Urban road traffic jam identification and detection method for jam prevention

Abstract

The invention relates to an urban road traffic jam identification and detection method for jam prevention. The method comprises the following steps: collecting road network section flow data; based on road network road section flow data, through traffic flow operation process modeling, an urban road network traffic flow state is predicted; on the basis of the urban road network traffic flow state prediction result, future congested road sections of the urban roads are judged; and based on the judgment result of the future congested road section of the urban road, the upstream source road section of the vehicle in the future congested road section is backtracked. Compared with the prior art, the method utilizes the real-time traffic flow data of the road section to predict the time-space traffic evolution process, identifies the road section which is in a congested state in the future in the road network, traces the upstream source road section of the vehicle on the congested road section in the future, and can realize accurate identification of the future congested source of the urban road. An upstream source is identified before congestion occurs, and the universality of congestion tracing is improved by using a traffic flow mechanism model.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic control, and in particular to an urban road traffic congestion identification and detection method for preventing congestion. Background Art

[0002] With the rapid development of social economy, traffic demand has increased dramatically, urban traffic has become increasingly tense, and the negative impact of traffic congestion has become more prominent. Urban roads are composed of complex sections and intersections. The composition and flow of traffic flows at different times and locations are different. The causes, locations, and scope of impact of traffic congestion are also different, showing obvious temporal and spatial distribution characteristics. Traffic congestion is a direct reflection of the imbalance between supply and demand of roads within a certain period of time. When traffic congestion occurs, the traffic flow upstream and downstream of the road changes significantly. In the upstream section, the traffic capacity decreases, the traffic density continues to rise, the road occupancy rate continues to increase, the driving speed decreases, the traffic flow through the road section continues to decrease, and the vehicle delay increases, thereby increasing the travel time to the destination.

[0003] Congestion in urban road networks usually shows spatiotemporal dependence on multiple upstream sections. Even upstream source sections far away from the congested area may contribute a large amount of traffic demand. The practical problem of "where does the traffic at the congested location come from" has been waiting for a clear answer. At the same time, preventing traffic congestion requires not only the implementation of traffic management strategies at congested nodes, but also active control measures for upstream source sections. For the source of congestion, measures such as boundary control, path induction and reversible lanes can be adopted to effectively restrict vehicles that are about to enter the congested area. Therefore, tracing the source of traffic congestion in urban roads is crucial for implementing active control strategies. Existing research on tracing the source of urban road traffic congestion faces two challenges: one is that it is difficult to finely trace the upstream source sections that cause future congestion based on historical aggregate data; the other challenge is that traditional congestion tracing methods often rely on data-driven models and lack consideration of the evolution mechanism of traffic flow, resulting in a lack of interpretability and universality of the methods. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide an urban road traffic congestion identification and detection method for blockage prevention. The method can use the real-time traffic flow data of the road section to predict the spatiotemporal traffic evolution process, identify the sections in the road network that will be congested in the future, and trace the upstream source sections of vehicles on the future congested sections.

[0005] The purpose of the present invention can be achieved by the following technical solution: A method for identifying and detecting urban road traffic congestion for preventing congestion, comprising the following steps:

[0006] S1, collecting road network section flow data;

[0007] S2. Based on the traffic flow data of road network sections, the traffic flow status of urban road network is predicted by modeling the traffic flow operation process;

[0008] S3, based on the prediction result of step S2, determining the future congested sections of urban roads;

[0009] S4. Based on the judgment result of step S3, trace back to the upstream source section of vehicles in the future congested section.

[0010] Furthermore, the step S1 specifically utilizes traffic detection technology to obtain the number of vehicles and queue length information of each road section in the road network.

[0011] Furthermore, the step S2 of predicting the traffic flow state of the urban road network includes predicting the traffic flow of the road section and predicting the queue length of the road section.

