Expressway accident congestion position detection method based on ETC big data

Abstract

The invention discloses a highway accident congestion position detection method based on ETC big data, and the method comprises the steps: collecting ETC transaction data of all toll stations on a highway, carrying out the data cleaning, and matching the passing record of a vehicle at an adjacent toll station according to the ETC transaction data; segmenting the ETC transaction data according to a fixed time interval, and respectively calculating the vehicle average passing time, the vehicle average speed and the traffic flow of each time period; calculating the flow change rate and the speed change rate in the current time period t, and comparing the flow change rate and the speed change rate with corresponding threshold values to judge whether a traffic accident occurs or not; calculating an accident position by using an accident position estimation model; the congestion length is calculated based on the average passing time of normal passing, and the starting and ending positions of congestion are determined in combination with the accident position. According to the invention, quick response to accidents is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of highway traffic management, and in particular to a method for detecting highway accident congestion locations based on ETC big data. Background Art

[0002] Frequent traffic accidents and congestion on highways pose a serious challenge to public safety and smooth traffic flow. Traditional detection methods are insufficient to address these complex issues. However, the widespread use of Electronic Toll Collection (ETC) systems has opened up a new avenue for accident and congestion detection. ETC systems accurately record the timestamps and location information of vehicles passing through toll booths, providing valuable data support for in-depth analysis of traffic flow and speed. They can monitor vehicle operating status in real time and promptly detect abnormal fluctuations in traffic flow, thereby accurately pinpointing the location of accidents and congestion on highways.

[0003] However, the field of highway accident and congestion detection has long faced challenges such as data imbalance and complex accident causes. In recent years, many scholars have introduced cutting-edge technologies and methods to provide new ideas for solving these problems. Existing technologies have the following technical disadvantages: (1) Traditional accident detection methods, such as manual reporting and video surveillance, often face the problems of slow information transmission and limited coverage. This means that once an accident occurs, relevant information may need to go through multiple links before it can be transmitted to the emergency response department, thereby delaying rescue and processing time. This delay not only affects the efficiency of accident handling, but may also lead to secondary accidents. (2) Accident detection methods that rely on surveillance cameras and roadside sensing equipment require a large amount of hardware investment and continuous maintenance costs. This high economic burden puts many local governments and traffic management departments under great pressure to achieve full road network coverage, especially when funds are tight. The high cost makes it impossible for some areas to achieve comprehensive monitoring and rapid response. (3) Although the electronic toll collection (ETC) system has been widely used in many areas and has generated a large amount of data, its use in accident detection is still insufficient. Due to the lack of effective data analysis and processing methods, the potential value of this data has not been fully realized. Effective use of this data will help improve the predictive capabilities and response speed of accident detection. (4) Existing accident detection methods are insufficient in the accuracy of accident location, making it difficult to meet the needs of precise highway management and emergency response. Information such as the specific location of the accident, the scope of impact, and its severity is often difficult to obtain accurately. This not only affects the efficiency of emergency response, but may also lead to waste of resources and unnecessary traffic congestion.

[0004] While current research and application have achieved some success, they primarily focus on estimating traffic parameters using floating vehicle data or GPS data, and the in-depth application of ETC data in highway accident detection remains insufficient. Furthermore, ETC data presents challenges such as large data volumes, high real-time requirements, and high levels of data noise. The efficient use of ETC big data to accurately locate accident locations remains a pressing issue. Summary of the Invention

[0005] The purpose of the present invention is to provide a highway accident congestion location detection method based on ETC big data, which is used to detect other types of traffic abnormal events and achieve rapid response to accidents.

[0006] The technical solution adopted in the present invention is:

[0007] A method for detecting highway accident congestion locations based on ETC big data includes the following steps:

[0008] Step 1: Data acquisition and preprocessing: After collecting and cleaning ETC transaction data from each toll station on the highway, the ETC transaction data is matched with the vehicle's travel records at adjacent toll stations to calculate the travel time of each vehicle passing through the road section.

[0009] Furthermore, the ETC transaction data in step 1 includes the timestamp of the vehicle passing through the toll station, the geographical location information of the toll station, and vehicle identification information (such as license plate number, vehicle model, etc.).

[0010] Furthermore, in step 1, traffic accident reports corresponding to the study time period are obtained synchronously to verify the accuracy of the model.

[0011] Furthermore, data cleaning in step 1 includes removing incomplete or erroneous records, unifying the time format, and correcting timestamp errors.

