Road congestion propagation prediction method based on double-layer hypergraph model

Abstract

The invention relates to the technical field of intelligent traffic, and provides a road congestion propagation prediction method based on a double-layer hypergraph model, and the method comprises the steps: constructing an FC-ISQ double-layer traffic congestion propagation model based on a hypergraph; determining a network layer of the model as a traffic road layer and a traffic information spreading layer; establishing a heterogeneous differential equation set of the model and verifying the positive property and the bounded property of a model solution; calculating a basic regeneration number R0 of the model, and taking the basic regeneration number R0 as a threshold value for judging whether the traffic jam is automatically dissipated or not; performing numerical simulation verification on the model; through simulation, prediction results of each group of the model under different initial parameter settings are obtained, and visual prediction of propagation trends and propagation conditions of traffic congestion under different conditions is realized. According to the invention, by constructing the FC-ISQ double-layer traffic jam propagation model based on the hypergraph, the dynamic propagation rule of the traffic jam state is systematically revealed.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent transportation, and particularly relates to a prediction method for road congestion propagation based on a double-layer hypergraph model. Background Art

[0002] Urban road traffic problems are closely related to residents' daily travel and are key elements indispensable in urban development and construction. In the process of modern urbanization, with the popularization of information technologies such as major map navigation Apps, people can obtain the traffic conditions of each region in real time through these tools, so as to better plan and execute travel plans. In addition, the rise of social media platforms has also provided diversified channels for the dissemination of urban traffic information, enabling residents to quickly share the real-time situation of the travel area and obtain real-time dynamic information of the travel roads.

[0003] However, the process of urban road construction often needs to comprehensively consider many factors, including sidewalk design, traffic signal layout, bridge planning, and geographical environment, etc. These factors are interrelated, making the road congestion problem show complex systematic characteristics. The existing technologies fail to fully consider the systematic correlation of multi-dimensional factors and lack in-depth analysis of the influence of group behavior. Traditional models are too simplified to handle the dynamics of complex systems. Summary of the Invention

[0004] In view of the above-mentioned defects of the existing technologies, the present invention proposes a prediction method for road congestion propagation based on a double-layer hypergraph model. The technical solution designed by the present invention includes the following steps:

[0005] S1: Construct a double-layer traffic congestion propagation model of FC-ISQ based on a hypergraph;

[0006] S2: Determine the network layer of the double-layer traffic congestion propagation model of FC-ISQ as the traffic road layer and the traffic information dissemination layer;

[0007] S3: Establish a heterogeneous differential equation system of the double-layer traffic congestion propagation model of FC-ISQ and verify the positivity and boundedness of the model solution;

[0008] S4: Calculate the basic reproduction number R 0 , and use it as the threshold for judging whether traffic congestion dissipates automatically;

[0009] S5: Conduct numerical simulation verification on the double-layer traffic congestion propagation model of FC-ISQ;

[0010] S6: Obtain the prediction results of each group of the double-layer traffic congestion propagation model of FC-ISQ under different initial parameter settings through simulation, and realize the visual prediction of the propagation trend and propagation situation of traffic congestion in different situations.

[0011] Preferably, the traffic road layer in S2 includes:

[0012] Classify the traffic road layer into free links F k and congested links C k ;

[0013] Introduce the concept of a hypergraph in the traffic road layer to refine the classification of road connection types. The intersections of the roads represent nodes, and any number of nodes are connected by a hyperedge, indicating the mutual relevance of multiple links under urban road construction; only consider unidirectional flow links. Based on the number of unidirectional out-flow links at the intersection and the size of the hyperedge, the refined classification of the road is expressed as k i = [k i 2 , k i 3 , k i 4 ,..., k i J , where k i J represents the number of intersections with a hyperedge size of, in the case of i out-flow links;

[0014] When the traffic flow increases, the free link F k becomes a congested link C with a probability of β 3 ; k ;

[0015] When the traffic flow decreases, the congested link C k becomes a free link F with a probability of φ 4 ; k .