[0012] Furthermore, the specific process of predicting the road section traffic flow is as follows:

[0013] Vehicles passing through the import section L e Flows into the road network and flows through the connecting section L c , and finally pass through the exit section L x Outflow road network, these three types of road segments constitute the set of all road segments in the road network: L = L e ∪L c ∪L x , based on the conservation law, the traffic flow operation process of the road section is modeled:

[0014]

[0015] Among them, N i (t) represents the number of vehicles on road section i at time t; and They represent the number of vehicles flowing into and out of road section i at time t respectively;

[0016] The number of vehicles entering a connecting or exiting section is determined by the traffic capacity, signal timing scheme, queue length and downstream section space. For an entry section, the number of vehicles entering is is determined by the inflow rate α; for the exit section, the number of outflowing vehicles It is determined by the outflow rate β:

[0017]

[0018] Among them, δ ji (t) is a binary variable representing the state of the traffic light from section j to section i; S ji represents the saturated flow rate from section j to section i; nji represents the number of lanes on segment j that flow to segment i; q ji (t) represents the queue length on segment j heading to segment i at time t; C i represents the maximum capacity of road section i; P ji It represents the turning ratio from section j to section i;

[0019] δ ik (t) is a binary variable representing the traffic light status from section i to section k; S ik represents the saturated flow rate from section i to section k; n ik represents the number of lanes on section i flowing to section k; q ik (t) represents the queue length on segment i heading to segment k at time t; C k represents the maximum capacity of road section k; P ik It represents the steering ratio from section i to section k.

[0020] Furthermore, the specific process of predicting the queue length of the road section is as follows:

[0021] To determine the number of vehicles flowing into and out of a road segment, the queue length of the road segment is predicted according to the following formula:

[0022]

[0023] Considering that vehicles entering from the upstream will not reach the stop line of the downstream section immediately, the free flow travel time is used to calculate the number of vehicles joining the queue. At the same time, the concept of point queue is introduced, the physical length of the queue is ignored, and the capacity of the section is considered. For the import section, the number of vehicles joining the queue is determined by the inflow rate:

[0024]

[0025] For connecting segments and entrance segments, the number of vehicles leaving the queue is:

[0026]

[0027] in, and They represent the number of vehicles joining and leaving the queue on segment i at time t and heading to segment k respectively; l i represents the length of road segment i; represents the free flow speed of road segment i.

[0028] Furthermore, step S3 includes the following process:

[0029] S31, define the road section congestion index, design the traffic congestion index TCI, and use the road section speed to identify urban road congestion;

[0030] S32, constructing a speed-density relationship of a road section, and fitting a speed-density relationship model of the road section by using the road section basic graph theory and measured data;

[0031] S33, converting the future speed of the road section, converting the predicted road section density into the road section speed through the fitted road section speed-density relationship model, and then calculating the traffic congestion index of the road section. When the TCI of road section i i (t) is less than the threshold μ c , it indicates that road section i is in a congested state at time t.

[0032] Furthermore, the traffic congestion index TCI in step S31 is defined as the ratio of the average speed of the road section to the free flow speed:

[0033]

[0034] TCI i (t) represents the traffic congestion index of road section i at time t; v i (t) represents the average speed of section i at time t.

[0035] Furthermore, the speed-density relationship model of the road section in step S32 is specifically:

[0036]

[0037] Among them, k i is the density of road section i, k t , θ1, θ2 are fitting parameters.

[0038] Furthermore, step S4 includes the following process:

[0039] S41, calculate the time when the vehicle enters the future congested section, assuming that the section k is obtained at time It will become congested, and the vehicles on section k include running vehicles and queued vehicles use Indicates time The set of all vehicles on the congested road section k at time t, firstly, needs to confirm the time when the vehicle enters the congested road section k, and the running vehicle at time The vehicle has not joined the queue or reached the stop line, so the running time of the vehicle on section k is less than the free flow travel time. The travel time of queued vehicles on road segment k includes waiting time and free flow travel time:

[0040]

[0041] in, represents the moment when vehicle ω enters road section k; represents the free flow travel time of road segment i; represents the time when the first queued vehicle enters section k;

[0042] S42, calculate the travel time of the running vehicle and the queued vehicle, when The number of vehicles entering section k reaches the queue length Then get By comparison and N ik (t), the upstream section of the vehicle is predicted. When the vehicle ω is at time Entering section k, indicating Based on this, it is confirmed that the upstream source section of vehicle ω is i;

[0043] Continue tracing back and calculate the travel time of vehicle ω on road section i, then calculate the time when vehicle ω enters road section i, and identify the upstream source road section j;

[0044] S43, trace back to the upstream source section, when the backtracking time reaches the set threshold t tra When , the tracing ends and the last road section s of vehicle ω is output ω , trace all vehicles on the congested section k, and calculate the number of vehicles contributed by each upstream section to the congested section k

[0045] Furthermore, the calculation of the time when the vehicle ω enters the road section i in step S42 specifically includes:

