Urban center road bearing capacity early warning method and device

Abstract

The invention discloses a city center road bearing capacity early warning method and device, and the method comprises the steps: obtaining the traffic data of a city center road, determining an equivalent lane, calculating the traffic bearing capacity of a basic unit based on a converted equivalent traffic flow, and carrying out the early warning of the traffic bearing capacity of the basic unit through the mapping relation between the equivalent traffic flow and a non-equivalent traffic flow. Calculating the traffic bearing capacity of the non-equivalent lane, and combining the relevance between the equivalent lane and the non-equivalent lane in time and space; a road network global traffic bearing capacity calculation model is established among the determined maximum number of running vehicles of the basic units, the road network structure parameters, the signal timing parameters and the average travel speed, to-be-detected city center road data is input into the road network global traffic bearing capacity calculation model to predict a result, and risk early warning is carried out. The urban center road quality risk early warning method improves the accuracy of urban center road quality risk early warning, has good interpretability, and is suitable for an urban center road bearing capacity early warning system.

Description

Technical Field

[0001] The present invention relates to the field of road bearing capacity early warning, and in particular to a method and device for early warning of road bearing capacity in a city center. Background Art

[0002] As the artery that restricts urban development, road traffic is a very complex integrated system. Its formation and development are restricted by many factors such as urban structure, scale, nature, economy, etc. The main cause of urban traffic problems is the imbalance between traffic supply and traffic demand. When the existing traffic supply resources cannot effectively meet the traffic needs of travelers, various traffic problems will arise.

[0003] In today's urban traffic management field, with the continuous increase in traffic flow and the intensification of congestion in the central business district (CBD) of the city, how to effectively evaluate and manage the traffic carrying capacity of the central business district has become an urgent problem to be solved. Traffic flow on urban roads will show periodic stop-and-go behavior due to the influence of traffic signals. In addition, due to the fluctuation of traffic flow, the number of vehicles in the road network will change dynamically, making it challenging to directly calculate the basic unit traffic carrying capacity, and it is even more difficult to establish the relationship between the average travel speed and the number of basic unit vehicles.

[0004] Traffic flow in urban roads will show periodic stop-and-go behavior due to the influence of traffic signals. In addition, due to the fluctuation of traffic flow, the number of vehicles in the road network will change dynamically, making it challenging to directly calculate the basic unit traffic carrying capacity, and it is even more difficult to establish the relationship between the average travel speed and the number of basic unit vehicles. To this end, it is necessary to first analyze the operation characteristics of traffic flow in the road network, explore the common change laws of vehicles in the process of travel, and then lay the foundation for the subsequent traffic flow modeling. In the road section driving area, due to the influence of surrounding traffic flow, the running speeds of different vehicles may be different; in the intersection driving area, due to the constraints of traffic lights, vehicles are forced to queue up and change the running mode, and pass through the intersection in the form of intermittent flow. Therefore, in this traffic environment, it is difficult to directly calculate the number of vehicles running in the road network by mathematical methods. Combined with the analysis results of traffic flow operation characteristics, a city center road carrying capacity early warning method and device are needed. Summary of the invention

[0005] The purpose of the present invention is to provide a method for early warning of the carrying capacity of urban central roads.

[0006] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0007] The present invention comprises the following steps:

[0008] A Obtain traffic data of the central roads in the city, determine the equivalent lanes, and convert the intermittent traffic flow controlled by signals on the roads into continuous release traffic flow while keeping the number of vehicles unchanged. The lanes that can continuously release traffic flow after equivalent are the equivalent lanes;

[0009] B. Calculating the traffic carrying capacity of the basic unit based on the converted equivalent traffic flow. The equivalent lane area in the basic unit can be divided into a stacked release area and a free driving area. By using the mapping relationship between the equivalent traffic flow and the non-equivalent traffic flow, the traffic carrying capacity of the non-equivalent lane is calculated and the temporal and spatial correlation between the equivalent lane and the non-equivalent lane is combined.

[0010] C established the relationship between the number of lanes in the road network structure parameters and the green-to-signal ratio in the signal timing parameters, determined the relationship between the maximum number of operating vehicles in the basic unit and the road network structure parameters, signal timing parameters, and average travel speed, established a road network global traffic carrying capacity calculation model, input the data of the city center roads to be tested into the prediction results of the road network global traffic carrying capacity calculation model, and carried out risk warning.

[0011] Furthermore, in step C, a basic unit travel time model is constructed, and the relationship between the section length and the average travel speed in the road network structure parameters is established. The travel time of the vehicle flow can be divided into free travel time and stack release time, and a travel time calculation model for each section is constructed. The formula is as follows:

[0012]

[0013]

[0014] Where: L u Indicates the length of the basic unit u; represents the distance traveled at free flow speed in the basic unit u; and They represent the length of the road section and the length of the intersection in the basic unit u respectively; n represents the number of vehicles; represents the expected average travel speed of the basic unit u;

[0015] Among them, the average travel time of the traffic flow in the basic unit u is T u Can be calculated based on the length of the basic unit and the average travel speed; average free travel time It can be determined based on the length of the free flow section and the free flow speed; average cascade release time The calculation is made using the time relationship between different sections of the basic unit.

[0016] Furthermore, in step C, the calculation of the road network global traffic carrying capacity calculation model includes the following steps:

[0017] According to the operation rules of traffic flow in different areas, the traffic carrying capacity of the stacking release area and the free driving area are calculated respectively;

[0018] Based on the interaction between the stacked release area and the free travel area, the traffic carrying capacity of the equivalent lane is determined;

[0019] By analyzing the mapping relationship between equivalent traffic flow and non-equivalent traffic flow, the traffic carrying capacity of non-equivalent lanes is calculated;

[0020] Combining the temporal and spatial correlation between equivalent lanes and non-equivalent lanes, the basic unit traffic carrying capacity is derived. The calculation formula of the basic unit traffic carrying capacity can be expressed as:

[0021]

[0022] Where: B represents the set of basic traffic units in the road network, B = {1, 2, 3, ..., u, ...}, which includes the upstream section driving area R ij and downstream intersection form area R j , A u represents the traffic carrying capacity of the u-th basic unit in the road network, u∈B, and They represent the traffic carrying capacity of the equivalent lane and the traffic carrying capacity of the non-equivalent lane in the u-th basic unit respectively;

[0023] and They respectively represent the traffic carrying capacity of the stacked release area and the traffic carrying capacity of the free driving area in the u-th basic unit.

[0024] Furthermore, the calculation formula of the road network traffic carrying capacity can be expressed as:

[0025]

[0026] The constraints are:

[0027]

[0028] Where: A represents the traffic carrying capacity of the road network; represents the average travel speed of the road network; Represents the traffic proportion in the tth direction.

[0029] Furthermore, the road network global traffic carrying capacity calculation model has a built-in traffic carrying capacity time warning model, which uses the actual utilization area S of the CBD area roads, the actual operation time T of the roads and the time and space consumption C of traffic during peak hours. i , and the road network carrying capacity model is obtained:

[0030]

[0031] Substituting CBD compactness, actual road operation efficiency, parking system correction coefficient, public transportation correction coefficient and policy correction value into the corrected road network carrying capacity, i.e., the traffic carrying capacity model, the formula is as follows:

[0032]

[0033] Among them, A is the traffic carrying capacity; A T is the traffic carrying capacity threshold of the road section in the CBD area. When the traffic volume on the road section satisfies the following formula, the road section i in the CBD will not be in a congested state;

[0034]

[0035] in, A is the traffic volume forecast value of road section i at (t+Δt); T Indicates traffic carrying capacity;

[0036] Let Ni represent the critical capacity of traffic flow on road section i, and (t+Δt)Ni represent the number of vehicles on road section i at (t+Δt), then:

[0037]

[0038] Where: Ni is the critical capacity of traffic flow on section i; (t+Δt)Ni is the number of vehicles on section i at (t+Δt); Li is the length of section i; vi is the critical speed of congested traffic flow on section i; (t+Δt)vi is the average speed of vehicles on section i at (t+Δt);

[0039] The speed prediction based on Seq2Seq-GCGRU algorithm and correction coefficient is used to predict the traffic data of a certain road section in the next stage. The formula is as follows;

[0040]

[0041] Among them, Φ represents the probability distribution of the prediction results, is the predictor variable, are input features, represents the training set, is the set of model parameters of the regression model, Λ is the variational distribution of the model parameters of the regression model,

[0042] Goodness of fit:

[0043]

[0044] Where Ξ represents the goodness of fit, ω is the parameter that adjusts the weight between the variational distribution and the likelihood function, ν is the model parameter, i is the index of ν, ρ is the variational distribution, Υ is the uniform prior distribution, Π is the likelihood function, ∈ is the uncertainty parameter,

[0045] Probability density function:

[0046]

[0047] Among them, τ n is the noise term, is the predicted value of the regression model, α is the adjustment function,

[0048] Goodness of fit for prediction uncertainty:

[0049]

[0050] Use the test set to test the accuracy of the regression model.