[0012] Step 2: Segment the ETC transaction data at fixed time intervals and calculate the average vehicle travel time, average vehicle speed, and traffic flow in each time period.

[0013] Furthermore, step 2 specifically includes the following steps:

[0014] Step 2-1, segment the data into statistics at a fixed time interval Δt (e.g., every 5 minutes or 15 minutes);

[0015] Step 2-2, calculate the average travel time T of vehicles in time period t t :

[0016]

[0017] Where n is the number of vehicles passing through in time period t, Ti is the travel time of the i-th vehicle.

[0018] Step 2-3, calculate the average speed V of the vehicle in time period t t :

[0019]

[0020] Wherein, L is the length of the road section between adjacent toll stations;

[0021] Step 2-4, calculate the traffic flow F in time period t t :

[0022]

[0023] Step 3, anomaly detection: calculate the flow rate change rate and speed change rate in the current time period t respectively, and compare them with the corresponding thresholds to determine whether there is a traffic accident;

[0024] Furthermore, step 3 specifically includes the following steps:

[0025] Step 3-1, calculate the rate of change of flow and speed:

[0026]

[0027]

[0028] Step 3-2, setting the flow rate and speed drop thresholds α and β respectively based on historical data or industry standards;

[0029] Step 3-3, determining whether the decrease values of the flow rate and speed are both less than the set threshold value; if so, determining that an abnormal event occurred during the time period; otherwise, determining that no abnormal event occurred during the time period;

[0030] Specifically, if ΔF≤-α and ΔV≤-β are both satisfied, it is considered that an abnormal event has occurred during the time period, which may be caused by a traffic accident.

[0031] Step 4: Calculate the accident location using the accident location estimation model. The specific calculation expression is:

[0032]

[0033] Where L is the length of the road section between adjacent toll stations; V t is the average speed in time period t; V t-1 is the average speed in time period t-1; T t is the average travel time in time period t; d is the distance from the accident point to the starting toll station.

[0034] Specifically, the variable definition is: L: the length of the road section between adjacent toll booths. t : Average speed in time period t. V t-1 : Average speed in time period t-1. T t : Average travel time in time period t. d: Distance between the accident point and the starting toll station. Accident location calculation: Based on the travel time of vehicles before and after the accident, establish the equation: Then solve the accident location d:

[0035] Step 5: Calculate the congestion length based on the average travel time of normal traffic, and determine the starting and ending locations of the congestion in combination with the accident location;

[0036] Furthermore, step 5 specifically includes the following steps:

[0037] Step 5-1: Calculate the average travel time T under normal circumstances 正常 , the specific calculation expression is:

[0038]

[0039] Where L is the length of the road section between adjacent toll stations; V t-1 is the average speed in time period t-1

[0040] Step 5-2, calculate the congestion length L 拥堵 , the specific calculation expression is:

[0041]

[0042] Among them, F t is the traffic flow in time period t; S is the number of lanes.

[0043] Step 5-3: Calculate the starting position of congestion. The specific calculation expression is:

[0044]

[0045] Step 5-4, calculate the congestion end position, the specific calculation expression is:

[0046]

[0047] Where d is the distance between the accident site and the starting toll station.

[0048] The present invention adopts the above technical solution. Compared with the existing technology, the present invention has the following beneficial effects: 1. Strong real-time performance: Compared with the traditional accident detection method, the present invention uses ETC real-time data to quickly detect traffic anomalies after an accident occurs, shortening the response time. 2. Low cost: No new hardware equipment is required, and the existing ETC data resources are fully utilized, reducing the cost of system construction and maintenance. 3. High precision: By establishing a scientific mathematical model and comprehensively analyzing indicators such as traffic flow, speed and travel time, the location of the accident can be accurately located with a small error. 4. Wide applicability: This method can be applied to all types of highways, is not restricted by region, and has good promotion value.

[0049] This invention utilizes the unique advantages of ETC data in traffic monitoring, combined with the current urgent need for efficient and accurate accident congestion detection technology, to significantly improve the safety management and congestion relief capabilities of expressways, and provide strong support for building a safer, more efficient and intelligent modern transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments;

[0051] Figure 1 The figure is a flow chart of the method for detecting the location of highway accident congestion based on ETC big data according to the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0053] like Figure 1 As shown, the present invention discloses a method for detecting highway accident congestion locations based on ETC big data, which includes the following steps:

[0054] Step 1: Data acquisition and preprocessing: After collecting and cleaning ETC transaction data from each toll station on the highway, the ETC transaction data is matched with the vehicle's travel records at adjacent toll stations to calculate the travel time of each vehicle passing through the road section.

