has insufficient detection accuracy for high-speed moving vehicles, and compared with the radar detector, although the principles are similar, it is not suitable for rainy and snowy weather.

[0074] As preferred, the image detection technology is realized by an image detector, which uses a camera to collect the image of a road traffic intersection or section, and transmits the image to an image analysis device, obtains the speed, profile and lane change information of the vehicle through image processing technology, the image detection method can obtain more information of the vehicle relying on image processing technology, but has high technical bottlenecks, the service life of the camera is limited, and it is easily disturbed by the external environment.

[0075] As preferred, the floating car detection technology uses a city public vehicle equipped with a GPS to travel information, applies a path estimation, map matching and other calculation models to match the vehicle with the road space and time, and calculates the traffic condition information, the floating car technology is an important means to obtain road traffic condition information in the current intelligent transportation system.

[0076] As preferred, the vehicle recognition system is a comprehensive traffic information collection system that uses multiple traffic information collection technologies to realize traffic flow data, traffic congestion condition analysis, license plate information collection and vehicle information judgment, the vehicle recognition system is usually applied to highways, city road monitoring, parking lots and the like, the vehicle recognition system has high recognition accuracy, wide detection range and high automation degree, greatly improves the collection degree of traffic information, but the vehicle recognition system integrated with multiple functions has high cost and is easily disturbed by the external environment.

[0077] As preferred, the interchange station detection system is composed of a series of interchange stations, collects traffic road information through a dedicated interchange detection device, and finally transmits the information to an interchange data center through a network.

[0078] As preferred, the radar detection technology, the image detection technology, the interchange station detection technology and the toll station detection technology are combined to collect traffic flow data.

[0079] As preferred, when storing and organizing the traffic flow data, the traffic flow data is organized with main roads as main nodes.

[0080] As Figure 2 The application also provides a traffic carrying capacity auxiliary analysis system based on CUBE data, which comprises:

[0081] a generating module for establishing a traffic flow prediction model;

[0082] a collecting module for collecting traffic flow data on the network and pre-processing the traffic flow data on the network;

[0083] a storage module for organizing and storing the collected and pre-processed traffic flow data;

[0084] an analysis module for taking the pre-processed traffic flow data as an input source of the traffic flow prediction model, predicting real-time traffic flow through the traffic flow prediction model, and warning and analyzing abnormal traffic flow data.

[0085] Compared with the prior art, the application has the beneficial effects that:

[0086] 1. The application divides the traffic road into different road sections, reduces the calculation workload of the traffic flow prediction model, and increases the prediction accuracy of the prediction model.

[0087] 2. The application can intuitively observe the traffic conditions in the road section by calculating the average driving speed, vehicle density, space occupancy rate and time occupancy rate in the road section, and can timely observe when an accident occurs in the road section.

[0088] 3. The application can collect accurate traffic flow data by using the combination of radar detection technology, image detection technology, intermodulation station detection technology and toll station detection technology to collect traffic flow data, thereby increasing the prediction accuracy.

[0089] The above is only the preferred embodiment of the application and is not used to limit the application, although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or make equivalent replacement for some technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application shall be included in the protection scope of the application.

Claims

1. A method based on CUBE The data-assisted traffic carrying capacity analysis method is characterized by: Includes the following steps: Establish traffic flow prediction models; Collect traffic flow data from the network and perform traffic flow data preprocessing on the network; Organize and store the collected and preprocessed traffic flow data; The processed traffic flow data is used as the input source for the traffic flow prediction model. The traffic flow prediction model is used to predict real-time traffic flow and to provide early warnings and analysis for abnormal traffic flow data. The traffic flow prediction model uses the formula Traffic volume is predicted for different lanes within the statistical road section; among them, q For traffic volume, i Number the lanes; T For the collection period, N for T The number of cars counted during the data collection period; The traffic flow prediction model uses the formula Calculate the vehicle density within the calculation period; where, N for T Collect the number of cars counted during the week. L To calculate the road length within a road section, the traffic flow prediction model uses the formula... Calculate the space occupancy rate within the statistical road section; among which, Z k To calculate the space occupancy rate within road sections, L To calculate the length of roads within a road section, N for T Collect the number of cars counted during the week. li For the first i The length of the vehicle body; The traffic flow prediction model uses the formula Calculate the time occupancy rate within the statistical road interval; among which, Z s To calculate the time occupancy rate within road sections, T The length of the data collection period. N for T The number of cars counted during the data collection period. t i For the first i The time it takes for a vehicle to pass through the monitored section of road; The traffic flow prediction model statistically analyzes the probability of vehicles entering adjacent road sections based on traffic flow data and defines it as the coherence coefficient of the road section. The model estimates the number of vehicles leaving the road section based on vehicle density and speed, and estimates the number of vehicles entering the road section based on the data of vehicles leaving adjacent road sections using the coherence coefficient. It then predicts the vehicle density. A combination of radar detection, image detection, traffic control station detection, and toll station detection technologies is used to collect traffic flow data. When storing and organizing traffic flow data, main roads are used as the primary nodes. An early warning is issued when the traffic flow prediction model predicts that vehicle density, spatial occupancy within the road section, and time occupancy within the road section exceed a certain threshold.

2. The method based on claim 1 CUBE The data-assisted traffic carrying capacity analysis method is characterized by: The modeling steps of the traffic flow prediction model include: Step 1: Using ADF The unit root test method is used to test the stationarity of the original flow series. If the original flow series is not stationary, a difference transformation is performed until the new series passes the stationarity test. Step Two: Application AIC The criterion determines the order of the model, that is, determines... AR ( p In the model p The value of is estimated using the least squares method; Step 3: Use the established model to predict traffic flow; When new sampling data is generated, repeat steps one through three to achieve dynamic prediction of traffic flow.

3. The method based on claim 2 CUBE The data-assisted traffic carrying capacity analysis method is characterized by... In utilizing ADF When using the unit root test to test the stationarity of the original flow series, the formula is used. Calculate the absolute percentage error of the prediction model; in, x i ’ For predicted values, x i This is the actual value.

4. The method based on claim 3 CUBE The data-assisted traffic carrying capacity analysis method is characterized by: In use ADF When using the unit root test to test the stationarity of the original flow series, the formula is used. The mean absolute percentage error of the prediction model is calculated; where, N For the number of predictions.

5. The method based on any one of claims 1 to 4 CUBE The data-driven traffic carrying capacity auxiliary analysis system is characterized by: include: The generation module is used to build traffic flow prediction models; The data acquisition module is used to collect traffic flow data from the network and to preprocess the traffic flow data from the network. The storage module is used to organize and store the collected and preprocessed traffic flow data; The analysis module uses the processed traffic flow data as the input source for the traffic flow prediction model, predicts real-time traffic flow through the traffic flow prediction model, and provides early warnings and analysis for abnormal traffic flow data.

Images