[0016] Preferably, the traffic information dissemination layer in S2 includes:

[0017] Unknown information holders, travel information dissemination promoters S TP , congestion information dissemination promoters S TC and known information keepers Q;

[0018] When an unknown information holder learns about the traffic conditions of the road, it becomes a travel information dissemination promoter S with a probability of β 1 ; TP ;

[0019] When an unknown information holder learns about the traffic conditions of the road, it becomes a congestion information dissemination promoter S with a probability of β 2 ; TC ;

[0020] When a travel information dissemination promoter S TP stops disseminating the traffic conditions of the road, it becomes an unknown information holder with a probability of φ 1The probability of becoming a silent person Q with known information;

[0021] When the congestion information propagator S TC stops propagating the traffic conditions of the road, with a probability of φ 2 becomes a silent person Q with known information;

[0022] When the congestion information propagator S TC propagates the traffic conditions of the road, the congested link C k with a probability of φ 3 becomes a free link F k .

[0023] Preferably, the heterogeneous differential equations in S3 are as follows:

[0024] dF k (t)=[φ 3 C k (t)S TC (t)+kφ 4 Θ(t)C k (t)-β 3 F k (t)S TP (t)]dt

[0025] dC k (t)=[β 3 F k (t)S TP (t)-φ 3 C k (t)S TC (t)-kφ 4 Θ(t)C k (t)]dt

[0026] dI(t)=[Λ - β 1 I(t)S TP (t)-β 2 I(t)S TC (t)-μI(t)]dt

[0027] dS TP (t)=[β 1 I(t)S TP (t)-φ 1 S TP (t)-μS TP (t)]dt

[0028] dS TC (t)=[β 2 I(t)S TC (t)-φ 2 S TC(t)-μS TC (t)]dt

[0029] dQ(t)=[φ 1 S TP (t)+φ 2 S TC (t)-μQ(t)]dt

[0030] in,

[0031]

[0032] In the formula, F k (t) is the free link F with excess k at time t k The number of k (t) is the congested link C that exceeds k at time t k The number of TC (t) is the person S who spreads the congestion information at time t TC , Θ(t) is the number of hyperedges pointing to free links F k The probability of S TP (t) is the person S who spreads the information promoting travel at time t TP , I(t) is the number of unknown information at time t, Λ is the migration rate of the traffic information dissemination layer group, μ is the migration rate of the traffic information dissemination layer group, Q(t) is the number of Q who remain silent at time t, ω j is the propagation weight of the hyperedge size j, k j is the superdegree corresponding to the starting node of the link, P(k j ) is for salvation k j The probability of the distribution, <k j > is the average hyperdegree corresponding to the hyperedge size j, J is the maximum hyperedge order, R is the maximum number of outgoing links of the starting node, is the free link F with the number of outgoing links i and hyperedge order j of the starting node k at time t k The number of is the total number of links with outgoing links from starting node i and hyperedge order j.

[0033] Preferably, verifying the positivity and boundedness of the model solution in S3 includes:

[0034] Assume that N1k(t) represents the total number of roads with starting node type k within the research scope at time t, N1 represents the total number of roads within the research scope, and N2(t) represents the total number of traffic information dissemination layer groups at time t, then:

[0035]

[0036] N 2(t) = I(t) + S TP (t) + S TC (t) + Q(t)

[0037] Assuming that all parameter values are non - negative, we have:

[0038]

[0039] All solutions are bounded, and the feasible region of the model is:

[0040]

[0041] where Γ is the feasible region of the model.

[0042] Preferably, the S4 includes:

[0043] Set the heterogeneous differential equations to zero, calculate the equilibrium points of the FC - ISQ double - layer traffic congestion propagation model, and calculate the basic reproduction number R 0 and use it as the threshold for judging whether traffic congestion will dissipate automatically;

[0044] The equilibrium points include the non - congestion equilibrium point E 0 , the boundary equilibrium point E 1 , the boundary equilibrium point E 2 and the congestion - persistent equilibrium point E * .

[0045] Preferably, the calculation of the basic reproduction number R 0 and using it as the threshold for judging whether traffic congestion will dissipate automatically includes:

[0046] Using the next - generation matrix method, obtain the inflow part F and the outflow part V composed of S TP , S TC and C k , which are respectively expressed as:

[0047]

[0048] The input matrix F and the transfer matrix V are the Jacobian matrices corresponding to F(S TP , S TC , C k ) and V(S TP , S TC , C k );

[0049] The corresponding largest eigenvalue is:

[0050]

[0051] where, in the case of no congestion threshold

[0052] When R 0 ≤ 1, there is a congestion-free equilibrium point C in the FC-ISQ two-layer traffic congestion propagation model. k

[0053] Beneficial effects:

[0054] 1. The two-layer traffic congestion propagation model constructed based on the hypergraph theory in the present invention adopts a method combining hypergraph and complex network theories, and conducts a more comprehensive macroscopic analysis of traffic congestion phenomena. Considering the multi-dimensional influencing factors existing in the process of urban road construction, it clearly expounds the hypergraph structure of the traffic road layer, and constructs the traffic information propagation layer from the characteristics of group behavior, highlighting the coupling relationship between the two-layer networks;

[0055] 2. By constructing a state transition diagram, the present invention establishes a set of dynamic equilibrium point analysis systems, and obtains the prediction results of each group under different initial parameter settings through simulation, realizing the visual prediction of the propagation trend and propagation situation of traffic congestion in different situations. This system can effectively analyze the existence of congestion-free and persistent congestion states, provide a more accurate theoretical basis for the traffic congestion prediction model, and provide scientific decision-making support for urban traffic planning in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a schematic flowchart of a preferred embodiment of the present invention;

[0057] Figure 2 is a schematic diagram of the hypergraph network of the traffic road layer of a preferred embodiment of the present invention;

[0058] Figure 3 is a schematic diagram of the model state transition of a preferred embodiment of the present invention;

[0059] Figure 4 is a schematic diagram of the congestion-free equilibrium point E 0 of a preferred embodiment of the present invention;

[0060] Figure 5 is a schematic diagram of the boundary equilibrium point E 1 of a preferred embodiment of the present invention;

[0061] Figure 6 is a schematic diagram of the boundary equilibrium point E 2 of a preferred embodiment of the present invention;

[0062] Figure 7 is a schematic diagram of the traffic congestion persistent equilibrium point E * of a preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION ​

[0063] The following is a detailed description of the embodiments of the present invention. The following embodiments are implemented on the premise of the technical solution of the present invention, and detailed implementation manners and specific operation processes are given. However, the protection scope of the present invention is not limited to the following embodiments.

[0064] The present invention designs a prediction method for road congestion propagation based on a double-layer hypergraph model, systematically reveals the dynamic propagation law of traffic congestion states, provides a scientific basis for urban traffic planning and management, and at the same time provides theoretical support for the improvement of travel convenience, such as Figure 1-7 shown, the technical solution includes the following steps, specifically including:

[0065] S1: Construct a FC-ISQ double-layer traffic congestion propagation model based on a hypergraph;

[0066] S2: Determine the network layer of the FC-ISQ double-layer traffic congestion propagation model as the traffic road layer and the traffic information propagation layer;

[0067] S3: Establish a heterogeneous differential equation system for the FC-ISQ double-layer traffic congestion propagation model and verify the positivity and boundedness of the model solution;

[0068] S4: Calculate the basic reproduction number R 0 of the FC-ISQ double-layer traffic congestion propagation model and use it as a threshold for judging whether traffic congestion dissipates automatically;

[0069] S5: Conduct numerical simulation verification on the FC-ISQ double-layer traffic congestion propagation model;

[0070] S6: Obtain the prediction results of each group of the FC-ISQ double-layer traffic congestion propagation model under different initial parameter settings through simulation, and realize the visual prediction of the propagation trend and propagation situation of traffic congestion in different situations.

[0071] Preferably, the traffic road layer in S2 includes:

[0072] Classify the traffic road layer into free links F k and congested links C k ;

[0073] Introduce the concept of a hypergraph in the traffic road layer to refine the classification of road connection types. The intersections of roads represent nodes, and any number of nodes are connected by a hyperedge, indicating the mutual correlation of multi-links under urban road construction; only consider one-way flow links. Based on the number of one-way out-flow links at intersections and the size of hyperedges, the refined classification of roads is expressed as k i = [k i 2 , k i 3 , k i4 ,..., k i J , k i J represents the number of intersections with hyper - edge size of at intersections with i out - flowing links;

[0074] When the traffic flow increases, the free link F k becomes a congested link C with a probability of β 3 ; k

[0075] When the traffic flow decreases, the congested link C k becomes a free link F with a probability of φ 4 ; k

[0076] Preferably, the traffic information dissemination layer in S2 includes:

[0077] Unknown information holders, dissemination - promoting travel information holders S TP , dissemination - promoting congestion information holders S TC and known - information keep - silent holders Q;

[0078] When the unknown information holder learns the traffic conditions of the road, it becomes a dissemination - promoting travel information holder S with a probability of β 1 ; TP

[0079] When the unknown information holder learns the traffic conditions of the road, it becomes a dissemination - promoting congestion information holder S with a probability of β 2 ; TC