[0051] Furthermore, during risk warning, the wireless communication module of the warning device control unit receives information including the number and direction of moving targets on the road from the road monitoring unit, and after analysis and processing, controls the warning screen and audio playback device to prompt and warn drivers entering the road section;

[0052] The warning control unit uses the STM32F103ZET6 chip as the control core, which controls the wireless communication module to receive information transmitted by the road monitoring unit and control the roadside warning equipment.

[0053] In a second aspect, an embodiment of the present application further provides an electronic device, including:

[0054] A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.

[0055] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0056] The beneficial effects of the present invention are:

[0057] The present invention is a method and device for early warning of the carrying capacity of urban central roads. Compared with the prior art, the present invention has the following technical effects:

[0058] The present invention simplifies the complex and changeable traffic environment in the actual road network, ensures that the number of vehicles in the road network remains unchanged before and after the conversion, and significantly reduces the complexity of the modeling process.

[0059] Thus, a possible quality crisis can be quickly responded to and warned. In addition, the present invention also has good interpretability, so that enterprises can clearly understand the basis of risk warning, and enhance the transparency and credibility of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of the steps of a method for early warning of the carrying capacity of urban central roads according to the present invention;

[0061] Figure 2 A system block diagram of a method for early warning of the carrying capacity of urban central roads according to the present invention;

[0062] Figure 3 A schematic diagram of lane equivalent changes in a method for early warning of the carrying capacity of a city center road according to the present invention;

[0063] Figure 4 A schematic diagram of an equivalent lane inspection at a two-way ten-lane cross signal intersection of a city center road carrying capacity early warning method according to the present invention;

[0064] Figure 5 A schematic diagram for calculating the time relationship between normal traffic flow and equivalently abstracted traffic flow in a method for early warning of the carrying capacity of a city center road according to the present invention;

[0065] Figure 6 It is a node speed diagram of a homogeneous traffic state and a non-homogeneous traffic state of a method for early warning of the carrying capacity of a city center road of the present invention;

[0066] Figure 7 This is a flow chart of the CBD traffic carrying capacity time warning in the urban center road carrying capacity warning method in the embodiment of this specification. DETAILED DESCRIPTION

[0067] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0068] like Figure 1 and 2 As shown, a method for early warning of the carrying capacity of a city center road of the present invention comprises the following steps:

[0069] A Obtain traffic data of the central roads in the city, determine the equivalent lanes, and convert the intermittent traffic flow controlled by signals on the roads into continuous release traffic flow while keeping the number of vehicles unchanged. The lanes that can continuously release traffic flow after equivalent are the equivalent lanes;

[0070] B. Calculating the traffic carrying capacity of the basic unit based on the converted equivalent traffic flow. The equivalent lane area in the basic unit can be divided into a stacked release area and a free driving area. By using the mapping relationship between the equivalent traffic flow and the non-equivalent traffic flow, the traffic carrying capacity of the non-equivalent lane is calculated and the temporal and spatial correlation between the equivalent lane and the non-equivalent lane is combined.

[0071] C established the relationship between the number of lanes in the road network structure parameters and the green-to-signal ratio in the signal timing parameters, determined the relationship between the maximum number of operating vehicles in the basic unit and the road network structure parameters, signal timing parameters, and average travel speed, established a road network global traffic carrying capacity calculation model, input the data of the city center roads to be tested into the prediction results of the road network global traffic carrying capacity calculation model, and carried out risk warning.

[0072] Step 1: Equivalent lane division

[0073] The present invention proposes a method that can not only simplify the complex and changeable traffic environment in the actual road network, but also ensure that the number of vehicles in the road network remains unchanged before and after the conversion, which will significantly reduce the complexity of the modeling process.

[0074] like Figure 3 As shown in the figure, the equivalent lane in the homogeneous traffic state converts the intermittent traffic flow controlled by the signal on it into a continuous release traffic flow while keeping the number of vehicles unchanged. The lane that can release the traffic flow continuously after the equivalent is the equivalent lane. The conversion process is determined by the number of entrance lanes and the green-signal ratio of the intersection. The core idea of ​​the equivalent lane is to realize the conversion of discontinuous traffic flow to continuous traffic flow based on the principle of constant average travel speed under the premise of the same number of vehicles. This transformation greatly simplifies the complexity of the traffic carrying capacity calculation process on the basis of ensuring the theoretical correctness of the equivalent concept. The specific idea of ​​the equivalent process is: when the traffic flow in the intersection is in a non-saturated release, while keeping the original average travel speed of the traffic flow unchanged, there is still remaining space on the road for more vehicles to use, that is, the road network has not yet reached the state of the maximum number of running vehicles. In other words, the traffic carrying capacity of the road network can only be obtained when the traffic flow is in a saturated release state. Therefore, based on the principle of constant number of vehicles, the number of vehicles released saturatedly during the green light period is equal to the number of vehicles released continuously on the equivalent lane within the period.

[0075] After lane equivalence, the number of entrance lanes at the intersection is reduced, the continuity of traffic flow is enhanced, and the average travel speed of vehicles and the number of vehicles in the road network remain unchanged. In other words, after equivalence, it is equivalent to converting the flat intersection into an overpass, allowing vehicles to pass through the intersection in the form of an overpass. Correspondingly, lanes that are not controlled by signals and continuously release traffic, such as right-turn lanes at signalized intersections and lanes at non-signalized intersections, are called non-equivalent lanes. Through lane equivalence, the relationship between the number of lanes in the road network structure parameters and the green-to-signal ratio in the signal timing parameters is established. It should be supplemented that the equivalent conversion only reduces the number of entrance lanes that remain open in the intersection, and the number of lanes in the road section remains unchanged. Therefore, the original storage capacity of the road will not change, that is, the equivalent process will not affect the maximum number of queued vehicles that the road can accommodate.

[0076] 1.2 Inspection of Equivalent Lanes

[0077] (1) Cross intersection

[0078] In terms of cross intersections, the present invention constructs a two-way ten-lane cross signal intersection. The intersection is a traditional four-phase intersection with a cycle time of 200s and a free flow speed of 65km / h. The schematic diagram before and after the equivalent is as follows Figure 4 shown.

[0079] The relationship between the number of vehicles on the road network and the average travel speed is the same as the result of the single-direction verification. The average relative error of the number of vehicles on the road network before and after the equivalence is 1.61%, and the maximum error does not exceed 6.50%. The standard deviation of the error is 2.84%.

[0080] (2) Roundabout

[0081] In terms of roundabouts, the present invention constructs a four-branch roundabout signalized intersection, all of which are bidirectional ten-lane roads. The intersection adopts a single entrance signal control mode, and each entrance is set as a separate phase. The left-turning traffic and straight traffic in the entrance can be released at the same time. The intersection is set with four phases, and the cycle length is 200s. The free flow speed is 65km / h.

