A road congestion degree identification method and early warning system under discrete time system

Abstract

The present invention discloses a road congestion degree identification method and early warning system under a discrete time system, the method comprising dividing any time period into two-dimensional time vectors according to the traffic flow characteristics of different time periods; calculating the predicted vehicle arrival rate and predicted vehicle detention probability distribution law under different two-dimensional time vectors; designing a quantitative standard to predict the road congestion degree index under different time vectors; and finally setting the road congestion degree early warning threshold and the road congestion degree level according to the road congestion degree index. The system comprises a traffic condition early warning data receiving module, a signal processing module, a signal light indication module, a buzzer early warning module, a computer storage module and a transmitting antenna module. The present invention predicts the road congestion condition according to the real-time traffic data of different road sections, improves the traffic congestion condition to a great extent, improves the judgment and attendance efficiency of traffic control personnel, reduces the prediction energy efficiency of traffic sections, and alleviates traffic pressure.

Description

Technical Field

[0001] The invention relates to information technology and operations research, and in particular to a road congestion degree identification method and an early warning system under a discrete time system. Background Art

[0002] Traffic conditions change rapidly. Due to its complexity and uncontrollability, it has always troubled traffic control personnel, and it is easy to cause untimely and lagging management. For example, when a road section is congested, the traffic management personnel go to check the congestion situation. It is likely that it is already serious and difficult to handle, or when the traffic management personnel arrive, the congestion situation has been resolved; and the road section with good traffic conditions may also hide hidden traffic problems. Although there are intelligent early warning systems now, due to the high cost of equipment and the short timeliness of the warning, it must be turned on for a long time, so the penetration rate is low, the cost and energy consumption are huge, and the practicality is poor. How to obtain quantitative indicators of traffic congestion and design a low-cost road congestion warning system without changing the existing traffic system has become an important problem that needs to be solved urgently. Summary of the invention

[0003] Purpose of the invention: The purpose of the present invention is to provide a road congestion degree identification method and early warning system under a discrete time system, so as to improve the traffic congestion situation to a great extent, improve the judgment and attendance efficiency of traffic control personnel, reduce the large amount of energy consumption of predicted traffic sections, increase the smoothness of traffic, and reduce traffic pressure.

[0004] Technical solution: The method for identifying road congestion level in a discrete time system according to the present invention comprises the following steps:

[0005] S1. Divide any time period into two-dimensional time vectors according to the traffic flow characteristics of different time periods;

[0006] The step S1 specifically includes: using statistical methods to analyze and obtain two time dimensions that have a greater impact on vehicle flow, and establishing a two-dimensional time vector, denoted as (t a ,t b ), where t a Indicates the time interval division in a day; t b Indicates the time interval division of different dates.

[0007] S2, extracting vehicle traffic information based on road vehicle identification data monitored by the road;

[0008] The step S2 specifically includes: using the road monitoring, selecting the camera facing the road section, extracting the vehicle traffic information under different two-dimensional time vectors described in step S1, and recording different two-dimensional time vectors (t a ,t b) corresponds to the statistical information of the number of vehicles arriving, that is, the vehicle traffic information.

[0009] S3. Based on the road supervision system, record the signal light indication cycle and traffic duration;

[0010] The step S3 is specifically as follows: the signal light indication cycle is recorded as T(s); the passage time, that is, the green light time in one cycle, is recorded as a(s).

[0011] S4, calculating the predicted vehicle arrival rate under different two-dimensional time vectors based on a large amount of passing vehicle traffic information under different two-dimensional time vectors;

[0012] The step S4 is specifically as follows: based on the vehicle traffic information under different time vectors described in step S2, a system prediction algorithm is designed through a large amount of past vehicle traffic information recorded by the road congestion level recognition and warning system to obtain the predicted vehicle arrival rate under different two-dimensional time vectors, which is recorded as

[0013] S5, calculating the predicted vehicle detention probability distribution law after passing the signal light indication period under different two-dimensional time vectors based on the predicted vehicle arrival rate and vehicle traffic information under different two-dimensional time vectors;

[0014] The step S5 is specifically as follows: according to the predicted average vehicle arrival rate under different two-dimensional time vectors described in step S4 and the road travel time of the observed traffic section described in step S3, the predicted vehicle detention probability distribution law after passing the traffic light indication cycle under different time vectors is calculated.