[0046] The travel time t ω,i Free Flow Travel Time With waiting time d ω,i ,When calculating the waiting time, two cases are considered: 1) vehicle ω does not experience queuing and passes directly through the intersection; 2) vehicle ω experiences queuing due to a red light;

[0047] First, obtain the green light elapsed time when the vehicle ω leaves the road section i, that is, the time when the vehicle leaves the road section Green light start time The maximum number of vehicles passing through the green light is calculated as follows:

[0048]

[0049] By comparing the queue length q of the lane where vehicle ω is located ω,ik (t) and the maximum number of vehicles that can pass during the green light time C ω,ik , to determine whether the vehicle is experiencing a queue, if q ω,ik (t) Always smaller than C ω,ik , then there is no queueing and the waiting time is 0;

[0050] On the contrary, if there is a queue, there exists q ω,ik (t) greater than C ω,ik The waiting time is the time when the vehicle starts Join the queue time The difference:

[0051]

[0052] Vehicle start time It is calculated by the green light start time, the maximum capacity of the green light time and the reverse wave speed:

[0053]

[0054] Among them, w ik represents the reverse wave velocity from section i to section k;

[0055] When vehicle ω joins the queue, q ω,ik With C ω,ik Similarly, the time when the vehicle joins the queue is calculated based on this:

[0056]

[0057] Among them, t ′ ω,i represents the time when vehicle ω joins the queue.

[0058] Compared with the prior art, the present invention has the following advantages:

[0059] The present invention first collects the traffic flow data of the road network section, and predicts the traffic flow state of the urban road network, determines the future congested sections of the urban roads, and traces back the upstream source sections of the vehicles in the future congested sections. Thus, a "prediction-identification-backtracking" framework is constructed, using the real-time traffic flow data of the section, based on the traffic flow evolution mechanism to predict the spatiotemporal traffic evolution process, predict the traffic flow of the section, identify the sections in the road network that will be congested in the future through the basic graph relationship of the section, and trace back the vehicles on the future congested sections, tracing back the upstream source sections of the vehicles on the future congested sections, so as to identify the upstream source before the congestion occurs, realize the accurate identification of the future congestion source of the urban road, and improve the universality of congestion tracing through the traffic flow mechanism model. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0061] Figure 2 It is a schematic diagram of the application process of the embodiment;

[0062] Figure 3a and 3b It is a driving trajectory diagram of the vehicle on the road section in the embodiment. DETAILED DESCRIPTION

[0063] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0064] Example

[0065] like Figure 1 As shown, a method for identifying and detecting urban road traffic congestion for preventing congestion includes the following steps:

[0066] S1, collecting road network section flow data;

[0067] S2. Based on the traffic flow data of road network sections, the traffic flow status of urban road network is predicted by modeling the traffic flow operation process;

[0068] S3, based on the prediction result of step S2, determining the future congested sections of urban roads;

[0069] S4. Based on the judgment result of step S3, trace back to the upstream source section of vehicles in the future congested section.

[0070] This embodiment applies the above solution to construct Figure 2 The congestion tracing model framework shown in the figure mainly includes:

[0071] Step 1: Collect traffic data on road network sections

[0072] Traffic detection technology is used to obtain information such as the number of vehicles and queue length on each section of the road network through radar, coils and other equipment, where the detected area is all sections in the road network.

[0073] Step 2: Predict the traffic flow status of urban road network

[0074] (2.1) Predict the traffic flow on a road section.

[0075] Vehicles passing through the import section (L e ) flows into the road network and flows through the connecting section (L c ), and finally through the exit section (L x ) outflow road network. These three types of road segments constitute the set of all road segments in the road network (L = L e ∪L c ∪L x ), the traffic flow operation process of the road section is modeled based on the conservation law:

[0076]

[0077] Where N i(t) represents the number of vehicles on road section i at time t; and They represent the number of vehicles flowing into and out of road section i at time t respectively.

[0078] The number of vehicles entering a connecting or exiting section is determined by the capacity, signal timing scheme, queue length, and downstream section space. For an entry section, the number of vehicles entering is determined by the inflow rate α.

[0079]

[0080] where δ ji (t) is a binary variable representing the state of the traffic light; S ji represents the saturated flow rate from section j to section i; n ji represents the number of lanes on segment j that flow to segment i; q ji (t) represents the queue length on segment j heading to segment i at time t; C i represents the maximum capacity of road section i; P ji It represents the steering ratio of the flow from section j to section i.