[0082] The relationship between the number of vehicles on the road network and the average travel speed is the same as the verification results of the cross intersection. The average relative error of the number of vehicles on the road network before and after the equivalence is 3.94%, and the maximum error does not exceed 9.36%. The standard deviation of the error is 3.56%. Based on the above series of verifications, the number of vehicles on the road network before and after the equivalence remains basically unchanged, and thus the maximum number of vehicles running in the road network will also be the same. This further demonstrates that the lane equivalence idea is reliable and effective in calculating the traffic carrying capacity of the road network.

[0083] Step 2: Calculation of basic unit traffic carrying capacity

[0084] 2.1 Construction of basic unit travel time model

[0085] After the lanes in the road network are equivalently converted, the number of open import lanes will be less than the number of lanes in the road section. Therefore, when there are many vehicles, some lanes in the intersection will be blocked. In order to calculate the basic unit traffic carrying capacity containing blocked traffic flow, the present invention proposes the concept of traffic flow abstraction. Traffic flow abstraction refers to extracting the key features of traffic flow in the basic unit operation process and summarizing the operation rules that can reflect its essential attributes. According to this idea, the traffic flow operation process can be abstracted into two processes: free travel and cascading release. Specifically, free travel refers to the traffic flow traveling at a free flow speed on the road section and the internal road of the intersection. The corresponding area is called the free travel area, referred to as the free area; Cascading release refers to the saturation release of traffic flow in a cascading form before the stop line. The corresponding area is called the cascading release driving area, referred to as the cascading area. This type of abstraction is achieved because no matter how fast the vehicle moves on the road, or in what form it queues to pass through the intersection, as long as the total travel time of the vehicle remains unchanged, the traffic flow can always be equivalent to the "free travel-cascading release" operation process, which ensures the reliability and accuracy of the abstract process.

[0086] 2.2 Calculation model of travel time between stacking area and free area

[0087] like Figure 5 As shown in the figure, corresponding to the abstract process of the stacking area and the free area in the basic unit, the travel time of the vehicle flow can be divided into free travel time and stacking release time. Among them, the average travel time T of the vehicle flow in the basic unit u u Can be calculated based on the length of the basic unit and the average travel speed; average free travel time It can be determined based on the length of the free flow section and the free flow speed; average cascade release time The time relationship between different sections of the basic unit is used for calculation, and then the travel time calculation model for each section is constructed. See the formula for details.

[0088]

[0089] Where: L u Indicates the length of the basic unit u; represents the distance traveled at free flow speed in the basic unit u; and They represent the length of the road section and the length of the intersection in the basic unit u respectively; n represents the number of vehicles; represents the expected average travel speed of the basic unit u.

[0090] The division of basic units simplifies the calculation object of the road network traffic carrying capacity, and the construction of travel time calculation models for different sections lays the foundation for determining the traffic carrying capacity. At the same time, by constructing the basic unit travel time model, the relationship between the section length and the average travel speed in the road network structure parameters is established.

[0091] Calculation of basic unit traffic carrying capacity based on equivalent vehicle flow

[0092] After a series of conversions of equivalent lanes and traffic flow abstractions, the traffic that can run continuously in the road network is defined as equivalent traffic. Based on the converted equivalent traffic, the traffic carrying capacity of the basic unit is calculated. According to the above analysis, the equivalent lane area in the basic unit can be divided into a stacked release area and a free driving area. Therefore, the calculation of the traffic carrying capacity of the basic unit can be decomposed into the following steps: First, according to the operation rules of traffic in different areas, the traffic carrying capacity of the stacked release area and the free driving area are calculated respectively; then, the interaction between the traffic between the stacked release area and the free driving area is discussed to determine the traffic carrying capacity of the equivalent lane; then, by analyzing the mapping relationship between equivalent traffic and non-equivalent traffic, the traffic carrying capacity of the non-equivalent lane is calculated; finally, the traffic carrying capacity of the basic unit is derived by combining the spatiotemporal correlation between equivalent lanes and non-equivalent lanes. Based on the above analysis, the calculation formula of the traffic carrying capacity of the basic unit can be expressed as:

[0093]

[0094] Where: B represents the set of basic traffic units in the road network, B = {1, 2, 3, ..., u, ...}, which includes the upstream section driving area R ij and downstream intersection form area R j , A u represents the traffic carrying capacity of the u-th basic unit in the road network, u∈B, and They respectively represent the equivalent lane traffic carrying capacity and non-equivalent lane traffic carrying capacity in the u-th basic unit. and They respectively represent the traffic carrying capacity of the stacked release area and the traffic carrying capacity of the free driving area in the u-th basic unit.

[0095] From the summary of the traffic carrying capacity calculation method above, it can be seen that there are currently two main problems with the definition and calculation of traffic carrying capacity: on the one hand, the traffic carrying capacity is the maximum value when it is static; on the other hand, the traffic carrying capacity obtained by calculation has little practical significance and cannot truly reflect the current traffic form of the CBD. However, the traffic carrying capacity of the road network has different carrying capacities under different circumstances, and it is dynamic and variable. Therefore, in response to the above problems, the present invention considers more influencing factors to make the results more accurate. Based on the traffic characteristics of the central business district of a large city and relying on the time-space consumption method, the present invention specifically quantifies these factors that have an impact on the road network, and at the same time introduces them as correction coefficients into the time-space consumption method model to obtain the model formula of traffic carrying capacity.

[0096] Before modeling, we basically assume that: within the urban road network system, all levels of roads (expressways, trunk roads, secondary trunk roads, and branch roads) are networked, and the urban road network infrastructure is relatively complete and stable. There is a set of traffic control and management measures that are compatible with the road network system to ensure the travel of motor vehicles, and the connection between motor vehicle travel and other modes of transportation is effective, and the proportion of travel modes is relatively stable. The calculation formula for the basic unit traffic carrying capacity is:

[0097]

[0098] Further in-depth analysis of the operation rules and interactive correlations of traffic in different sections shows that although the driving characteristics of vehicles in equivalent lanes and non-equivalent lanes are significantly different, the two are driving on their own roads and have little interference with each other; in equivalent lanes, there is an interactive influence between vehicles in the free driving area and vehicles in the stacked release area. Specifically, in order to maintain the equilibrium steady state of the basic unit and make the number of vehicles in the road network always at the maximum value, the vehicles connecting the free driving area and the stacked release area need to maintain corresponding replenishment and stability, and the stacked release area must not be unbalanced, resulting in a decrease or increase in the number of vehicles. That is, in order to ensure that the average travel speed of the basic unit of the road network remains unchanged, in the interval of unit headway, every time a vehicle leaves the stacked release area, the vehicle in the free driving area needs to be replenished in time, otherwise the basic unit will be unbalanced and unable to reach the original average travel speed. In other words, the average headway in the two states should be equal, that is Combined with the relationship between headway and headway, The calculation formula of the basic unit traffic carrying capacity can be further derived as follows:

[0099]

[0100] So far, the present invention has determined the interactive relationship between the maximum number of operating vehicles in the basic unit and the road network structure parameters, signal timing parameters, other related parameters, and average travel speed, and has preliminarily achieved the modeling goal of the road network traffic carrying capacity.

[0101] Step 3: Establishment and verification of the calculation model of the road network's overall traffic carrying capacity

[0102] Model variable selection

[0103] The three important parameters that affect traffic carrying capacity are the actual road area, the actual road operation time and the time and space consumption of individual traffic peak hours. In addition, the six main factors affecting traffic carrying capacity need to be considered, namely, land spatial structure, urban road network scale, road network hierarchy structure, traffic mode structure, traffic facilities and traffic management policies. Among them, the factors affecting urban road network scale and road network hierarchy structure have been reflected in the parameters of urban road effective area, so they are no longer considered separately. At the same time, since traffic carrying capacity is a capacity after all vehicles on the road are uniformly converted into standard cars, the factors affecting traffic mode structure are not considered for the time being.