[0015] S6. Designing a quantitative standard based on the predicted vehicle detention probability distribution law to predict the road congestion index under different time vectors; the step S6 specifically comprises: obtaining the number of vehicles detained after any indication period by calculation according to the predicted vehicle detention probability distribution law under different two-dimensional time vectors described in step S5; considering the wide applicability of the road congestion index in a large number of application scenarios of different road sections, the present invention provides a road congestion index after any indication period, denoted as Cr, which is based on the limit vehicle passing number of the traffic section, i.e., A c The predicted vehicle detention probability distribution law is used to calculate a universal road congestion degree measurement standard.

[0016] S7. Setting a road congestion warning threshold and a road congestion level according to road congestion indicators under different two-dimensional time vectors.

[0017] The step S7 specifically includes: setting a road congestion warning threshold, denoted as Cr hBy comparing the road congestion level and the road congestion warning threshold, the different working states of the indicator lights and buzzers can be identified, so as to predict and identify the vehicle congestion level on the observed road section, which plays an important role in assisting and guiding the work of traffic management personnel.

[0018] A road congestion level early warning system in a discrete time system, the system adopts the above-mentioned road congestion level identification method in a discrete time system, and includes the following modules:

[0019] Traffic condition warning data receiving module: used to receive real-time vehicle traffic information and indicator light information under different two-dimensional time vectors of the road supervision system and road monitoring, and store it in the computer storage module;

[0020] Signal processing module: used to process the data of the data receiving module, and obtain the control signal of the signal light and buzzer and the propagation signal of the transmitting antenna through the processing of the signal processing module;

[0021] Traffic light indication module: used to receive control signals from the signal processing module, and to remind traffic control personnel of their judgment and attendance through the changes of red, yellow and green lights;

[0022] Buzzer warning module: used to receive the control signal of the signal processing module, and remind the traffic control personnel whether they should go out to control the traffic through the status change of the buzzer;

[0023] Transmitting antenna module: used to receive control signals from the signal processing module, and broadcast wireless signals to inform passers-by in advance through the antenna, so that they can prepare for congestion and change to alternative sections in time;

[0024] Computer storage module: used to store real-time traffic data and road congestion identification and warning system programs, ensure data persistence, and provide fast access to support the normal operation of the computer. In addition, it also includes data management, transmission and security functions to effectively organize, protect and share stored information, so that the computer can efficiently complete various computing tasks and operations.

[0025] A computer storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned method for identifying the degree of road congestion in a discrete time system.

[0026] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for identifying the degree of road congestion in a discrete time system is implemented.

[0027] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0028] 1. The present invention establishes a discrete time system, creatively establishes a two-dimensional time vector according to the correlation between vehicle arrival rate and time, sets a prediction algorithm, predicts the vehicle detention situation under different two-dimensional time vectors, quantifies the average road congestion index under different time vectors, sets the traffic condition warning threshold, and transmits warning information through buzzer sounding and wireless signal transmitter, which can simply and efficiently warn the congested road condition information of the monitored road section, thereby improving the attendance effectiveness of traffic control personnel;

[0029] 2. The present invention can predict the congestion level of the monitored road section in advance, so that the passing vehicles can select the alternative road section in advance. This two-pronged approach has a good early warning function for the vehicle congestion situation on the monitored road section, which can greatly improve the smoothness of traffic and control efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of a method for identifying road congestion level in a discrete time system;

[0031] Figure 2 It is a schematic diagram of the vehicle detention probability distribution law under different two-dimensional time vectors;

[0032] Figure 3 It is a schematic diagram of the predicted road congestion index under different two-dimensional time vectors;

[0033] Figure 4 The invention discloses a road congestion degree identification method under a discrete time system and a schematic diagram of a terminal structure of an early warning system. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0035] like Figure 1 As shown, a method for identifying road congestion level in a discrete time system includes the following steps:

[0036] S1. Divide any time period into two-dimensional time vectors according to the traffic flow characteristics of different time periods. According to the statistical method, the two time dimensions that have a greater impact on traffic flow are analyzed and a two-dimensional time vector is established, which is recorded as (t a ,t b ), where t a Indicates the time interval division in a day; t b Indicates the time interval division of different dates.

