Traffic network robustness evaluation method and system based on multi-source data fusion

Abstract

The invention relates to the technical field of intelligent traffic, in particular to a traffic network robustness evaluation method and system based on multi-source data fusion, and the method specifically comprises the following steps: obtaining a plane axis diagram according to an obtained research map and road data, and marking the obtained data information; importing the plane axis diagram into the depthmapX to construct a traffic road network topology model, and performing spatial syntactic analysis; exporting a traffic road network topology model and a space syntactic analysis result, accessing real-time data in the acquired data, and mapping the traffic road network topology model to a road network space coordinate system through a space-time alignment algorithm; constructing a nonlinear coupling model to calculate the robustness weight of each road section, and optimizing parameters in the model through a loss function and a gradient descent method; and combining the robustness weight of each road section after parameter optimization with the disaster probability factor, and outputting a comprehensive evaluation result. The method solves the problems of data isolation and evaluation lag of a traditional method, and is suitable for urban road network optimization and emergency management.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation, and particularly relates to a method and system for evaluating the robustness of a traffic road network based on multi-source data fusion. Background Art

[0002] The stability of the road traffic network is directly related to the normal operation of the entire urban social and economic activities. Studying the robustness of the road traffic network helps improve the anti-interference ability of the urban traffic system, so as to effectively reduce losses when facing natural disasters and traffic congestion and other emergencies.

[0003] Space syntax explores the importance of a certain space to people, uses data-based methods to represent people's subjective images, analyzes the relationship between spaces with quantitative indicators such as integration degree and choice degree, and describes the impact of building space and urban space on people; Sheng Qiang (2018) conducted an empirical study on the cross-section passenger flow between subway stations in Beijing, Tianjin and Chongqing with integration degree and choice degree as analysis factors; Yu Yang (2023) analyzed the spatial layout of Kazanqi folk culture block through road connection value, road control value, road choice degree and pedestrian simulation; in complex networks, an important node is called a key node. In the study of complex networks, scholars have found that these key nodes in the network can determine the structure and function of the entire network.

[0004] The research on the robustness of the road traffic network not only helps to improve the stability and anti-interference ability of the urban traffic system, but also provides a scientific basis for traffic planning and management, thus promoting the sustainable development of urban social and economic activities. Traditional road network design usually focuses on optimizing the efficiency and capacity of the network. However, with the acceleration of the urbanization process, the challenges faced by the road network are increasing day by day. How to enhance its anti-interference ability by improving the robustness of the road network has become the main research goal at present. The complex network theory provides a new perspective for this problem. By abstracting the road network into a topological structure, the relationship between its centrality, vulnerability and robustness is studied.

[0005] Therefore, the present invention proposes a method and system for evaluating the robustness of a traffic road network based on multi-source data fusion to solve the above problems. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the present invention develops a method and system for evaluating the robustness of a traffic road network based on multi-source data fusion. Through the evaluation model based on space syntax, the present invention can more comprehensively understand the performance of the road network when being disturbed, and can more deeply understand the stability and adaptability of the urban road network when facing various disturbances by combining complex network theory and space syntax analysis method.

[0007] On the one hand, the technical solution of the present invention for solving the technical problem is a method for evaluating the robustness of a traffic road network based on multi-source data fusion, including the following steps: Data acquisition: Obtain the research map and road data, import the obtained data into AutoCAD software for processing to obtain a plane axis diagram, and mark the obtained data information in the plane axis diagram; Spatial syntax analysis: Import the plane axis diagram into depthmapX to construct a traffic road network topological model and perform spatial syntax analysis. The spatial syntax analysis includes integration degree analysis, connectivity value analysis, and choice degree analysis; Mapping of the traffic road network topological model: Export the traffic road network topological model and the results of spatial syntax analysis, and access the real-time data in the obtained data. Map the traffic road network topological model to the road network spatial coordinate system through a spatio-temporal alignment algorithm; Model construction and optimization: Construct a non-linear coupling model to calculate the robustness weights of each road section, and optimize the parameters in the model through a loss function and the gradient descent method; Output of evaluation results: Combine the robustness weights of each road section after parameter optimization with the disaster probability factor, and output the comprehensive evaluation results.