[0080] When the dissemination - promoting travel information holder S TP stops disseminating the traffic conditions of the road, it becomes a known - information keep - silent holder Q with a probability of φ 1 ;

[0081] When the dissemination - promoting congestion information holder S TC stops disseminating the traffic conditions of the road, it becomes a known - information keep - silent holder Q with a probability of φ 2 ;

[0082] When the dissemination - promoting congestion information holder S TC disseminates the traffic conditions of the road, the congested link C k becomes a free link F with a probability of φ 3 ; k

[0083] Preferably, the heterogeneous differential equation system in S3 is as follows:

[0084] dF k (t)=[φ 3 ​​​​​C k (t)S TC (t)+kφ 4 Θ(t)C k (t)-β 3 F k (t)S TP (t)]dt

[0085] dC k (t) = [β 3 F k (t)S TP (t)-φ 3 C k (t)S TC (t)-kφ 4 Θ(t)C k (t)]dt

[0086] dI(t)=[Λ-β 1 I(t)S TP (t)-β 2 I(t)S TC (t)-μI(t)]dt

[0087] dS TP (t) = [β 1 I(t)S TP (t)-φ 1 S TP (t)-μS TP (t)]dt

[0088] dS TC (t) = [β 2 I(t)S TC (t)-φ 2 S TC (t)-μS TC (t)]dt

[0089] dQ(t)=[φ 1 S TP (t)+φ 2 S TC (t)-μQ(t)]dt

[0090] in,

[0091]

[0092] In the formula, F k (t) is the free link F with excess k at time t k The number of k (t) is the congested link C that exceeds k at time t k The number of TC(t) is the number of people S who spread congestion information at time t TC , Θ(t) is the probability that any hyperedge points to the free link F k . TP (t) is the number of people S who spread travel-promoting information at time t TP ,,(t) is the number of people with unknown information at time t, Λ is the immigration rate of the traffic information dissemination layer group, μ is the emigration rate of the traffic information dissemination layer group, Q(t) is the number of people Q who keep silent about known information at time t, ω j is the propagation weight when the hyperedge size is j, k j is the hyperdegree corresponding to the starting node of the link, P(k j ) is the probability of the hyperdegree k j distribution, <k j > is the average hyperdegree corresponding to the hyperedge size j, J is the maximum hyperedge order, R is the maximum number of out-flowing links of the starting node, is the number of free links F with the out-flowing link number i of the starting node with hyperdegree k and hyperedge order j at time t k . is the total number of links with the out-flowing link number i and hyperedge order j of the starting node.

[0093] Preferably, verifying the positivity and boundedness of the model solution in S3 includes:

[0094] Let N1k(t) represent the total number of roads with the starting node type k within the research scope at time t, N1 be the total number of roads within the research scope, and N2(t) be the total number of the traffic information dissemination layer group at time t, then there is:

[0095]

[0096] N 2 (t) = I(t) + S TP (t) + S TC (t) + Q(t)

[0097] Assuming that all parameter values are non-negative, then there is:

[0098]

[0099] All solutions are bounded, and the feasible region of the model is:

[0100]

[0101] In the formula, Γ is the feasible region of the model.

[0102] Preferably, S4 includes:

[0103] Set the heterogeneous differential equations to zero, calculate the equilibrium points of the FC-ISQ two-layer traffic congestion propagation model, and calculate the basic reproduction number R 0 and use it as the threshold for judging whether traffic congestion dissipates automatically;

[0104] The equilibrium points include the non-congestion equilibrium point E 0 , the boundary equilibrium point E 1 , the boundary equilibrium point E 2 and the congestion persistence equilibrium point E * .

[0105] Specifically, for the non-congestion equilibrium point E 0 , the calculation is as follows:

[0106] Let S TP (t)=S TC (t)=Q(t)=C k (t)=0, and the solution is

[0107] Then

[0108]

[0109] In the formula, for I 0 , E 0 , and the superscript represents the population quantity at the corresponding non-congestion equilibrium point E 0 ; the subscript represents different values of k;

[0110] For the boundary equilibrium point E 1 , the calculation is as follows:

[0111] Let S TP (t)=0, C k (t)=0, and after arranging the equation, the following equation is obtained:

[0112]

[0113] The solution is:

[0114]

[0115] Then

[0116]

[0117] In the formula, for I 1 , Q 1 , and the superscript represents the corresponding boundary equilibrium point E 1The population quantity at, the subscript represents different values of k.