[0037] The specific implementation method of the two-dimensional time vector division includes: according to the strong correlation between the arrival rate of vehicles and time, according to the statistical method analysis, the priority ranking of the influence of time on the arrival rate of vehicles is obtained, the two time dimensions with the highest priority are taken, and the two-dimensional time vector (t a ,t b ). Wherein, t in the two-dimensional time vector coordinates a Specifically expressed as: a ∈{t a1 ,t a2 ,t a3 ,t a4 ,t a5 ,t a6 ,t a7}, the t a Represents the time interval division of a day, which is divided into: t a1 , morning (2-6 am); t a2 , morning peak (6:00-9:00); t a3 , morning (9:00-12:00); t a4 , noon (12:00-14:00); t a5 , afternoon (14:00-17:00); t a6 , evening peak (17:00-20:00); t a7 , night (20:00-2:00 the next day). Among them, the two-dimensional time vector coordinates t b Specifically expressed as: b ∈{t b1 ,t b2 ,t b3}, the t b Indicates the time interval division of different dates, which is divided into: t b1 , ordinary working day; t b2 , connecting day (the last working day and the last holiday day); t b3 , day off.

[0038] S2. Extracting vehicle traffic information based on road vehicle identification data monitored by the road. Using road monitoring, select a camera facing the road section to extract vehicle traffic information under different two-dimensional time vectors described in step S1, and record different two-dimensional time vectors (t a ,t b ) corresponds to the statistical information of the number of vehicles arriving, that is, the vehicle traffic information.

[0039] The specific process of extracting vehicle traffic information includes: using road monitoring, selecting a camera facing the observed road section, and extracting vehicle traffic information based on road vehicle identification data monitored by the road, including: the time slot number of the vehicle arriving at the traffic section and the time slot number of the vehicle leaving the observed section; and according to the vehicle traffic information under different two-dimensional time vectors, the number of vehicle arrivals under different two-dimensional time vectors can be easily obtained, for example: the two-dimensional time vector coordinate (t ak ,t bo ) under the i-th group of observation statistics, the number of vehicle arrival statistics is recorded as:

[0040] S3. Based on the road supervision system, record the signal light indication cycle and the duration of the traffic flow. The signal light indication cycle is recorded as T(s); the duration of the traffic flow is the duration of the green light in one cycle, recorded as a(s).

[0041] Based on the road supervision system, the traffic light information in the direction opposite to the observed section is selected: the green light duration data, i.e. the passing time, is extracted, denoted as a(s); the total duration data of the red light and the yellow light is extracted, denoted as b(s); then one indication cycle is denoted as T=(a+b)(s).

[0042] S4, based on a large amount of passing vehicle traffic information under different two-dimensional time vectors, calculate the predicted vehicle arrival rate under different two-dimensional time vectors. Based on the vehicle traffic information under different time vectors described in step S2, a system prediction algorithm is designed through a large amount of passing vehicle traffic information recorded by the road congestion level recognition and early warning system to obtain the predicted vehicle arrival rate under different two-dimensional time vectors, denoted as

[0043] Based on a large amount of past vehicle traffic information at different time vectors, a weighted algorithm is used to obtain the predicted average vehicle arrival rate at different time vectors, which is used as the predicted vehicle arrival rate at the next same two-dimensional time vector, denoted as: but:

[0044]