[0081] Similarly, the number of vehicles outflowing from a connecting or importing segment can also be calculated. For an exporting segment, the number of outflowing vehicles is determined by the outflow rate β.

[0082]

[0083] (2.2) Predict the queue length of the road section.

[0084] In order to determine the number of vehicles flowing into and out of the road segment, the queue length of the road segment needs to be predicted and calculated:

[0085]

[0086] in and They respectively represent the number of vehicles joining and leaving the queue on section i heading to section k at time t.

[0087] Considering that vehicles entering from upstream will not reach the stop line of the downstream section immediately, the free flow travel time is used to calculate the number of vehicles joining the queue. At the same time, the concept of point queue is introduced, ignoring the physical length of the queue, but still considering the capacity of the section. For the import section, the number of vehicles joining the queue is determined by the inflow rate.

[0088]

[0089] Among them l i represents the length of road segment i; represents the free flow speed of road segment i.

[0090] For connecting segments and entrance segments, the number of vehicles leaving the queue is:

[0091]

[0092] In practical applications, δ ji (t) can be obtained through signal data; S ji 、n ji , C i , l i , Can be obtained through road network topology data; ji Available through historical traffic data.

[0093] Step 3: Determine the future congested sections of urban roads

[0094] (3.1) Define the indicators for determining road congestion.

[0095] The present invention proposes a traffic congestion index (TCI) to identify urban road congestion using the road segment speed, which is defined as the ratio of the average road segment speed to the free flow speed:

[0096]

[0097] TCI i (t) represents the traffic congestion index of road section i at time t; v i (t) represents the average speed of section i at time t.

[0098] (3.2) Construct the speed-density relationship of the road section.

[0099] The existence of the basic graph of the road section proves that there is a relationship between the average speed of the road section and the average density of the road section. Using the basic graph theory of the road section and the measured data, the speed-density relationship model of the road section is fitted:

[0100]

[0101] where k i is the density of road section i, k t , θ1, θ2 are fitting parameters.

[0102] (3.3) Convert the future speed of the road section.

[0103] The predicted road density is converted into road speed through the road speed-density relationship model fitted in step (3.2). Then the traffic congestion index of the road can be calculated. When the TCI of the road i (t) is less than the threshold μ c , then the road section i is considered to be in a congested state at time t.

[0104] In step 3, according to relevant standards, this embodiment sets the congestion index threshold μ c The specific value of is taken as (0,0.3]; In addition, the specific process of fitting the speed-density relationship of the road section is:

[0105] ① Collect measured data on average speed and average density of road sections;

[0106] ② According to relevant experimental results, the model parameters in the road section speed-density relationship have the following relationship:

[0107] θ1=0.1612k t +0.0337

[0108] θ2=0.0093k t -0.0507

[0109] ③ According to the measured data, the Levenberg–Marquardt algorithm can be used to fit k t .

[0110] Step 4: Trace back the upstream source sections of vehicles on future congested sections

[0111] (4.1) Calculate the time when the vehicle enters the future congested section.

[0112] Assume that the road segment k is obtained at time The vehicles on section k include running vehicles and queued vehicles use Indicates time The set of all vehicles on the road section k at time . First, we need to confirm the time when the vehicle enters the congested road section k. The running vehicle at time The vehicle has not joined the queue or reached the stop line, so the running time of the vehicle on section k is less than the free flow travel time. The travel time of queued vehicles on road segment k includes waiting time and free flow travel time.

[0113]

[0114] in represents the moment when vehicle ω enters road section k; represents the free flow travel time of road segment i; represents the time when the first queuing vehicle enters section k.

[0115] (4.2) Calculate the travel time of running vehicles and queued vehicles.

[0116] When The number of vehicles entering section k reaches the queue length You can get By comparison and N ik (t), the upstream segment of the vehicle can be predicted. For example, when vehicle ω is at time Entering section k means Based on this, it can be confirmed that the upstream source section of vehicle ω is i.

[0117] To continue tracing, it is necessary to calculate the travel time of vehicle ω on road section i. Travel time t ω,i Free Flow Travel Time With waiting time d ω,i When calculating the waiting time, there are two cases: 1) Vehicle ω does not experience queuing and passes through the intersection directly (such as Figure 3a 2) Vehicle ω experiences a queue due to a red light (such as Figure 3b The key is to determine whether the vehicle ω has experienced a queue. Get the time when the green light of the vehicle ω leaves the road section i, that is, the time when the vehicle leaves the road section Green light start time Then calculate the maximum number of vehicles that pass during the green light elapsed time:

[0118]

[0119] By comparing the queue length q of the lane where vehicle ω is located ω,ik (t) and the maximum number of vehicles that can pass during the green light time C ω,ik , it can be determined whether the vehicle has experienced queuing.