[0104] ① Effective road area S in CBD area 车

[0105] The road area that can ensure normal use by individuals is also called the effective use area of ​​urban road facilities. Usually the road area occupied by motor vehicles is used as a reference to minimize the impact on traffic individuals, and the remaining area is the effective use area. The specific factors are:

[0106] 1) Grade correction factor R1

[0107] Motor vehicle lanes are usually based on the area of ​​main roads. The remaining roads need to be converted into the area of ​​main roads. The relative proportions of roads of different levels are also required, which is also called the comprehensive reduction coefficient R1, and R1=0.75~0.90.

[0108] 2) Lane correction factor R2

[0109] The width of urban roads in my country is generally 3.75m-4.0m. In order to meet the requirements of mixed vehicle operation, the correction factor is 1.0, but when the road width is 3.5m, it is 0.85, and the bicycle lane is 1.0.

[0110] 3) Average route usage frequency coefficient R3

[0111] Different road grades and functions take different values. It is also necessary to understand the driver's familiarity with the route, the lane selection preference when driving on the road, and the location of the route. High-grade roads have large traffic because of their advantages, and low-grade roads have small traffic, so high and low values ​​are taken respectively. The four-level road network takes 0.6 to 0.9 respectively.

[0112] 4) Route interference correction factor R4

[0113] Bus and taxi stops will affect the operation of vehicles, so it is recommended to take 0.7 to 0.9. Multi-lane roads need to take a higher value because usually only the outer lanes are affected by lateral interference. Therefore, the effective area of ​​the CBD internal road is:

[0114] S 车 =S×R1×R2×R3×R4

[0115] ②Effective road operation time T

[0116] T is the actual use time of the road. Generally speaking, whether the road capacity meets the requirements is determined by whether it meets the requirements during peak hours. The traffic organization of different intersections is different. The comprehensive reduction coefficient R5 is determined based on the survey results. The comprehensive reduction coefficient is between 0.6 and 0.9, and generally R5 = 0.6 ~ 0.8. CBD internal road capacity: T = R5

[0117] ③ Time and space consumption of traffic monomer during peak hours C i

[0118] C i It is the space occupied by individual traffic units during travel during peak hours. Based on the lane width, some parameters are transformed to obtain the following formula:

[0119]

[0120] C i is the time and space consumption of the traffic individual, hour·m 2 / times; h is the headway, m; t is the average travel time of a traffic unit during peak hours, h; d is the lateral width required for a traffic unit to move at a speed that reaches capacity, m; L is the average travel distance of a traffic unit during peak hours, m; v is the average speed of a traffic unit during peak hours, km / h; k is the traffic density, vehicles / km; q is the road network capacity, vehicles / h.

[0121] ④Compactness b

[0122] The compactness of CBD can be used to determine whether the internal spatial structure of the CBD area meets the requirements. The closer the b value is to 1, the denser it is, and it can also be considered that its internal spatial structure meets the requirements.

[0123] B is the compactness of CBD; Q is the area of ​​the region; l is the outer perimeter of the regional spatial shape.

[0124] ⑤ Actual road operation efficiency E

[0125] E is the actual operating efficiency of the road. It mainly describes the operating status of the road network at all levels within the CBD. It is derived from the following formula:

[0126] M is the total number of road sections; q i is the traffic volume of road section i; L i is the length of road section i; V i is the average travel speed on road section i.

[0127] ⑥Parking system correction factor

[0128] P—parking system capacity; P0—parking demand.

[0129] Among them, the parking system capacity P:

[0130] P=C0×ω

[0131] C0 is the number of parking spaces in the parking lot; ω is the parking turnover rate.

[0132] ⑦ Public transport correction factor θ

[0133] On the road, public transportation will affect the operation of motor vehicles, especially the travel of buses. However, the passenger capacity of buses is higher than that of private cars, so the use of buses can increase the service level of roads and improve the carrying capacity of the road network.

[0134]

[0135] Where: ε a is the proportion of cars used in transportation; b is the share of private cars in traffic; x is the bus occupancy rate; y is the private car occupancy rate; λ is the ratio of the road space occupied by a bus to the space occupied by a private car under the same service level.

[0136] ⑧ Traffic management policy correction value p

[0137] This factor is concretized using the classification of the intensity of public transportation development priority, as shown in the following table.

[0138] Prioritize the development of public transportation policies Very strong Stronger force Normal strength Weak strength No Value range 0.75～1 0.5～0.75 0.25～0.5 0～0.25 0

[0139] 3.2 Impact of traffic status differences on road network traffic carrying capacity

[0140] The calculation process of the road network traffic carrying capacity can be converted into two steps: the construction of the road network basic unit traffic carrying capacity calculation model and the establishment of the relationship between the basic unit traffic carrying capacity and the road network traffic carrying capacity. The present invention establishes the road network traffic carrying capacity calculation model based on the basic unit traffic carrying capacity calculation model. During the operation of traffic flow, since the overall traffic state of the road network is not stable and uniform, the average travel speed between the basic units may be different, which in turn affects the establishment of the direct mapping relationship between the average travel speed of the road network and the road network traffic carrying capacity.

[0141] After the basic unit traffic carrying capacity is determined, the next step is to analyze the relationship between the basic unit traffic carrying capacity and the road network traffic carrying capacity, and then construct a road network traffic carrying capacity calculation model. During the operation of traffic flow, the overall traffic state of the road network is not stable and uniform, and the average travel speeds between basic units may be different, resulting in the same road network average travel speed corresponding to different basic unit operation conditions. Specifically, the first case is that the average travel speed of the road network is the same as the average travel speed of each basic unit. In this scenario, the operation state of the entire road network traffic is uniform, so it is called a homogeneous traffic state; the second case is that the average travel speed of the road network is different from the average travel speed of each basic unit. In this scenario, the speed of some basic units is higher than the road network speed, while some are lower than the road network speed. The operation state of the road network traffic is uneven, so it is called a non-homogeneous traffic state.

[0142] In a homogeneous traffic state, since the average travel speeds at the road network level and the basic unit level are the same, the results will be the same whether the road network speed or the basic unit speed is used to determine the road network traffic carrying capacity. In other words, according to the service level mapping attribute of traffic carrying capacity, in a homogeneous state, the road network traffic carrying capacity is equal to the sum of the traffic carrying capacities of each basic unit obtained based on the average travel speed of the road network, and the average travel speed at the road network level has an intuitive mapping relationship with the traffic carrying capacity.

[0143] like Figure 6 As shown in Figure 1, under non-homogeneous traffic conditions, the average travel speed of each basic unit is different, and the corresponding basic unit traffic carrying capacity will also be different, and the traffic carrying capacity of the local road network may change. In this case, it is impossible to know the average travel speed of each basic unit based on the speed of each basic unit. The traffic carrying capacity of the road network A is obtained net,each and average travel speed based on the road network The traffic carrying capacity of the road network A is obtainednet,average The relationship between them cannot be determined with the road network traffic carrying capacity A net,each The corresponding average travel speed of the road network, and thus the mapping relationship between the road network traffic carrying capacity and the average travel speed of the road network cannot be directly established. However, according to the basic definition and modeling core of the road network traffic carrying capacity, it aims to establish a one-to-one mapping interaction between the average travel speed of the road network and the maximum number of operating vehicles in the road network. When the reader specifies the expected average travel speed at a road network level, the corresponding road network traffic carrying capacity can be determined; if only the mapping relationship between the average travel speed at the basic unit level and the traffic carrying capacity is known, the interaction relationship between the average travel speed at the road network level and the traffic carrying capacity cannot be determined, and the established modeling goals cannot be achieved, making it difficult to carry out subsequent practical applications. Therefore, it is crucial to explore the impact of non-uniform traffic conditions on the road network traffic carrying capacity.