[0045] Where n is the latest n-time two-dimensional time vector (t ak ,t bo ) under the vehicle traffic data; n1 is the latest n1-time two-dimensional time vector (t ak ,t bo ) under the vehicle traffic data; θ1 is the weight value of the latest n1 vehicle arrivals recorded by the road congestion recognition and early warning system, where θ1∈(0.5,1); The latest i-th two-dimensional time vector (t ak ,t bo ) total number of vehicle arrivals under; is a two-dimensional time vector (t ak ,t bo ) represents the length of the time period, the specific expression is: in, Represents a two-dimensional time vector (t ak ,t bo )'s end slot number; Represents a two-dimensional time vector (t ak ,t bo ) starting time slot number. Additional explanation: a k Indicates the first time parameter in the two-dimensional time vector, which has 7 different states; b o It represents the second time parameter in the two-dimensional time vector, and has three different states.

[0046] S5. Calculate the predicted vehicle retention probability distribution law after passing the signal light indication period under different two-dimensional time vectors based on the predicted vehicle arrival rate and vehicle traffic information under different two-dimensional time vectors. The predicted vehicle retention probability distribution law after passing the signal light indication period under different time vectors is calculated based on the predicted vehicle average arrival rate under different two-dimensional time vectors described in step S4 and the road travel time of the observed traffic section described in step S3.

[0047] Since the signal light indication cycle T, the passage time a in a single cycle, and the prohibited passage time b in a single cycle in the road supervision system are all positive integer time lengths, establishing a discrete time system can very accurately describe the vehicle traffic conditions of the traffic system under different time vectors, and can better quantify and predict the degree of traffic congestion.

[0048] The discrete time system specifically includes: the entry protocol adopted is an early arrival system, that is, the vehicle arrival can only occur at the beginning of the time slot (k, k + ), k represents the number of time slots; when a vehicle passes through the observation section, the service completion and departure can only occur at the end of the time slot (k -,k). Considering the actual situation of vehicle arrival on the observation section, vehicles can arrive at any time slot, regardless of the direction of travel indicated by the indicator light. Therefore, the process of vehicle arrival at the system conforms to the geometric arrival process with an arrival rate of α, that is, each time slot is equally likely to receive an arriving vehicle, and the vehicle arrives with an arrival probability α. There is a probability (1-α) that no vehicle arrives at the observation section, and the arrival process of each vehicle is independent of each other and does not affect each other. Considering the actual situation of vehicle traffic on the observation section, vehicles can only pass through the observation section during the green light time slot, which is strongly related to the direction of travel indicated by the indicator light. According to the number of vehicles passing through the observation section recorded in a single indication cycle, the maximum value is taken as the limit number of vehicles passing through the observation section, that is:

[0049]

[0050] in, represents the number of vehicles passing in the i-th indication period, A c Indicates the maximum number of vehicles passing through the observation section.

[0051] The vehicle arrival rate under different two-dimensional time vectors and the road travel time of the system observation section predict the vehicle detention probability distribution law after the indication period under different time vectors. The initial system vehicle detention number is preset to be n0=0, and the system detention number after the (k-1)th indication period is n k-1 , after the next indication cycle ends, the number of vehicles stranded in the system is n k , the number of vehicles arriving in the kth indication period is x k , the number of vehicles passing through the kth indication period is After the kth indication cycle, the number of stranded vehicles in the system is n k In the known (n k-1 ,x k ,a ck ) is:

[0052] n k =n k-1 +x k -a ck ,k≥1

[0053] Due to a ck The randomness of the trial system determines the value A c Replace the random variable a ck , and obtain the number of vehicles stranded in the system after the kth indication cycle n k In the known (n k-1 ,x k ,A c ) is:

[0054]

[0055] Due to the Markov property of the road congestion identification and warning system, the number of vehicles stranded in the system after the kth indication cycle is related to the number of vehicles arriving in the kth indication cycle, the number of vehicles leaving in the green light time slot, and the number of vehicles stranded in the system after the (k-1)th indication cycle, and has nothing to do with the number of vehicles stranded in the system after the previous (k-2) indication cycle and the process of vehicles arriving and leaving the system in the previous (k-2) indication cycle. Therefore, the transfer probability formula of stranded vehicles in the system can be obtained as follows:

[0056] P{n k =l|n0=l0,n1=l1,…,n k-1 = l k-1}=P{n k =l|n k-1 = l k-1}

[0057] According to the transfer probability of stranded vehicles in the above system, the probability formula of the number of vehicles after any indication period can be obtained after the number of vehicles after the previous indication period is known. According to the above preset initial system, the number of stranded vehicles is n0=0, that is:

[0058] P{n0=0}=1

[0059] The probability formula for the number of vehicles stranded under n1=l can be obtained:

[0060]

[0061] In the known n k-1 In the case of n, the state and transition probability after the previous indication cycle can be used to obtain n k =l, the probability formula for the number of vehicles stranded and the steady-state probability distribution law:

[0062]

[0063] According to the above preset initial system stranded vehicle number and the obtained n k =l vehicles stranded number probability formula can be obtained P{n k = l}, that is, the predicted vehicle detention probability distribution law, is recorded as: P{n = l}. According to the different two-dimensional time vectors (t ak ,t bo ) By bringing the predicted vehicle arrival rates under different two-dimensional time vectors into the calculation method of different vehicle detention probability distribution laws of the system, the predicted vehicle detention probability distribution laws under different two-dimensional time vectors can be easily and accurately predicted.

[0064] S6. Design a quantitative standard to predict the road congestion index under different time vectors based on the predicted vehicle detention probability distribution law. The number of vehicles detained after any indication period can be obtained according to the predicted vehicle detention probability distribution law under different two-dimensional time vectors described in step S5. Considering the wide applicability of setting the road congestion index, according to the limit vehicle passing number of the traffic section, it is recorded as A c , considering the predicted vehicle detention probability distribution law, an algorithm is designed to quantify the road congestion index, and the road congestion index after any indication period can be obtained, which is denoted as Cr.

[0065] According to the predicted vehicle detention probability distribution law under the different two-dimensional time vectors, the predicted vehicle detention probability distribution law after any indication period can be obtained. Due to the average vehicle detention time, the number of stranded vehicles in the traffic section can effectively reflect the road congestion level of the observed section. When there is no vehicle detention in the kth period, that is, the system stranded vehicles and the periodic arriving vehicles are cleared at the time slot of the green light in the kth period, the weighting of the system congestion level is 0, which shows that the traffic system has a strong processing capacity for arriving vehicles and stranded vehicles, the traffic capacity is sufficient, and the system is not congested; when there is a vehicle detention in the kth period, that is, the system stranded vehicles and the periodic arriving vehicles are not cleared at the time slot of the green light in the kth period, and there are l vehicles stranded, the weighting is:

[0066]

[0067] Considering the wide applicability of setting the average road congestion index, the maximum number of vehicles passing through the road section A is considered. c , then the road congestion index after any indication period can be expressed as:

[0068]

[0069] The size of the positive indicator Cr is used to represent the real-time predicted road congestion level of the observed section. The larger the Cr value is, the higher the predicted road congestion level of the observed section is. The closer the Cr value is to zero, the better the traffic conditions are. The value of Cr is always a non-negative number.

[0070] S7. Set the road congestion warning threshold and road congestion level according to the road congestion level indicators under different two-dimensional time vectors. Set the road congestion warning threshold, denoted as Cr h By comparing the road congestion level and the road congestion warning threshold, the different working states of the indicator lights and buzzers can be identified, which plays an important role in predicting and identifying the vehicle congestion level on the observed road section and assisting in guiding the work of traffic management personnel.