[0008] In the specific implementation manner, the data acquisition is specifically as follows: The obtained data includes road axes, intersection nodes, road attribute information, real-time data, and historical accident records. Mark the road axes, intersection nodes, and road attribute information in the plane axis diagram; Among them, the road axes, intersection nodes, and road attribute information are obtained by parsing a publicly available geographic information database. The road attribute information includes design vehicle speed, regional speed limit, road grade, and number of lanes; The real-time data includes real-time vehicle speed, congestion index, real-time heavy rain occurrence probability, and earthquake occurrence probability. The real-time vehicle speed and congestion index are obtained by accessing a web map through an API, and the real-time heavy rain occurrence probability and earthquake occurrence probability are obtained by accessing the meteorological bureau API; The historical accident records are obtained from the government data open platform. The historical accident records include accident location, recovery time, and influence range.

[0009] In the specific implementation manner, the spatial syntax analysis is specifically as follows: Import the plane axis diagram into the DepthmapX spatial syntax analysis software for spatial syntax analysis. The analysis process is as follows: (1)Construct a traffic road network topology model based on the plane axis diagram and the marked data. Convert the actual roads in the plane axis diagram into line segments, with each line segment representing a road section, so that the connection relationships between intersections and road sections are consistent with the traffic road network topology model, and correct the adjacency relationship according to the actual traffic rules. Simplify the topology of complex intersections and only retain the main traffic paths; (2)Integration analysis: Select the integration analysis module in the DepthmapX space syntax analysis software. Calculate the topological depth between each road section based on the shortest path algorithm to generate a global integration value, which represents the aggregation degree of the th line segment in the entire traffic road network topology model and represents the total number of road sections in the traffic road network topology model; (3)Connectivity value analysis: In the DepthmapX space syntax analysis software, count the number of road sections directly connected to each road section in the traffic road network topology model to obtain the connectivity value corresponding to each road section; (4)Choice analysis: Select the choice analysis module in the DepthmapX space syntax analysis software. Count the frequency of each road section passing through all the shortest paths. This frequency is the choice value corresponding to each road section. The road sections with high choice values in the entire traffic road network topology model are the key road sections with high traffic pressure.

[0010] In the specific implementation manner, the mapping of the traffic road network topology model is as follows: Map the traffic road network topology model to the road network space coordinate system through the spatio-temporal alignment algorithm combined with the space syntax analysis results to unify the data in terms of time and space. Define the time window of the road network spatio-temporal coordinate system as , and the data within the time window is regarded as the same moment. For the high-frequency data in the traffic road network topology model, take its last frame. For the low-frequency data in the traffic road network topology model, perform interpolation for alignment, and then perform spatial registration through the implicit neural representation network. Verify the coordinate points of the road network space according to the known calibration points, and train the alignment results of the spatio-temporal alignment algorithm. The calibration points are the data information marked and obtained in the plane axis diagram.

[0011] In the specific implementation manner, the non-linear coupling model is as follows: Calculate the S-shaped function gain term of the connectivity value of each road section, the power-law correction term of the choice and the ratio attenuation term of the integration and the vehicle speed through the non-linear coupling model to calculate the robustness weight of each road section in the traffic road network topology model. The calculation formula is as follows: , where, Denote the robustness weight of the th section in the traffic road network topology model, denote the gain term of the S-shaped function for the connection value, denote the gain steepness coefficient, denote the power-law correction term for the selectivity, denote the selectivity intensity coefficient, denote the selectivity nonlinear exponent, denote the selectivity bias term, denote the ratio attenuation term of the integration degree and the vehicle speed, denote the th section's real-time speed, denote the th section's maximum speed limit, denote the anti-overflow constant.

[0012] In the specific implementation manner, the loss function is specifically as follows: The loss function includes a prediction error term and a regularization term, and the calculation formula is as follows: , where, is the actual robustness score deduced from historical accident data, denote the regularization coefficient used to suppress parameter overfitting, ; The actual robustness score is deduced and calculated from the accident recovery time retrieved by the traffic management department, and the calculation formula is as follows: , where, denote the accident recovery time of the th section retrieved by the traffic management department, denote the th section's preset maximum tolerance time.