[0118] Boundary equilibrium point E 2 , calculated as follows:

[0119] Let S TC (t)=0, F k (t)=0, after arranging the equations, the following equations are obtained:

[0120]

[0121] The solutions are:

[0122]

[0123] Then

[0124]

[0125] In the formula, for, 2 , Q 2 , and The superscript represents the population quantity at the corresponding boundary equilibrium point E 2 The subscript represents different values of k.

[0126] Congestion persistent equilibrium point E * , calculated as follows:

[0127] The solutions are obtained from the system of equations:

[0128]

[0129] Among them

[0130]

[0131] Then

[0132]

[0133] In the formula, for I * , S TC * , S TP * , A * , Θ * , Q * , and The superscript represents the population quantity at the corresponding congestion persistent equilibrium point E * The subscript represents different values of k.

[0134] Preferably, the basic reproduction number R is calculated 0 and used as a threshold for judging whether traffic congestion will automatically dissipate, including:

[0135] Using the next-generation matrix method, the inflow part F and the outflow part V composed of S TP , S TC and C k are obtained, and are respectively expressed as:

[0136]

[0137] The input matrix F and the transfer matrix V are the Jacobian matrices corresponding to F(S TP , S TC , C k ) and V(S TP , S TC , C k );

[0138] The corresponding maximum eigenvalue is:

[0139]

[0140] Among them, in the case of no congestion threshold

[0141] When R 0 ≤1, there is an equilibrium point of no congestion C k in the FC-ISQ two-layer traffic congestion propagation model.

[0142] Specifically, for S6, the present application reproduces the results of the theoretical derivation of each equilibrium point through simulation. This solution realizes the specific setting of the parameters of the FC-ISQ model, and verifies the reliability and practicability of the model. The prediction results output of the model for each equilibrium point reflect the propagation trend and propagation situation of the congested link C k under different conditions. Comprehensive analysis of the model parameters and the propagation characteristics of each group obtained by prediction can help to deeply analyze the influencing factors of traffic congestion and the dynamic law of congestion propagation. The specific numerical simulation parameters are shown in the following table:

[0143] Table 1 Numerical simulation parameters

[0144]

[0145]

[0146] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by those skilled in the art in this technical field based on the concept of the present invention through logical analysis, reasoning, or limited experiments on the basis of the prior art should fall within the protection scope determined by the claims.

Claims

1. A method for predicting the propagation of road congestion based on a two-layer hypergraph model, characterized in that: include: S1: Construct a hypergraph-based FC-ISQ two-layer traffic congestion propagation model; S2: The network layers of the FC-ISQ two-layer traffic congestion propagation model are determined as the traffic road layer and the traffic information propagation layer; S3: Establish a heterogeneous differential equation system for the FC-ISQ two-layer traffic congestion propagation model and verify the positivity and boundedness of the model solution; S4: Calculate the basic reproduction number R0 of the FC-ISQ two-layer traffic congestion propagation model and use it as the threshold to determine whether the traffic congestion will automatically dissipate; S5: Numerical simulation verification of the FC-ISQ two-layer traffic congestion propagation model; S6: Through simulation, the prediction results of each group under different initial parameter settings of the FC-ISQ two-layer traffic congestion propagation model are obtained, and the visual prediction of the propagation trend and propagation status of traffic congestion under different situations is realized.

2. The method for predicting the propagation of road congestion based on a double-layer hypergraph model according to claim 1, characterized in that: The traffic road layer in S2 includes: Classify the traffic road layer into free links F k and congested link C k ; The concept of hypergraph is introduced at the traffic road layer to refine the connection types of roads. The intersections of roads represent nodes. Any number of nodes are connected by a hyperedge, which represents the interrelationship of multiple links under urban road construction. Only one-way flow links are considered. Based on the number of one-way outflow links at the intersection and the size of the hyperedge, the refined classification of roads is represented by k i =[k i 2 , k i 3 , k i 4 , ..., k i J ],k i J represents the number of intersections with hyperedge size J among i outgoing links; When the traffic volume increases, the free link F k Becomes congested link C with a probability of β3 k ; When the traffic volume decreases, the congested link C k F becomes a free link with a probability of φ4 k .