[0071] According to the road congestion degree index, three different road congestion degree levels and system feedback are set, specifically including: when Cr ≤ 0.1, it indicates that the traffic prediction status of the observed section is good and there is almost no congestion, and the system indicator light shows green; when 0.1 < Cr ≤ 0.3, it indicates that the traffic prediction status of the observed section is average, there is congestion but the system can still handle it, and the system indicator light shows yellow; the road congestion degree warning threshold of the system is expressed as: Cr h , the system predicts Cr h = 0.3. When 0.3 < Cr, it indicates that the traffic prediction status of the observed section is poor, there is congestion and the vehicle congestion waiting time is long, and the system indicator light shows red. The warning system buzzer alarms intermittently for many times, and the traffic police should immediately manage the traffic of vehicles on the section. The warning system broadcasts warning information, warning the vehicles passing through the observed section that they can choose to change to an alternative route, so as to effectively warn of road congestion conditions, dispatch police in a timely and targeted manner, greatly improve the efficiency of police dispatch, and reduce the congestion probability that may occur on the section.

[0072] Taking a certain section in Jiangsu as an example, predict the predicted vehicle arrival rate of the section under different two-dimensional time vectors Through the predicted vehicle arrival rate formula, the predicted data is calculated as shown in Table 1:

[0073] Table 1 Predicted vehicle arrival rate under different two-dimensional time vectors

[0074]

[0075] The limit vehicle passing number A of the section c = 50; the green light duration is 30S, and the red light and yellow light durations are 60S, that is, the indication cycle is 90S. Table 1 shows the predicted vehicle arrival rates under 21 different two-dimensional time vectors Now, a two-dimensional time vector is taken to describe the prediction method, and the prediction methods under the other two-dimensional time vectors are the same and will not be elaborated. Considering predicting the congestion status during the morning rush hour of the next connecting day, extract the predicted vehicle arrival rate in Table 1 under this two-dimensional time vector From the above vehicle retention quantity distribution law solving method, the vehicle retention distribution law after the passing cycle during the morning rush hour of the next connecting day can be obtained. According to the above preset initial system retention vehicle number n0 = 0, predicted vehicle arrival rate A c = 50, and the predicted vehicle retention distribution law data during the prediction period is calculated as shown in Table 2:

[0076] Table 2 Predicted vehicle retention distribution law during the prediction period

[0077]

[0078] From Table 2, we can see that the total probability of the number of stranded vehicles being greater than or equal to 10 is less than or equal to 1 / 1000. The system considers the situation with a probability less than 1 / 1000 as an extremely small probability event, so it is omitted. Considering the probability distribution law of the number of stranded vehicles from 0 to 9, the road congestion index can be expressed as:

[0079]

[0080] According to the road congestion index Cr and the road congestion warning threshold Cr h By comparing the sizes of , we can obtain the predicted result of the congestion condition during the morning rush hour of the next connecting day: because Cr≤0.1, the traffic forecast condition of the road section during the morning rush hour of the next connecting day is good, with almost no congestion, and the system indicator light is green.

[0081] like Figure 2 As shown in Figure 2, the vehicle detention probability distribution law under 21 different two-dimensional time vectors is shown. Figure 2 It can be seen that when the predicted arrival rate When it is higher, the vehicle detention probability distribution curve will have a second probability area maximum as the number of vehicle detention increases. Figure 2 As shown, in the two-dimensional time vector (t a2 ,t b1 ) and the two-dimensional time vector (t a6 ,t b1 ), the second probability maximum occurs when the number of stranded vehicles is large. Since the probability of vehicle stranding does not decrease monotonically with the number of stranded vehicles, the fluctuation of the probability of vehicle stranding often leads to vehicle congestion on the road.

[0082] like Figure 3 As shown in Figure 2, the values ​​and trends of the predicted road congestion indicators under different two-dimensional time vectors are shown. The three line graphs represent t b1 , ordinary working day; t b2 , connecting day (the last working day and the last holiday day); t b3 , the predicted road congestion index on holidays. It is not difficult to see from the figure that the congestion level fluctuates most significantly on the next ordinary working day, and the road congestion on the next holiday is the lightest; from the position of the warning threshold line, it can be seen that the predicted traffic conditions on this observed section are generally good, but during the morning and evening peaks on the next ordinary working day, traffic management personnel need to dispatch police in time to ease road traffic pressure and inform passers-by in advance to change to alternative sections.