[0013] In the specific implementation manner, the gradient descent method is specifically as follows: Set the initial parameters of to be , , denote 's set, and adjust the parameters in iteratively until convergence, and the calculation formula is as follows: , , where, denote the absolute difference of the loss values between two adjacent iterations, denote the updated parameters, represents the current value that has not been updated, and its initial value is , represents the partial derivative of the loss function with respect to the parameter, represents the learning rate, and the learning rate has a dynamic decay rule of , represents the initial value of the learning rate, represents the number of iterations; If the loss function does not decrease after exceeding the preset number of iterations, the training is terminated and the optimal parameters are saved.

[0014] In the specific implementation manner, the output of the evaluation result is specifically as follows: Calculate the disaster probability factor for each road section based on the obtained data , and the calculation formula is as follows: , where, represents the heavy rain occurrence probability of the th road section, represents the earthquake occurrence probability of the th road section, represents the road density of the th road section, represents the number of lanes of the th road section, represents the superposition effect of compound disasters, represents the degree of dilution of disasters by the road network redundancy; Combine the robustness weight of each road section with the disaster probability factor to obtain the comprehensive evaluation result, and the calculation formula is as follows: , where, represents the disaster weight coefficient, represents the congestion index.

[0015] On the other hand, the present invention also provides a traffic road network robustness evaluation system based on multi-source data fusion, including modules for executing the processing instructions of each step in the traffic road network robustness evaluation method based on multi-source data fusion.

[0016] The effects provided in the invention content are only the effects of the embodiments, rather than all the effects of the invention. The above technical solutions have the following advantages or beneficial effects: The present invention discloses a method and system for evaluating the robustness of a traffic road network based on multi-source data fusion. The evaluation model based on space syntax not only considers the topological structure of the road network but also combines multiple morphological variables in space syntax, such as integration and choice, which help to more comprehensively understand the performance of the road network when disturbed; the evaluation model of the robustness of the traffic road network based on space syntax can more deeply understand the stability and adaptability of the urban road network in the face of various disturbances by combining complex network theory and space syntax analysis methods; through multi-source data fusion and non-linear coupling models, the present invention comprehensively and accurately depicts the state of the road network, and combines the gradient descent method to quickly optimize parameters, significantly improving the model training efficiency; the design of the loss function ensures the convergence and generalization ability of the model, while the dynamic adjustment of the disaster factors realizes the real-time identification and emergency response of high-risk areas.

[0017] Therefore, the present invention can not only efficiently evaluate the robustness of the road network, optimize the existing road network design, but also provide reliable decision-making support for traffic management and disaster emergency, provide a scientific basis for the planning of future urban traffic systems, and has certain engineering practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification, and are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention.

[0019] Figure 1 It is a schematic flow chart of the method of the present invention.

[0020] Figure 2 It is the main traffic path retained after topological simplification.

[0021] Figure 3 It is a schematic diagram of the experimental results of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to clearly illustrate the technical features of the present solution, the present invention will be described in detail below through specific embodiments and in conjunction with its drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below.

[0023] Embodiment 1 A method for evaluating the robustness of a traffic road network based on multi-source data fusion in the present invention is used to evaluate the robustness of the traffic road network within a one-kilometer range around Xuejiapozi Subway Station in the West Coast New Area of Qingdao, specifically as follows: Step 1: Data acquisition: Obtain the research map and road data within one kilometer around Xuejiabozi Subway Station in the West Coast New Area of Qingdao. The obtained data includes road plane alignment, intersection nodes, road attribute information, real-time data, and historical accident records; The road plane alignment, intersection nodes, and road attribute information are obtained by parsing the publicly available geographic information database. The road attribute information includes design speed, regional speed limit, road grade, and number of lanes; The real-time data includes real-time vehicle speed, congestion index, probability of rainstorm occurrence, and probability of earthquake occurrence. The real-time vehicle speed and congestion index are obtained by accessing the online map through API, and the real-time probability of rainstorm occurrence and probability of earthquake occurrence are obtained by accessing the meteorological bureau API; The historical accident records are obtained from the government data open platform. The historical accident records include accident location, recovery time, and impact scope; Import the obtained data into AutoCAD software for processing to obtain a plane axis diagram, and mark the obtained data information in the plane axis diagram; Step 2: Space syntax analysis: (1) Import the obtained road plane alignment data into AutoCAD software for processing to obtain a plane axis diagram, and mark the intersection nodes and road attribute information in the plane axis diagram; construct a traffic road network topological model based on the plane axis diagram and the marked data. Convert the actual roads in the plane axis diagram into line segments, and each line segment represents a road section, so that the connection relationship between intersections and road sections is consistent with the traffic road network topological model, and correct the adjacency relationship according to the actual traffic rules. Simplify the topology of complex intersections, and only retain the main traffic paths. The retained main traffic paths are as Figure 2 shown; (2) Integration analysis: Select the integration analysis module in the DepthmapX space syntax analysis software, calculate the topological depth between each road section based on the shortest path algorithm, and generate a global integration value, which represents the aggregation degree of the th line segment in the entire traffic road network topological model, and represents the total number of road sections in the traffic road network topological model; (3) Connectivity value analysis: Count the number of road sections directly connected to each road section in the traffic road network topological model in the DepthmapX space syntax analysis software to obtain the connectivity value corresponding to each road section; (4) Choice analysis: Select the choice analysis module in the DepthmapX space syntax analysis software, and count the frequency of each road section passing through all the shortest paths. This frequency is the choice value corresponding to each road section. The road sections with high choice values in the entire traffic road network topological model are the key road sections with high traffic pressure.