[0144] Construction and verification of calculation model for road network traffic carrying capacity

[0145] Regardless of the traffic status of the road network, the road network traffic carrying capacity is the sum of the basic unit traffic carrying capacities calculated based on the average travel speed of the road network. Alternatively, each basic unit can first use its own average travel speed to calculate its own traffic carrying capacity, and then the cumulative value of the carrying capacity of each basic unit is also the road network traffic carrying capacity. Based on the above, it can be seen that under a certain average travel speed of the road network, the calculation formula of the road network traffic carrying capacity can be expressed as:

[0146]

[0147] The constraints are:

[0148]

[0149] Where: A represents the traffic carrying capacity of the road network; represents the average travel speed of the road network; Represents the traffic proportion in the tth direction. The meanings of other variables are the same as above.

[0150] The present invention uses VISSIM software based on Visual Basic programming to build a traffic experiment scene, conducts a simulation of the urban road network, and verifies the effectiveness of the proposed model by comparing the traffic carrying capacity obtained based on the simulation with the traffic carrying capacity obtained based on the invention method. To this end, two example road networks are constructed, and the specific situation is as follows:

[0151] (1) Road network 1: This road network consists of 24 road sections and 12 nodes, of which 8 external nodes are used to meet traffic needs and 4 internal nodes are controlled by traffic signals. The north-south direction has eight lanes in both directions and the east-west direction has six lanes in both directions. The cycle of each node is 160s, and the green-signal ratios for straight driving and left turning in the north-south direction are 0.30 and 0.20 respectively, and the corresponding green-signal ratios in the east-west direction are 0.23 and 0.17 respectively.

[0152] (2) Road Network 2: This road network consists of 80 road sections and 32 nodes, of which 16 external nodes are used to provide traffic demand and 16 internal nodes are controlled by traffic signals; all directions are two-way six-lane; the cycle of the external nodes is 180s, and the green signal ratios for straight and left turns are 0.28 and 0.22 respectively; the cycle of the internal nodes is 140s, and the green signal ratios for straight and left turns are 0.29 and 0.21 respectively. The determination of the green signal ratios in different directions is based on the road network structure design, traffic flow input and actual traffic survey experience, and all nodes are conventional four-phase.

[0153] When traffic flows in a road network, the number of vehicles cannot remain constant, but it will not change suddenly. Fluctuations in the number of vehicles in the road network are usually accompanied by changes in the average travel speed. This is the process of changes in the traffic state of the road network. When inputting traffic flow in a simulated road network, different combinations of traffic flow at the input endpoints will lead to differences in the number of vehicles running on the road network, which in turn will lead to the situation where the same average travel speed may correspond to multiple numbers of vehicles. However, each average travel speed will only correspond to a maximum number of vehicles, that is, the traffic carrying capacity of the road network. Among them, the road network will have a critical average travel speed under high-speed conditions. In this case, as long as the number of vehicles does not exceed n limit , the traffic can always travel freely at this speed. This phenomenon is also consistent with the actual traffic situation.

[0154] Step 4: Traffic carrying capacity warning for the city's central business district

[0155] Early warning is "pre-warning". It is a process of determining the state of a certain system that deviates from the warning line and issuing a warning signal. Early warning also includes two aspects, prevention and control, based on prediction. First, regular monitoring of target events and risk assessment of the state of events to prevent abnormal operation of the situation. This layer of meaning focuses on prevention and monitoring; second, the target event causes a major crisis due to risk accumulation or amplification, or due to an emergency, and the crisis needs to be regulated to resolve the crisis and restore normal operation. This layer of meaning is mainly to control and eliminate the crisis. Before the alarm occurs, the development laws and specific precursors of past things can be summarized to judge and predict the changing trends of things, and a forecast warning can be issued to related parties, so as to ensure that the warning subject takes timely response measures to avoid losses or minimize losses.

[0156] The early warning of traffic carrying capacity in the central business district is to apply the principles and methods of early warning to the traffic carrying capacity of the commercial district. This is a complex system that integrates many factors such as land spatial structure, urban road network scale, road network hierarchy structure, traffic mode structure, internal parking system and public transportation system, and traffic management policy. From a macro perspective, the content of the early warning of the carrying capacity of the central business district of the city includes the design of early warning index system, early warning evaluation, early warning level identification and early warning response strategy. In the present invention, an early warning index system will be established through time warning and space warning respectively, and for the spatial warning of traffic carrying capacity, the indicators reflecting the warning situation will be standardized, and the fuzzy judgment method will be used to evaluate the carrying capacity of the central business district of the city and divide the early warning level, and finally the response strategy for each early warning level will be proposed.

[0157] Traffic carrying capacity time warning

[0158] Traffic carrying capacity time warning means that in a given area, the traffic carrying capacity of the area is first calculated, and then the traffic volume during peak hours is predicted. As time goes by, the traffic volume surges until it reaches the threshold of the traffic carrying capacity of the affected area. In this process, the average speed of the traffic flow is predicted using a method based on Seq2Seq-GCGRU and a correction coefficient, and then the warning level is divided into different speed ranges, so as to achieve the purpose of CBD traffic carrying capacity warning in time.

[0159] (1) Congestion characteristics of roads in the CBD-affected area

[0160] The degree of traffic congestion is described by the average speed of motor vehicles on the main roads of the CBD. Smooth traffic: the average speed of motor vehicles on the main roads of the CBD is not less than 48km / h; Slightly delayed: the average speed of motor vehicles on the main roads of the CBD is less than 48km / h but higher than 40km / h; Mildly congested: the average speed of motor vehicles on the main roads of the CBD is less than 40km / h but higher than 32km / h; Congested: the average speed of motor vehicles on the main roads of the CBD is less than 32km / h but higher than 24km / h; Severely congested: the average speed of motor vehicles on the main roads of the CBD is less than 24km / h.

[0161] (2) Traffic carrying capacity time warning model

[0162] The actual area of ​​roads in the CBD area is S, the actual operation time of the roads is T, and the time and space consumption of traffic during peak hours is C. i , and obtain the road network carrying capacity model.

[0163]

[0164] On this basis, the CBD compactness, actual road operation efficiency, parking system correction coefficient, public transportation correction coefficient and policy correction value are brought into the corrected road network carrying capacity, that is, the traffic carrying capacity model

[0165]

[0166] Among them, A is the traffic carrying capacity, and the other symbols are the same as above; A T is the traffic carrying capacity threshold of the road section in the CBD area. When the traffic volume on the road section satisfies the following formula, the road section i in the CBD will not be in a congested state.

[0167]

[0168] in, A is the traffic volume forecast value of road section i at (t+Δt); T Indicates traffic carrying capacity.

[0169] Let Ni represent the critical capacity of traffic flow on road section i, and (t+Δt)Ni represent the number of vehicles on road section i at (t+Δt), then:

[0170]

[0171] Where: Ni is the critical capacity of traffic flow on section i; (t+Δt)Ni is the number of vehicles on section i at (t+Δt); Li is the length of section i; vi is the critical speed of congested traffic on section i; (t+Δt)vi is the average speed of vehicles on section i at (t+Δt), which is obtained by the speed prediction based on the Seq2Seq-GCGRU algorithm and the correction coefficient below.

[0172] ①Speed ​​prediction based on Seq2Seq-GCGRU algorithm and correction coefficient

[0173] In order to predict the speed of the next cycle, a speed prediction model based on the Seq2Seq-GCGRU algorithm and correction coefficient is selected. This model can predict the degree of traffic congestion on a certain road section in the next period of time, capture the spatiotemporal characteristics of traffic speed, and analyze the periodicity and discreteness of traffic data at time points. The prediction principle can be summarized as: "vertical prediction, horizontal correction".