[0120] like Figure 3a and Figure 3b As shown, if q ω,ik (t) Always smaller than C ω,ik , then there is no queue and the waiting time is 0. On the contrary, if there is a queue, there is q ω,ik (t) greater than C ω,i The waiting time is the time when the vehicle starts and the time t of joining the queue ′ ω,i The difference:

[0121]

[0122] Vehicle start time It can be calculated by the green light start time, the maximum capacity during the green light time and the reverse wave speed:

[0123]

[0124] where w ikIt represents the reverse wave velocity from section i to section k.

[0125] When vehicle ω joins the queue, q ω,ik With C ω,ik The same. Based on this, the time when the vehicle joins the queue can be calculated:

[0126]

[0127] where t ′ ω,i represents the time when vehicle ω joins the queue.

[0128] Based on the above formula, the travel time of vehicle ω on section i can be calculated, and then the moment when vehicle ω enters section i can be calculated, so as to identify the upstream source section j.

[0129] (4.3) Trace back to the upstream source section.

[0130] When the backtracking time reaches the set threshold t tra When , the tracing ends and the last road section s of vehicle ω is output ω By tracing back all vehicles on the congested road section k, the number of vehicles contributing to the congested road section k from each upstream road section can be calculated.

[0131] In step 4, the reverse wave speed w of the road segment ik Can be obtained through historical vehicle trajectory data.

[0132] In summary, this solution uses real-time traffic flow data of road sections to predict the spatiotemporal traffic evolution process, identify sections in the road network that will be congested in the future, and trace the upstream source sections of vehicles on future congested sections. By constructing a "prediction-identification-backtracking" framework and using traffic flow mechanism models to improve the universality of congestion tracing, this solution can achieve accurate identification of future sources of congestion on urban roads, identify upstream sources before congestion occurs, and provide a decision-making basis for subsequent active control strategies.

Claims

1. A method for identifying and detecting urban road traffic congestion for preventing congestion, characterized in that: The following steps are involved: S1, collecting road network section flow data; S2. Based on the traffic flow data of road network sections, the traffic flow status of urban road network is predicted by modeling the traffic flow operation process; S3, based on the prediction result of step S2, determining the future congested sections of urban roads; S4. Based on the judgment result of step S3, trace back to the upstream source section of vehicles in the future congested section.

2. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 1, characterized in that: The step S1 specifically uses traffic detection technology to obtain the number of vehicles and queue length information of each road section in the road network.

3. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 1, characterized in that: The step S2 predicts the traffic flow state of the urban road network, including predicting the traffic flow of the road section and predicting the queue length of the road section.

4. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 3, characterized in that: The specific process of predicting the traffic flow of a road section is as follows: Vehicles passing through the import section L e Flows into the road network and flows through the connecting section L c , and finally pass through the exit section L x Outflow road network, these three types of road segments constitute the set of all road segments in the road network: L = L e ∪L c ∪L x , based on the conservation law, the traffic flow operation process of the road section is modeled: Among them, N i (t) represents the number of vehicles on road section i at time t; and They represent the number of vehicles flowing into and out of road section i at time t respectively; The number of vehicles entering a connecting or exiting section is determined by the traffic capacity, signal timing scheme, queue length and downstream section space. For an entry section, the number of vehicles entering is is determined by the inflow rate α; for the exit section, the number of outflowing vehicles It is determined by the outflow rate β: Among them, δ ji (t) is a binary variable representing the state of the traffic light from section j to section i; S ji represents the saturated flow rate from section j to section i; n ji represents the number of lanes on segment j that flow to segment i; q ji (t) represents the queue length on segment j heading to segment i at time t; C i represents the maximum capacity of road section i; P ji It represents the turning ratio from section j to section i; δ ik (t) is a binary variable representing the traffic light status from section i to section k; S ik represents the saturated flow rate from section i to section k; n ik represents the number of lanes on section i flowing to section k; q ik (t) represents the queue length on segment i heading to segment k at time t; C k represents the maximum capacity of road section k; P ik It represents the steering ratio from section i to section k.

5. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 4, characterized in that: The specific process of predicting the queue length of the road section is as follows: To determine the number of vehicles flowing into and out of a road segment, the queue length of the road segment is predicted according to the following formula: Considering that vehicles entering from the upstream will not reach the stop line of the downstream section immediately, the free flow travel time is used to calculate the number of vehicles joining the queue. At the same time, the concept of point queue is introduced, the physical length of the queue is ignored, and the capacity of the section is considered. For the import section, the number of vehicles joining the queue is determined by the inflow rate: For connecting segments and entrance segments, the number of vehicles leaving the queue is: in, and They represent the number of vehicles joining and leaving the queue on segment i at time t and heading to segment k; l i represents the length of road segment i; represents the free flow speed of road segment i.

6. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 1, characterized in that: Step S3 The process includes: S31, define the road section congestion index, design the traffic congestion index TCI, and use the road section speed to identify urban road congestion; S32, constructing a speed-density relationship of a road section, and fitting a speed-density relationship model of the road section by using the road section basic graph theory and measured data; S33, converting the future speed of the road section, converting the predicted road section density into the road section speed through the fitted road section speed-density relationship model, and then calculating the traffic congestion index of the road section. When the TCI of road section i i (t) is less than the threshold μ c , it indicates that road section i is in a congested state at time t.

7. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 6, characterized in that: The traffic congestion index TCI in step S31 is defined as the ratio of the average speed of the road section to the free flow speed: TCI i (t) represents the traffic congestion index of road section i at time t; v i (t) represents the average speed of section i at time t.

8. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 7, characterized in that: The speed-density relationship model of the road section in step S32 is specifically: Among them, k i is the density of road section i, k t , θ1, θ2 are fitting parameters.

9. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 1, characterized in that: The step S4 includes the following process: S41, calculate the time when the vehicle enters the future congested section, assuming that the section k is obtained at time It will become congested, and the vehicles on section k include running vehicles and queued vehicles use Indicates time The set of all vehicles on the congested road section k at time t, firstly, needs to confirm the time when the vehicle enters the congested road section k, and the running vehicle at time The vehicle has not joined the queue or reached the stop line, so the running time of the vehicle on section k is less than the free flow travel time. The travel time of queued vehicles on road segment k includes waiting time and free flow travel time: in, represents the moment when vehicle ω enters road section k; represents the free flow travel time of road segment i; represents the time when the first queued vehicle enters section k; S42, calculate the travel time of the running vehicle and the queued vehicle, when The number of vehicles entering section k reaches the queue length Then get By comparison and N ik (t), the upstream section of the vehicle is predicted. When the vehicle ω is at time Entering section k, indicating Based on this, it is confirmed that the upstream source section of vehicle ω is i; Continue tracing back and calculate the travel time of vehicle ω on road section i, then calculate the time when vehicle ω enters road section i, and identify the upstream source road section j; S43, trace back to the upstream source section, when the backtracking time reaches the set threshold t tra When , the tracing ends and the last road section s of vehicle ω is output ω , trace all vehicles on the congested section k, and calculate the number of vehicles contributed by each upstream section to the congested section k 10. The method for identifying and detecting urban road traffic congestion for preventing congestion according to claim 9, characterized in that: The calculation of the time when the vehicle ω enters the road section i in step S42 specifically includes: The travel time t ω，i Free Flow Travel Time With waiting time d ω，i ,When calculating the waiting time, two cases are considered: 1) vehicle ω does not experience queuing and passes directly through the intersection; 2) vehicle ω experiences queuing due to a red light; First, obtain the green light elapsed time when the vehicle ω leaves the road section i, that is, the time when the vehicle leaves the road section Green light start time The maximum number of vehicles passing through the green light is calculated as follows: By comparing the queue length q of the lane where vehicle ω is located ω,ik (t) and the maximum number of vehicles that can pass during the green light time C ω，ik , to determine whether the vehicle is experiencing a queue, if q ω,ik (t) Always smaller than C ω，ik , then there is no queueing and the waiting time is 0; On the contrary, if there is a queue, there exists q ω,ik (t) greater than C ω，ik The waiting time is the time when the vehicle starts and the time t′ when joining the queue ω，i The difference: Vehicle start time It is calculated by the green light start time, the maximum capacity of the green light time and the reverse wave speed: Among them, w ik represents the reverse wave velocity from section i to section k; When vehicle ω joins the queue, q ω,ik With C ω，ik Similarly, the time when the vehicle joins the queue is calculated based on this: Among them, t′ ω，i represents the time when vehicle ω joins the queue.

Images