[0083] like Figure 4 As shown, a road congestion warning system in a discrete time system includes the following modules:

[0084] Traffic condition warning data receiving module: used to receive real-time vehicle traffic information and indicator light information under different two-dimensional time vectors of the road supervision system and road monitoring, and store it in the computer storage module.

[0085] Signal processing module: used to process the data from the data receiving module. The processing algorithm is the above-mentioned road congestion degree identification method under a discrete time system. The signal processing module processes the control signals of the traffic light and buzzer and the propagation signal of the transmitting antenna.

[0086] Traffic light indication module: used to receive control signals from the signal processing module, and to clearly prompt traffic control personnel to make judgments and attendance through the changes of red, yellow and green lights.

[0087] Buzzer warning module: used to receive control signals from the signal processing module, and prompt traffic control personnel whether they should go out to control traffic through changes in the status of the buzzer.

[0088] Transmitting antenna module: used to receive control signals from the signal processing module, and broadcast wireless signals to passers-by in advance through the antenna, so that they can be prepared for congestion and change to alternative sections in time.

[0089] Computer storage module: used to store real-time traffic data and road congestion identification and warning system programs, ensure data persistence, and provide fast access to support the normal operation of the computer. In addition, it also includes data management, transmission and security functions to effectively organize, protect and share stored information, so that the computer can efficiently complete various computing tasks and operations.

Claims

1. A method for identifying road congestion level in a discrete time system, characterized in that: The following steps are involved: S1. Divide any time period into two-dimensional time vectors according to the traffic flow characteristics of different time periods; S2, extracting passing vehicle traffic information based on road vehicle identification data monitored by the road; S3. Based on the road supervision system, record the signal light indication cycle and traffic duration; S4, calculating the predicted vehicle arrival rate under different two-dimensional time vectors based on the passing vehicle traffic information under different two-dimensional time vectors; S5, calculating the predicted vehicle detention probability distribution law after passing the signal light indication cycle under different two-dimensional time vectors based on the predicted vehicle arrival rate under different two-dimensional time vectors and the passing information of past vehicles; S6. Based on the predicted vehicle detention probability distribution law, a quantitative standard is designed to predict the road congestion index under different two-dimensional time vectors; S7, setting a road congestion warning threshold and a road congestion level according to the road congestion level indicators under different two-dimensional time vectors; The step S3 is specifically as follows: the signal light indication cycle is recorded as T; the passage time, i.e. the green light duration in one cycle, is recorded as a, and the total duration data of the red light and the yellow light is extracted, which is recorded as b; then one indication cycle is recorded as T=a+b; The predicted vehicle detention probability distribution law is specifically as follows: the number of vehicles detained in the initial system is preset to be b0=0, and the number of vehicles detained in the system after the k-1th indication cycle is n k-1 , after the next indication cycle ends, the number of vehicles stranded in the system is n k , the number of vehicles arriving in the kth indication period is x k , the number of vehicles passing through the kth indication period is After the kth indication cycle, the number of stranded vehicles in the system is n k In the known (n k-1 ,x k ,a ck ) is: n k =n k-1 +x k -a ck ,k≥1 Due to a ck The randomness of the observed road section is used to calculate the maximum number of vehicles passing through A. c Replace the random variable a ck , and obtain the number of vehicles stranded in the system after the kth indication cycle n k In the known (n k-1 ,x k ,A c ) is: The transfer probability formula of stranded vehicles in the system is: P{n k =l|n0=l0,n1=l1,…,n k-1 =l k-1 }=P{n k =l|n k-1 =l k-1 } According to the transfer probability of stranded vehicles in the above system, the probability formula of the number of vehicles after any indication period is obtained after the number of vehicles after the previous indication period is known. According to the above preset initial system, the number of stranded vehicles is n0=0, that is: The probability formula for the number of vehicles stranded under n1=l is obtained: In the known n k-1 In the case of n, according to the state and transition probability after the previous indication cycle, we get n k =l, the probability formula for the number of vehicles stranded and the steady-state probability distribution law: According to the above preset initial system stranded vehicle number and the obtained n k =l vehicles stranded number probability formula to obtain P{n k =l}, that is, the predicted vehicle detention probability distribution law, denoted as: P{n=l}.