[0174] Longitudinal prediction: A sequence-to-sequence graph convolution gated recurrent unit (Seq2Seq-GCGRU) traffic speed prediction model is used, where the GRU consists of two gates, one for the reset gate r t , which is used to control the degree of neglect of the state information at the previous moment; the other is the update gate z t, which is used to control the degree to which the state information of the previous moment enters the current state. In order to extract the features of the time dimension and space dimension of traffic data at the same time, the matrix multiplication in GRU is replaced by graph convolution to obtain GCGRU.

[0175] The sequence-to-sequence (seq2seq) architecture is an encoder-decoder network, where the encoder is the encoder and the decoder is the decoder. In the encoding stage, the historical time series is taken as input, the initial state is set to 0, and the output is the intermediate state vector C, which is the last hidden layer state of the encoder; in the decoding stage, the intermediate state vector C output in the encoding stage is used as the initial state, and the input of the decoder is determined by the training method. A one-dimensional convolutional neural network is used to extract the feature information of the historical time series as the decoder input. Such feature information can be obtained in both the training and testing stages, solving the problem of different distributions between training and testing. In addition, due to the introduction of the feature information of the historical time series, the prediction accuracy of the model will also be improved to a certain extent. Therefore, the input of the decoder is expressed as follows:

[0176] X (l) = SeLU(θ * X (l-1) )

[0177] Where: * is a one-dimensional convolution operation; θ is the convolution kernel parameter; the activation function is SeLU.

[0178] A sequence-to-sequence graph convolutional gated recurrent unit (Seq2Seq-GCGRU) traffic speed prediction model is proposed to capture the spatiotemporal characteristics of traffic speed.

[0179] The model is used to extract the weekly, daily and near-term information of traffic speed. The input of the three parts are the weekly data x w , daily cycle data x d and the near-term data x h , three sequence-to-sequence architectures with the same structure are constructed using the GCGRU network to obtain the Seq2Seq-GCGRU architecture.

[0180] Horizontal correction: The traffic structure in the prediction stage m+1 will affect the speed. Therefore, it is important to quantitatively reflect factors that affect the speed, such as the traffic structure, and it is also the most critical step in determining the correction coefficient. To reflect invisible factors such as the traffic structure, it is necessary to trace back to the road network speed before that moment. When we correct the lateral factors, we use the correction coefficient σ as the average relative error of the lateral prediction. Under this traffic structure, the greater the speed fluctuation, the greater the average relative error, and the prediction accuracy must be improved comprehensively; on the contrary, if the average relative error is small, it means that the speed is relatively stable under the traffic composition structure, and only the prediction accuracy needs to be fine-tuned. Therefore, a prediction model based on "longitudinal prediction, lateral correction" is proposed:

[0181]

[0182] Where: is the speed prediction value; v t is the longitudinal predicted value of speed; σ is the correction coefficient.

[0183] The steps of CBD traffic carrying capacity time warning are as follows: first, calculate the CBD area traffic carrying capacity, then use the traffic department's equipment to measure the speed of traffic on each road section in the CBD in real time at different time periods of the day, and predict the average speed of traffic flow on each road section in the next time period through the collected speed data, and then divide the warning situation according to the specific speed range corresponding to the predicted speed, and finally use the warning device to predict and warn the drivers and passengers of the traffic situation, such as Figure 7 shown.

[0184] The warning situations corresponding to the predicted speed are divided into the following table:

[0185]

[0186] 4.2 Spatial warning of traffic carrying capacity

[0187] The spatial early warning of traffic carrying capacity is an early warning of the reasonable distribution of traffic flow at each section node in the CBD area to the road network. Mainly based on the distribution of traffic flow, the traffic carrying capacity of each main radiating section in the CBD area is obtained, and then the obtained traffic carrying capacity is compared with the traffic carrying capacity of the area. According to the relationship between traffic carrying capacity and traffic carrying volume, the present invention uses the traffic carrying capacity saturation indicator to reflect the warning situation of the traffic carrying capacity of the central business district. Traffic carrying capacity saturation refers to the degree to which the traffic carrying capacity of the city CBD meets the traffic demand at a certain stage of development. Its size is mainly determined by the size of the traffic carrying volume and the size of the traffic carrying capacity. The calculation method is the ratio of traffic carrying volume to traffic carrying capacity in a specific period, which reflects the adaptive relationship between carrying capacity and carrying volume.

[0188] (1) Traffic flow distribution in CBD area

[0189] Assume that a worker lives in R i and R i+1 In the fan-shaped area between i , or use R i+1 There must be a point (r,θ) on the circular arc with a distance r from the CBD, so that the time taken by the staff living at this point to go to the CBD is the same no matter which road they choose. Such a set of points will form a boundary contour Bi. Bi divides the staff living in the sector into two groups: i , or use R i+1 There are two major categories, so the traffic flow assignment problem becomes a problem of determining Bi.

[0190] Assume θi(r) is R i With B i At a distance r from CBD, B i The points on must satisfy the following conditions:

[0191]

[0192] Where: Φi is R i With R i+1 v(r) is the speed on the circular path at a distance r from the CBD; Ti(r) is the speed at R i The running time from the upper CBD to a point with a distance r from it (or vice versa).

[0193] This formula shows the running time from the residence (located on Bi) to the CBD. After sorting out the above formula and solving it, the target equation is given as follows:

[0194]

[0195] This is the basic equation for determining the traffic flow allocation problem, but to apply it, we must first find the exact expression of Ti(r). Since the speed on the radial main road depends only on the local traffic flow, in R i Any point on The passing speed is given by:

[0196] v j (x) = V0[1+A j q j (x) -1 ]

[0197] Where: vi(x) is R i The speed of the upper side when the distance from CBD is x; qi(x) is R i The traffic flow at x on the iThe velocity (flow gradient). And:

[0198]

[0199] Where: qi(x) is R i The traffic flow at a distance x from the city center. Under the equilibrium assumption, the traffic flow at x of Ri is proportional to the probability Pi(x) at that location. Pi(x) indicates that the residence is greater than x and uses R i The proportion of CBD workers. The area where the residence is located between Bi and Bi+1 radius x and R satisfies:

[0200]

[0201] Assuming that the proportional constant in this example is the global traffic flow generation rate, absorbed into the gradient parameter Oi, then:

[0202]

[0203] Finally we get:

[0204]

[0205] In particular, when f(ζ,τ) is symmetric (i.e. When f(ζ,r1)=f(ζ,τ2)), it becomes:

[0206]

[0207] This is the traffic flow distribution model of each radial road section in the CBD area, and the total traffic carrying capacity model of the CBD area is:

[0208]

[0209] (2) Spatial early warning model of traffic carrying capacity

[0210] The traffic carrying capacity saturation is introduced as a warning indicator, that is, the ratio of traffic carrying volume to traffic carrying capacity:

[0211]

[0212] Where: P is the traffic carrying capacity saturation; F is the traffic carrying capacity of the CBD area; T C For traffic carrying capacity.