3. The method for predicting the propagation of road congestion based on a double-layer hypergraph model according to claim 2, characterized in that: The traffic information dissemination layer in S2 includes: Those who do not know the information I, those who spread information that promotes travel S TP , those who spread congestion information S TC Q who keeps silent about known information; When the unknown informant I learns about the traffic conditions of the road, he becomes the information disseminator S who promotes travel with a probability of β1. TP ; When the unknown informant I learns about the traffic conditions of the road, he becomes the congestion information disseminator S with a probability of β2. TC ; When disseminating information that promotes travel TP When the transmission of the traffic situation on the road is stopped, Q becomes the known information and remains silent with a probability of φ1; When the congestion information is transmitted by S TC When the transmission of the traffic situation on the road is stopped, Q becomes the known information and remains silent with a probability of φ2; When the congestion information is transmitted by S TC When the traffic situation of the road is propagated, the congested link C k F becomes a free link with a probability of φ3 k .

4. The method for predicting the propagation of road congestion based on a double-layer hypergraph model according to claim 3, characterized in that: The heterogeneous differential equation system in S3 is as follows: dF k (t)=[φ3C k (t)S TC (t)+kφ4Θ(t)C k (t)-β3F k (t)S TP (t)]dt dC k (t)=[β3F k (t)S TP (t)-φ3C k (t)S TC (t)-kφ4Θ(t)C k (t)]dt dI(t)=[Λ-β1I(t)S TP (t)-β2I(t)S TC (t)-μ I (t)]dt dS TP (t)=[β1I(t)S TP (t)-φ1S TP (t)-μS TP (t)]dt dS TC (t)=[β2I(t)S TC (t)-φ2S TC (t)-μS TC (t)]dt dQ(t)=[φ1S TP (t)+φ2S TC (t)-μQ(t)]dt in, In the formula, F k (t) is the free link F with excess k at time t k The number of k (t) is the congested link C that exceeds k at time t k The number of TC (t) is the person S who spreads the congestion information at time t TC , Θ(t) is the number of hyperedges pointing to free links F k The probability of S TP (t) is the person S who spreads the information promoting travel at time t TP , I(t) is the number of unknown information I at time t, Λ is the migration rate of the traffic information dissemination layer group, μ is the migration rate of the traffic information dissemination layer group, Q(t) is the number of known information Q who remain silent at time t, ω j is the propagation weight of the hyperedge size j, k j is the superdegree corresponding to the starting node of the link, P(k j ) is for salvation k j The probability of the distribution, <k j > is the average hyperdegree corresponding to the hyperedge size j, J is the maximum hyperedge order, R is the maximum number of outgoing links of the starting node, is the free link F with the number of outgoing links i and hyperedge order j of the starting node k at time t k The number of is the total number of links with outgoing links from starting node i and hyperedge order j.

5. The method for predicting the propagation of road congestion based on a double-layer hypergraph model according to claim 4, characterized in that: The positivity and boundedness of the model solution are verified in S3, including: Let N 1k (t) represents the total number of roads with starting node type k within the research scope at time t, N1 is the total number of roads within the research scope, and N2(t) is the total number of traffic information dissemination layer groups at time t, then: N2(t)=I(t)+S TP (t)+S TC (t)+Q(t) Assuming all parameter values ​​are non-negative, we have: All solutions are bounded, and the feasible domain of the model is: Where Γ is the feasible region of the model.

6. The method for predicting the propagation of road congestion based on a double-layer hypergraph model according to claim 5, characterized in that: The S4 includes: The heterogeneous differential equations are set to zero, the equilibrium point of the FC-ISQ two-layer traffic congestion propagation model is calculated, and the basic reproduction number R0 is calculated and used as the threshold to determine whether the traffic congestion will automatically dissipate; The balance points include a non-congestion balance point E0, a boundary balance point E1, a boundary balance point E2, and a congestion-continuous balance point E * .

7. The method for predicting the propagation of road congestion based on a double-layer hypergraph model according to claim 6, characterized in that: The calculation of the basic reproduction number R0 and using it as a threshold for determining whether the traffic congestion is automatically dissipated includes: Using the next generation matrix method, we get TP , S TC and C k The incoming part F and the outgoing part V are respectively expressed as: The input matrix F and the transfer matrix V are F(S TP , S TC , C k ) and V(S TP , S TC , C k )The corresponding Jacobian matrix; The corresponding maximum eigenvalue is: Among them, in the absence of congestion Threshold When R0≤1, the FC-ISQ two-layer traffic congestion propagation model has a non-congestion C k balance point.

Images