[0024] Step 3: Mapping of the traffic road network topological model: Export the traffic road network topology model generated by space syntax analysis and the analysis results of integration, connectivity value, and choice degree; Access the real-time data source and define the time window of the road network spatio-temporal coordinate system = 5 minutes, and the data within the time window is regarded as the same moment; for high-frequency data, extract the last frame data of the time window, and for low-frequency data, use the linear interpolation algorithm for time alignment; Perform spatial registration on the traffic road network topology model and the road network spatial coordinate system through an implicit neural representation network. The implicit neural representation network consists of a multi-layer perceptron. The input is the node coordinates of the traffic road network topology model, and the spatial high-frequency features are captured through the sine activation function and position encoding, and the registered coordinates are output; the network parameters are supervised and trained through the true coordinates of the calibration points, and the loss function is the mean square error; the structure of the implicit neural representation network includes a position encoding module and a multi-layer perceptron, which are used to map the topology model nodes to the road network spatial coordinate system. The implicit neural representation network includes the following sequentially connected modules; Input module: Receive the node coordinates (x, y) and timestamp t of the traffic road network topology model, and normalize them to the interval [−1, 1]; Position encoding module: Perform Fourier feature mapping on the input coordinates to generate high-frequency feature vectors where p is the input coordinate, L is the frequency series, and the specific encoding parameters are: L = 10, covering the spatial scale from 10 meters to 0.1 meters; Multi-layer perceptron: It contains 4 hidden layers, with 256 neurons in each layer, and the activation function is the sine function (Sin), which is used to fit the non-linear relationship of coordinate mapping; Output module: The linear layer outputs the registered coordinates, and the loss function is the mean square error (MSE) of the calibration point coordinates.

[0025] The training process of the network is synchronized with the gradient descent method, and the dynamic learning rate Decays according to ; Based on the pre-marked calibration points (including intersection nodes and accident locations) in the plane axis diagram, verify the coordinate registration accuracy to ensure that the error is lower than the preset threshold; Use the calibration points as training samples to optimize the parameters of the spatio-temporal alignment algorithm; by iteratively adjusting the weights of the implicit neural representation network, minimize the spatial position deviation of the calibration points after mapping until the convergence condition is met; Output the traffic road network topology model after spatio-temporal alignment to make it completely unified with the real-time data in the spatio-temporal coordinate system.

[0026] Step 4. Model construction and optimization: Calculate the connectivity value of each road section through a non-linear coupling model S-shaped function gain term and selectivity Power-law correction term and integration and vehicle speed Ratio decay term to calculate the robustness weight of each road segment in the traffic network topology model. The calculation formula is as follows: , where, represents the robustness weight of the th road segment in the traffic network topology model, represents the S-shaped function gain term of the connection value, represents the gain steepness coefficient, represents the power-law correction term of the selectivity, represents the selectivity intensity coefficient, represents the selectivity nonlinear exponent, represents the selectivity bias term, represents the ratio decay term of the integration and vehicle speed, represents the th road segment's real-time speed, represents the th road segment's maximum speed limit, represents the anti-overflow constant.