[0055] Furthermore, the ETC transaction data in step 1 includes the timestamp of the vehicle passing through the toll station, the geographical location information of the toll station, and vehicle identification information (such as license plate number, vehicle model, etc.).

[0056] Furthermore, in step 1, traffic accident reports corresponding to the study time period are obtained synchronously to verify the accuracy of the model.

[0057] Furthermore, data cleaning in step 1 includes removing incomplete or erroneous records, unifying the time format, and correcting timestamp errors.

[0058] Step 2: Segment the ETC transaction data at fixed time intervals and calculate the average vehicle travel time, average vehicle speed, and traffic flow in each time period.

[0059] Furthermore, step 2 specifically includes the following steps:

[0060] Step 2-1, segment the data into statistics at a fixed time interval Δt (e.g., every 5 minutes or 15 minutes);

[0061] Step 2-2, calculate the average travel time T of vehicles in time period t t :

[0062]

[0063] Where n is the number of vehicles passing through in time period t, T i is the travel time of the i-th vehicle.

[0064] Step 2-3, calculate the average speed V of the vehicle in time period t t :

[0065]

[0066] Wherein, L is the length of the road section between adjacent toll stations;

[0067] Step 2-4, calculate the traffic flow F in time period t t :

[0068]

[0069] Step 3, anomaly detection: calculate the flow rate change rate and speed change rate of the current time period t respectively, and compare them with the corresponding thresholds to determine whether there is a traffic accident;

[0070] Furthermore, step 3 specifically includes the following steps:

[0071] Step 3-1, calculate the rate of change of flow and speed:

[0072]

[0073] Step 3-2, setting the flow rate and speed drop thresholds α and β respectively based on historical data or industry standards;

[0074] Step 3-3, determining whether the decrease values of the flow rate and speed are both less than the set threshold value; if so, determining that an abnormal event occurred during the time period; otherwise, determining that no abnormal event occurred during the time period;

[0075] Specifically, if ΔF≤-α and ΔV≤-β are both satisfied, it is considered that an abnormal event has occurred during the time period, which may be caused by a traffic accident.

[0076] Step 4: Calculate the accident location using the accident location estimation model. The specific calculation expression is:

[0077]

[0078] Where L is the length of the road section between adjacent toll stations; V t is the average speed in time period t; V t-1 is the average speed in time period t-1; T t is the average travel time in time period t; d is the distance from the accident point to the starting toll station.

[0079] Specifically, the variable definition is: L: the length of the road section between adjacent toll booths. t : Average speed in time period t. V t-1 : Average speed in time period t-1. T t : Average travel time in time period t. d: Distance between the accident point and the starting toll station. Accident location calculation: Based on the travel time of vehicles before and after the accident, establish the equation: Then solve the accident location d:

[0080] In addition, as a feasible implementation method, the impact of weather, road conditions, special events, etc. on traffic flow is considered to improve the model.

[0081] Step 5: Calculate the congestion length based on the average travel time of normal traffic, and determine the starting and ending locations of the congestion in combination with the accident location;

[0082] Furthermore, step 5 specifically includes the following steps:

[0083] Step 5-1: Calculate the average travel time T under normal circumstances 正常 , the specific calculation expression is:

[0084]

[0085] Where L is the length of the road section between adjacent toll stations; V t-1 is the average speed in time period t-1

[0086] Step 5-2, calculate the congestion length L 拥堵 , the specific calculation expression is:

[0087]

[0088] Among them, F t is the traffic flow in time period t; S is the number of lanes.

[0089] Step 5-3: Calculate the starting position of congestion. The specific calculation expression is:

[0090]

[0091] Step 5-4, calculate the congestion end position, the specific calculation expression is:

[0092]

[0093] Where d is the distance between the accident site and the starting toll station.

[0094] Effect description: Actual data comparison: Considering that the calculation of the congestion location depends on the accuracy of the accident location, the accident location d calculated by the model is compared with the location in the actual traffic accident report to evaluate the accuracy of the model.

[0095] Error analysis and model adjustment: Calculate the accident location error, analyze the error source and adjust the model parameters, such as the flow and speed drop thresholds -α, -β and other coefficients, to improve model accuracy. The error calculation expression is:

[0096] Error = |d 模型 -d 实际 |.