2. The method for identifying road congestion level in a discrete time system according to claim 1, characterized in that: The step S1 specifically includes: using statistical methods to analyze and obtain two time dimensions that have a greater impact on vehicle flow, and establishing a two-dimensional time vector, denoted as (t a ,t b ), where t a Indicates the time interval division in a day; t b Indicates the time interval division of different dates.

3. The method for identifying road congestion level in a discrete time system according to claim 1, characterized in that: The step S2 specifically includes: using road monitoring, selecting a camera facing the road section, extracting vehicle traffic information under different two-dimensional time vectors, and recording different two-dimensional time vectors (t a ,t b ) corresponds to the arrival quantity statistics of the vehicles under.

4. The method for identifying road congestion level in a discrete time system according to claim 1, characterized in that: The step S4 is specifically as follows: based on the passing vehicle information in step S2, a system prediction algorithm is designed through the passing vehicle information recorded by the road congestion level recognition and warning system to obtain the predicted vehicle arrival rate under different two-dimensional time vectors, which is recorded as 5. The method for identifying road congestion level in a discrete time system according to claim 1, characterized in that: The step S5 is specifically as follows: according to the predicted average vehicle arrival rate under different two-dimensional time vectors and the road travel time of the observed traffic section, the predicted vehicle detention probability distribution law after passing the signal light indication cycle under different time vectors is calculated.

6. The method for identifying road congestion level in a discrete time system according to claim 1, characterized in that: The step S6 specifically comprises: obtaining the number of vehicles stranded after any indication period by calculating the predicted vehicle stranding probability distribution law under different two-dimensional time vectors; considering that the road congestion degree index of the road congestion degree identification and early warning system should have wide applicability in a large number of application scenarios of different road sections, a road congestion degree index after any indication period is proposed, denoted as Cr, which is based on the limit vehicle passing number A of the traffic section. c and a universal road congestion degree measurement standard calculated based on the predicted vehicle detention probability distribution law.

7. The method for identifying road congestion level in a discrete time system according to claim 1, characterized in that: The step S7 specifically includes: setting a road congestion warning threshold, denoted as Cr h By comparing the road congestion level and the road congestion warning threshold, the different working states of the indicator lights and buzzers can be identified, thereby predicting and identifying the vehicle congestion level on the observed road section.

8. A road congestion warning system in a discrete time system using a road congestion identification method in a discrete time system as claimed in any one of claims 1 to 7, characterized in that: The system includes the following modules: Traffic condition warning data receiving module: used to receive real-time vehicle traffic information and indicator light information under different two-dimensional time vectors of the road supervision system and road monitoring, and store it in the computer storage module; Signal processing module: used to process the data of the data receiving module, and obtain the control signal of the signal light and buzzer and the propagation signal of the transmitting antenna through the processing of the signal processing module; Traffic light indication module: used to receive control signals from the signal processing module, and to remind traffic control personnel of their judgment and attendance through the changes of red, yellow and green lights; Buzzer warning module: used to receive the control signal of the signal processing module, and remind the traffic control personnel whether they should go out to control the traffic through the status change of the buzzer; Transmitting antenna module: used to receive control signals from the signal processing module, and broadcast wireless signals to inform passers-by in advance through the antenna, so that they can prepare for congestion and change to alternative sections in time; Computer storage module: used to store real-time traffic data and road congestion identification and warning system programs, providing fast access to support the normal operation of the computer; also includes data management, transmission and security functions to effectively organize, protect and share stored information.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, a method for identifying the degree of road congestion in a discrete-time system according to any one of claims 1 to 7 is implemented.

Images