[0027] Optimize the parameters in the model through the loss function and gradient descent method: Retrieve the accident recovery time from the traffic management department , and calculate the actual robustness score: , where, represents the accident recovery time of the th road segment retrieved by the traffic management department, represents the th road segment's preset maximum tolerance time.

[0028] The loss function formula is as follows: , where, is the actual robustness score of the th road segment inferred from historical accident data, represents the regularization coefficient used to suppress parameter overfitting, ; The gradient descent method is as follows: Set 's initial parameter to = 0.5, , represents Set, circular adjustment The parameters in it are adjusted until convergence, and the calculation formula is as follows: , , where, represents the absolute difference of the loss values between two adjacent iterations 。 represents the updated parameter, represents the current value that has not been updated, The initial value of is , represents the partial derivative of the loss function with respect to the parameter, represents the learning rate, and the learning rate The dynamic decay rule of is , represents the number of iterations, and the initial value of the learning rate is set = 0.1.

[0029] Step 5. Evaluation result output: Calculate the disaster probability factor for each section according to the obtained data , and the calculation formula is as follows: , where, represents the heavy rain occurrence probability of the th section, represents the earthquake probability of the th section, represents the road density of the th section, represents the number of lanes of the th section, represents the superposition effect of compound disasters, represents the dilution degree of the road network redundancy on disasters; Combine the robustness weight of each section with the disaster probability factor to obtain the comprehensive evaluation result, and the calculation formula is as follows: , where, represents the disaster weight coefficient, , represents the congestion index.

[0030] Finally, connect the data to each section in Arc GIS and display the high, medium, and low risk nodes in grades. The road network robustness evaluation result is as Figure 3 shown.

[0031] Although the specific embodiments of the invention have been described above in conjunction with the accompanying drawings, they are not intended to limit the scope of protection of the present invention. Based on the technical solutions of the present invention, various modifications or deformations that can be made by those skilled in the art without creative efforts are still within the scope of protection of the present invention.

Claims

1. A method for evaluating the robustness of a traffic network based on multi-source data fusion, characterized in that: The following steps are involved: Data acquisition: obtain research maps and road data, import the acquired data into AutoCAD software for processing to obtain a plane axis diagram, and mark the acquired data information in the plane axis diagram; Space syntax analysis: Import the plane axis map into depthmapX to build a traffic network topology model and perform space syntax analysis, which includes integration analysis, connection value analysis and selectivity analysis; Traffic network topology model mapping: export the traffic network topology model and space syntax analysis results, access the real-time data in the acquired data, and map the traffic network topology model to the road network space coordinate system through the space-time alignment algorithm; Model construction and optimization: Construct a nonlinear coupling model to calculate the robustness weight of each road section, and optimize the parameters in the model through loss function and gradient descent method; Evaluation result output: Combine the robustness weights of each road section after parameter optimization with the disaster probability factor to output a comprehensive evaluation result.

2. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 1 is characterized in that: The data acquisition is as follows: The acquired data includes road axes, intersection nodes, road attribute information, real-time data and historical accident records, and the road axes, intersection nodes and road attribute information are marked in the plane axis diagram; Among them, the road axis, intersection nodes and road attribute information are obtained by parsing the public geographic information database. The road attribute information includes the design speed, regional maximum speed limit, road grade and number of lanes; Real-time data includes real-time vehicle speed, congestion index, real-time probability of heavy rain and earthquake. Real-time vehicle speed and congestion index are obtained by accessing the network map through API, and real-time probability of heavy rain and earthquake is obtained through the API of the Meteorological Bureau. Historical accident records are obtained from the government data open platform, including the accident location, recovery time and impact scope.

3. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 2 is characterized in that: The syntactic analysis is as follows: Import the plane axis diagram into DepthmapX space syntax analysis software for space syntax analysis. The analysis process is as follows: (1) Construct a traffic network topology model based on the plane axis map and the marked data. Convert the actual roads in the plane axis map into line segments. Each line segment represents a road segment. Make the connection relationship between intersections and road segments consistent with the traffic network topology model. Modify the adjacency relationship according to the actual traffic rules. Simplify the topology of complex intersections and retain only the main traffic paths. (2) Integration analysis: Select the integration analysis module in the DepthmapX spatial syntax analysis software, calculate the topological depth between each road segment based on the shortest path algorithm, and generate a global integration value, which represents the topological depth of the first road segment in the traffic network topology model. The degree of aggregation of line segments in the entire traffic network topology model represents the total number of road segments in the traffic network topology model; (3) Connection value analysis: The number of sections directly connected to each section in the traffic network topology model is counted in the DepthmapX spatial syntax analysis software to obtain the connection value corresponding to each section; (4) Selectivity analysis: Select the selectivity analysis module in the DepthmapX spatial syntax analysis software to count the frequency of passing through each road section in all shortest paths. This frequency is the selectivity value corresponding to each road section. The road sections with high selectivity values ​​in the entire traffic network topology model are the key sections with high traffic pressure.

4. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 3 is characterized in that: The traffic network topology model mapping is as follows: The traffic network topology model is mapped to the road network spatial coordinate system through the spatiotemporal alignment algorithm combined with the results of space syntax analysis to unify the data in time and space. The time window of the road network spatiotemporal coordinate system is defined as The data in the time window are regarded as the same moment. For the high-frequency data in the traffic network topology model, the last frame is taken. For the low-frequency data in the traffic network topology model, the alignment is interpolated, and then the implicit neural representation network is used for spatial registration. The coordinate points of the road network space are verified according to the known calibration points. The alignment results of the spatiotemporal alignment algorithm are trained. The calibration points are the data information marked in the plane axis diagram.

5. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 4 is characterized in that: The nonlinear coupling model is as follows: Calculate the connection value of each road segment through nonlinear coupling model The S-type function gain term and selectivity The power law correction term and integration degree of With speed The robustness weight of each road section in the traffic network topology model is calculated by using the ratio attenuation term. The calculation formula is as follows: ， in, Represents the topological model of the traffic network. The robustness weight of each road segment, The S-shaped function gain term representing the connection value, represents the gain steepness coefficient, represents the power-law correction term of the selectivity, represents the selectivity intensity coefficient, represents the nonlinear index of selectivity, represents the selectivity bias term, represents the ratio attenuation term of integration degree and vehicle speed, Indicates Real-time speed of each road segment, Indicates The maximum speed limit for each road section, Represents an overflow-proof constant.

6. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 5 is characterized in that: The loss function is as follows: The loss function includes the prediction error term and the regularization term, and the calculation formula is as follows: ， in, is the actual robustness score based on the back-calculation of historical accident data, represents the regularization coefficient used to suppress parameter overfitting, ; The actual robustness score is the accident recovery time retrieved by the traffic management department The calculation formula is as follows: ， in, Indicates that the traffic management department has requested The accident recovery time for each road section, Expressing the The maximum tolerance time preset for each road section.

7. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 6 is characterized in that: The gradient descent method is as follows: set up The initial parameters are , , express A collection of loop adjustments The parameters in are calculated until convergence, and the calculation formula is as follows: ， ， in, Represents the absolute difference of the loss value between two adjacent iterations, represents the updated parameters, Indicates the current value that has not been updated. The initial value is , represents the partial derivative of the loss function with respect to the parameters, Represents the learning rate, learning rate The dynamic attenuation law is , represents the initial value of the learning rate, Indicates the number of iterations; If the loss function does not decrease after the preset number of iterations, the training is terminated and the optimal parameters are saved.

8. The method for evaluating the robustness of a traffic network based on multi-source data fusion according to claim 7 is characterized in that: The output of the evaluation results is as follows: Calculate the disaster probability factor for each road section based on the acquired data , the calculation formula is as follows: ， in, Indicates The probability of heavy rain on each road section, Indicates The probability of an earthquake occurring on a road section is Indicates The road density of each road segment, Indicates The number of lanes in a road section, represents the superposition effect of compound disasters, Indicates the dilution effect of road network redundancy on disasters; The robustness weight of each road section is combined with the disaster probability factor to obtain a comprehensive evaluation result. The calculation formula is as follows: ， in, represents the disaster weight coefficient, Represents the congestion index.

9. A traffic network robustness assessment system based on multi-source data fusion, characterized by: Including for executing claims A module for processing instructions for each step in a method for evaluating robustness of a traffic network based on multi-source data fusion as described in any one of the claims 1 to 5.

Images