[0097] The present invention adopts the above technical solution. Compared with the existing technology, the present invention has the following beneficial effects: 1. Strong real-time performance: Compared with the traditional accident detection method, the present invention uses ETC real-time data to quickly detect traffic anomalies after an accident occurs, shortening the response time. 2. Low cost: No new hardware equipment is required, and the existing ETC data resources are fully utilized, reducing the cost of system construction and maintenance. 3. High precision: By establishing a scientific mathematical model and comprehensively analyzing indicators such as traffic flow, speed and travel time, the location of the accident can be accurately located with a small error. 4. Wide applicability: This method can be applied to all types of highways, is not restricted by region, and has good promotion value.

[0098] This invention utilizes the unique advantages of ETC data in traffic monitoring, combined with the current urgent need for efficient and accurate accident congestion detection technology, to significantly improve the safety management and congestion relief capabilities of expressways, and provide strong support for building a safer, more efficient and intelligent modern transportation system.

[0099] Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

Claims

1. A highway accident congestion location detection method based on ETC big data, characterized by: It includes the following steps: Step 1: Collect ETC transaction data from each toll station on the highway and perform data cleaning. Then, match the vehicle's passage records at adjacent toll stations based on the ETC transaction data. Step 2: Segment the ETC transaction data at fixed time intervals and calculate the average vehicle travel time, average vehicle speed, and traffic flow in each time period; Step 3: Calculate the traffic flow rate change rate and speed change rate in the current time period t, and compare them with the corresponding thresholds to determine whether there is a traffic accident; Step 4: Calculate the accident location using the accident location estimation model. The specific calculation expression is: Where L is the length of the road section between adjacent toll stations; V t is the average speed in time period t; V t-1 is the average speed in time period t-1; T t is the average travel time in time period t; d is the distance between the accident site and the starting toll booth; Step 5: Calculate the congestion length based on the average travel time of normal traffic, and determine the starting and ending locations of the congestion in combination with the accident location.

2. The method for detecting highway accident congestion locations based on ETC big data according to claim 1, characterized in that: The ETC transaction data in step 1 includes the timestamp of the vehicle passing through the toll station, the geographic location information of the toll station, and vehicle identification information; the vehicle identification information includes the license plate number and vehicle model.

3. The method for detecting highway accident congestion locations based on ETC big data according to claim 1, characterized in that: In step 1, traffic accident reports corresponding to the study period are obtained synchronously to verify the accuracy of the model.

4. The method for detecting highway accident congestion locations based on ETC big data according to claim 1, characterized in that: Data cleaning in step 1 includes removing incomplete or erroneous records, unifying the time format, and correcting timestamp errors.

5. The method for detecting highway accident congestion locations based on ETC big data according to claim 1, characterized in that: Step 2 specifically includes the following steps: Step 2-1, segment the data into statistics at fixed time intervals Δt; Step 2-2, calculate the average travel time T of vehicles in time period t t : Where n is the number of vehicles passing through in time period t, T i is the travel time of the i-th vehicle; Step 2-3, calculate the average speed V of the vehicle in time period t t : Wherein, L is the length of the road section between adjacent toll stations; Step 2-4, calculate the traffic flow F in time period t t :

6. The method for detecting highway accident congestion locations based on ETC big data according to claim 1, characterized in that: Step 3 specifically includes the following steps: Step 3-1, calculate the rate of change of flow and speed: Step 3-2, setting the flow rate and speed drop thresholds α and β respectively based on historical data or industry standards; Step 3-3, determine whether the decrease values of flow rate and speed are both not greater than the set threshold value; if so, determine that an abnormal event occurred in the time period; otherwise, determine that no abnormal event occurred in the time period.

7. The method for detecting highway accident congestion locations based on ETC big data according to claim 1, characterized in that: Step 5 specifically includes the following steps: Step 5-1: Calculate the average travel time T under normal circumstances 正常 , the specific calculation expression is: Where L is the length of the road section between adjacent toll stations; V t-1 is the average speed in time period t-1 Step 5-2, calculate the congestion length L 拥堵 , the specific calculation expression is: Among them, F t is the traffic flow in time period t; S is the number of lanes; Step 5-3: Calculate the starting position of congestion. The specific calculation expression is: Step 5-4, calculate the congestion end position, the specific calculation expression is: Where d is the distance between the accident site and the starting toll station.

Images