[0213] Warning device control unit

[0214] The early warning device control unit receives information about the number and direction of moving targets on the road from the road monitoring unit through the wireless communication module, and also controls the warning screen and audio playback device through data analysis to prompt and warn drivers entering the road section. The warning control unit uses the STM32F103ZET6 chip as the control core. On the one hand, it controls the wireless communication module to receive information transmitted by the road monitoring unit, and on the other hand, it controls the roadside warning equipment. The early warning device calculates the traffic carrying capacity of the roads in the CBD area and then issues an early warning to improve the traffic system in the urban commercial area. The main technical requirements are as follows:

[0215] (1) Control unit: Balances performance and power consumption, is cost-effective, and helps increase system popularity;

[0216] (2) Warning equipment: Use character display and voice broadcast to remind drivers of road information ahead, and have the secondary development capability to modify corresponding warning information in different scenarios;

[0217] (3) Communication method: Wireless communication is used to achieve data communication between the road monitoring unit and the warning control unit, avoiding installation restrictions on communication lines in complex environments;

[0218] (4) System scalability: The control platforms of the system monitoring and warning units are equipped with a certain number of data communication interfaces, and some communication interfaces are reserved to ensure the subsequent expansion and improvement of the system;

[0219] (5) Power supply system: to ensure normal operation of the system and enable the early warning system to work 24 hours a day. Early warning evaluation of traffic carrying capacity in the central business district

[0220] (1) Classification of traffic carrying capacity warning evaluation levels

[0221] The emergence of urban renewal projects is bound to be accompanied by changes in traffic flow. For example, if the road traffic project is not reasonably allocated for urban traffic flow, the new project may cause urban traffic to fall into a vicious cycle and cause new traffic jams. In order to quantitatively express the impact of existing or new commercial area projects in the city on the road traffic carrying capacity, the concept of traffic carrying capacity saturation is introduced. Traffic carrying capacity saturation is a standard for measuring the impact of the proposed project on the road traffic carrying capacity within its influence range. It reflects the impact of the traffic volume caused by the proposed project on the future road network, judges whether the impact is within the acceptable range of the road network, and issues an early warning, so as to propose effective improvement measures for the early warning level and reduce the traffic impact. The impact of traffic carrying capacity saturation is analyzed, that is, the traffic carrying capacity saturation is evaluated through the traffic model. When the proposed project is completed and put into use, the road carrying capacity within the influence range can be warned whether it can withstand the road carrying capacity that will be generated. Using traffic carrying capacity saturation to warn the project of the degree of traffic impact has two main characteristics: ① operability; ② comprehensiveness.

[0222] Since the determination of the impact of new commercial area projects on road network traffic saturation and the classification of road traffic service levels are the result of the interaction between traffic flow and traffic facilities, the classification of traffic carrying capacity saturation can refer to the classification standards of traffic service levels, and the specific classification is as follows:

[0223]

[0224] Traffic carrying capacity saturation is a comprehensive evaluation indicator that qualitatively reflects the impact of existing CBD or planned CBD projects on roads within the affected area. Drawing on the classification of urban road service levels, the warning levels can be divided into five levels: Level I, Level II, Level III, Level IV, and Level V, namely:

[0225]

[0226]

[0227] The present invention uses the indicator "traffic carrying capacity saturation" to measure the traffic carrying capacity of the existing CBD or the proposed CBD project in the target year. The warning limit division is to determine the range of changes in the traffic carrying capacity of the CBD project, and the traffic carrying capacity saturation is used as the warning limit division critical value and as the interval division standard, so as to divide the warning situation into 5 levels, namely: no warning, light warning, medium warning, heavy warning and over warning. Setting: When the traffic carrying capacity saturation of the existing CBD or the proposed CBD project is less than 0.6, it can be regarded as no warning; when the traffic carrying capacity saturation is greater than or equal to 0.6 but less than 0.7, it is regarded as a light warning; when the traffic carrying capacity saturation is greater than or equal to 0.7 but less than 0.9, it is regarded as a medium warning; when the traffic carrying capacity saturation is greater than or equal to 0.9 but less than 1, it is regarded as a heavy warning, and when the traffic carrying capacity saturation is greater than or equal to 1, it is regarded as a huge warning. According to this standard, the warning limit range of the traffic carrying capacity of the existing CBD or the proposed CBD project can be obtained.

[0228] (2) Evaluation method based on attribute measurement function

[0229] Let X be the entirety of the research objects, which is the research object space, referred to as the object space. F is a certain property of the object space in X, called the attribute space. For example, X = {all roads in the city}, the element x in X represents a certain road. If the service level of the road needs to be studied, then F can be set to {service level}. An attribute set refers to any case in the attribute space F. For example, A = {Grade A}, B = {Grade B}, A and B are both cases in the attribute space F, therefore, A and B are both attribute sets and can be regarded as subsets of F. If (C1, C2,…, Cn) are N attribute sets in the attribute space, if {C1, C2,…, Cn} satisfies Then {C1, C2, ..., Cn} is called the partition of the attribute space, C k (1≤k≤n) is called an attribute set. The elements in X, that is, the evaluated object x has the attribute or level C k The degree of xk =δ(x∈C k ). According to the attribute set and attribute measurement theory, δ xk Should meet:

[0230]

[0231] For the evaluated object x in index I, X, its measurement value t has attribute or level C k The degree of xk =δ(x∈C k ) should be determined according to specific problems, data, expert experience and certain mathematical processing methods. The commonly used method is to determine the attribute measurement function to express the change of the attribute measurement δ when the measurement value t of indicator I changes. xk =δ(x∈Ck )The changing situations that will occur. The determination of the attribute measurement function needs to be specifically analyzed according to the actual problem. The following is to establish relevant attribute measurement functions for the general situations in the above table for illustration.

[0232] Index level division table

[0233]

[0234] Among them, a k satisfies a1 < a2 < … < a n-1 , let:

[0235] b k =(a k +a k+1 ) / 2, k = 1, 2,... s, n - 2

[0236] d = a p -b p -min{a k -b k , k - 1, 2,... s, n - 2}

[0237] Suppose the index value of x is t, and the index attribute measurement function δ k (t) established from the above table is as follows:

[0238]

[0239] If d = a p -b p , p > 1, then

[0240]

[0241] ②Attribute recognition analysis: The purpose of attribute recognition analysis is to make an effective judgment on which evaluation level x belongs to through the attribute measurement δ xk (1 ≤ k ≤ n). In the problem of comprehensive evaluation, the evaluation set (C1, C2,..., Cn) is usually an ordered class. If the evaluation set (C1, C2,..., Cn) is a partition of the attribute space F and satisfies C1 > C2 > ··· > Cn or C1 < C2 < ··· < Cn, then (C1, C2,..., Cn) is called an ordered partition of the attribute space F. For the ordered evaluation set (C1, C2,..., Cn), if it is necessary to identify the category of x, the confidence criterion can be used.

[0242] Suppose the evaluation set (C1, C2,..., Cn) is an ordered partition, C1 > C2 > ··· > Cn or C1 < C2 < ··· < Cn, λ is the confidence level, 0.5 < λ ≤ 1. If:

[0243]

[0244] Then x belongs to C k0 level.

[0245] The above criteria require that the "strong" level accounts for a considerable proportion. In applications, the confidence level λ is generally taken between 0.6 and 0.7.

[0246] (3) Traffic carrying capacity warning model based on attribute measurement function

[0247] According to the basic theory of attribute measurement function mentioned above and combined with the actual research, in this model, the research object space X = {central business district road network}, the attribute space F = {traffic carrying capacity warning}, the indicator is the traffic carrying capacity saturation, and from the previous division of traffic carrying capacity influence, it can be seen that it is divided into five levels from small to large: Ⅰ, Ⅱ, Ⅲ, Ⅳ, Ⅴ, and C1, C2, C3, C4, C5 are used to represent these five levels. Then (C1, C2, C3, C4, C5) is the ordered partition set of the attribute space F, and C1>C2>C3>C4>C5, which is a strong order partition.

[0248] The grading standards of the indicators have been defined above, and are shown in the following table:

[0249]

[0250] ① Based on the attribute measurement function evaluation method, the attribute measurement function model is based on the traffic carrying capacity saturation index of the city center business district, and combined with the data in the above table to establish five indicator attribute measurement functions as follows:

[0251]

[0252] ② Attribute identification: Substituting the traffic carrying capacity saturation value into the above function, the corresponding attribute measurement can be obtained, and then the attribute identification analysis is performed according to the confidence criterion (taking λ = 0.6). The warning level of the project traffic carrying capacity under the traffic carrying capacity saturation can be obtained from the following formula.

[0253]

[0254] In order to solve this problem, the present invention converts the intermittent traffic flow controlled by signals into continuously released traffic flow while keeping the number of vehicles unchanged. After the equivalence, the corresponding lanes can continuously release the traffic flow, and the control process is jointly determined by the number of import lanes and the green-signal ratio of the intersection. The core idea of ​​equivalent lanes is to realize the conversion of discontinuous traffic flow to continuous traffic flow based on the principle of constant average travel speed under the premise of the same number of vehicles. This transformation greatly simplifies the complexity of the traffic carrying capacity calculation process on the basis of ensuring the theoretical correctness of the equivalent concept. At the same time, in terms of time warning, in order to predict the speed of the next cycle, a speed prediction model based on the Seq2Seq-GCGRU algorithm and the correction coefficient is selected. The model can predict the average driving speed of a certain section of road in the next stage, capture the spatiotemporal characteristics of traffic speed, and analyze the discreteness of traffic data at periodicity and time points. In terms of spatial warning, emphasis is placed on the temporal distribution characteristics of traffic flow and the rational use of road space resources. The traffic flow at each section node in the CBD area is rationally allocated to the road network, and the size of the traffic carrying capacity saturation is introduced as a warning indicator. When the traffic flow exceeds the threshold, the system will issue a warning to remind traffic managers to take timely measures to alleviate traffic pressure.

[0255] Through real-time monitoring and early warning, traffic managers can promptly identify traffic problems and take effective measures to guide and manage them, effectively improving the efficiency of urban traffic and the travel experience of citizens. At the same time, this early warning equipment also provides strong support for urban planners and decision makers, helping them to plan and build urban traffic more scientifically.

[0256] In summary, the present invention has constructed a comprehensive and scientific early warning indicator system for traffic carrying capacity in urban central business districts. This system can not only help us better understand and grasp the laws of urban traffic operation, but also provide strong support for urban traffic management and planning.

[0257] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for early warning of the carrying capacity of urban central roads, characterized in that: The following steps are involved: A Obtain traffic data of the central roads in the city, determine the equivalent lanes, and convert the intermittent traffic flow controlled by signals on the roads into continuous release traffic flow while keeping the number of vehicles unchanged. The lanes that can continuously release traffic flow after equivalent are the equivalent lanes; B. Calculating the traffic carrying capacity of the basic unit based on the converted equivalent traffic flow. The equivalent lane area in the basic unit can be divided into a stacked release area and a free driving area. By using the mapping relationship between the equivalent traffic flow and the non-equivalent traffic flow, the traffic carrying capacity of the non-equivalent lane is calculated and the temporal and spatial correlation between the equivalent lane and the non-equivalent lane is combined. C established the relationship between the number of lanes in the road network structure parameters and the green-to-signal ratio in the signal timing parameters, determined the relationship between the maximum number of operating vehicles in the basic unit and the road network structure parameters, signal timing parameters, and average travel speed, established a road network global traffic carrying capacity calculation model, input the data of the city center roads to be tested into the prediction results of the road network global traffic carrying capacity calculation model, and carried out risk warning.

2. According to claim 1, a method for early warning of the carrying capacity of urban central roads is characterized in that: In step C, a basic unit travel time model is constructed, and the relationship between the section length and the average travel speed in the road network structure parameters is established. The travel time of the vehicle flow can be divided into free travel time and stack release time, and the travel time calculation model of each section is constructed. The formula is as follows: Where: L u Indicates the length of the basic unit u; represents the distance traveled at free flow speed in the basic unit u; and They represent the length of the road section and the length of the intersection in the basic unit u respectively; n represents the number of vehicles; represents the expected average travel speed of the basic unit u; Among them, the average travel time of the traffic flow in the basic unit u is T u Can be calculated based on the length of the basic unit and the average travel speed; average free travel time It can be determined based on the length of the free flow section and the free flow speed; average cascade release time The calculation is made using the time relationship between different sections of the basic unit.

3. According to claim 1, a method for early warning of the carrying capacity of urban central roads is characterized in that: The calculation of the road network global traffic carrying capacity calculation model in step C includes the following steps: According to the operation rules of traffic flow in different areas, the traffic carrying capacity of the stacking release area and the free driving area are calculated respectively; Based on the interaction between the stacked release area and the free travel area, the traffic carrying capacity of the equivalent lane is determined; By analyzing the mapping relationship between equivalent traffic flow and non-equivalent traffic flow, the traffic carrying capacity of non-equivalent lanes is calculated; Combining the temporal and spatial correlation between equivalent lanes and non-equivalent lanes, the basic unit traffic carrying capacity is derived. The calculation formula of the basic unit traffic carrying capacity can be expressed as: Where: B represents the set of basic traffic units in the road network, B = {1, 2, 3, ..., u, ...}, which includes the upstream section driving area R ij and downstream intersection form area R j , A u represents the traffic carrying capacity of the u-th basic unit in the road network, u∈B, and They represent the traffic carrying capacity of the equivalent lane and the traffic carrying capacity of the non-equivalent lane in the u-th basic unit respectively; and They respectively represent the traffic carrying capacity of the stacked release area and the traffic carrying capacity of the free driving area in the u-th basic unit.

4. The method for early warning of the carrying capacity of urban central roads according to claim 1, characterized in that: The calculation formula of the road network traffic carrying capacity can be expressed as: The constraints are: Where: A represents the traffic carrying capacity of the road network; It represents the average travel speed of the road network; Represents the traffic proportion in the tth direction.

5. The method for early warning of the carrying capacity of urban central roads according to claim 1, characterized in that: The road network global traffic carrying capacity calculation model has a built-in traffic carrying capacity time warning model, which uses the actual area of ​​CBD area roads S, the actual operation time of the roads T and the time and space consumption of traffic during peak hours C. i , and the road network carrying capacity model is obtained: Substituting CBD compactness, actual road operation efficiency, parking system correction coefficient, public transportation correction coefficient and policy correction value into the corrected road network carrying capacity, i.e., the traffic carrying capacity model, the formula is as follows: Among them, A is the traffic carrying capacity; A T is the traffic carrying capacity threshold of the road section in the CBD area. When the traffic volume on the road section satisfies the following formula, the road section i in the CBD will not be in a congested state; in, A is the traffic volume forecast value of road section i at (t+Δt); T Indicates traffic carrying capacity; Let Ni represent the critical capacity of traffic flow on road section i, and (t+Δt)Ni represent the number of vehicles on road section i at (t+Δt), then: Where: Ni is the critical capacity of traffic flow on section i; (t+Δt)Ni is the number of vehicles on section i at (t+Δt); Li is the length of section i; vi is the critical speed of congested traffic flow on section i; (t+Δt)vi is the average speed of vehicles on section i at (t+Δt); The speed prediction based on Seq2Seq-GCGRU algorithm and correction coefficient is used to predict the traffic data of a certain road section in the next stage. The formula is as follows; Among them, Φ represents the probability distribution of the prediction results, is the predictor variable, are input features, represents the training set, is the set of model parameters of the regression model, Λ is the variational distribution of the model parameters of the regression model, Goodness of fit: Where Ξ represents the goodness of fit, ω is the parameter that adjusts the weight between the variational distribution and the likelihood function, ν is the model parameter, i is the index of ν, ρ is the variational distribution, Υ is the uniform prior distribution, Π is the likelihood function, ∈ is the uncertainty parameter, Probability density function: Among them, τ n is the noise term, is the predicted value of the regression model, α is the adjustment function, Goodness of fit for prediction uncertainty: Use the test set to test the accuracy of the regression model.

6. The method for early warning of the carrying capacity of urban central roads according to claim 1, characterized in that: During risk warning, the wireless communication module of the warning device control unit receives information including the number and direction of moving targets on the road from the road monitoring unit, and after analysis and processing, controls the warning screen and audio playback device to prompt and warn drivers entering the road section; The warning control unit uses the STM32F103ZET6 chip as the control core, which controls the wireless communication module to receive information transmitted by the road monitoring unit and control the roadside warning equipment.

7. An electronic device comprising: processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, enables the electronic device to execute the method according to any one of claims 1 to 6.

Images