Urban road network traffic flood vulnerability assessment method based on multi-source and multi-modal data

Abstract

The invention discloses an urban road network traffic flood vulnerability assessment method based on multi-source and multi-modal data, which constructs a road network disaster vulnerability assessment model through a multi-source data fusion technology in combination with multi-source data such as hydrology, water conservancy, land utilization, DEM, road network attributes, traffic flow, social economy, social media and mobile phone signaling. A weighted fusion method is adopted to efficiently integrate multi-modal data, a high-performance hydrodynamic integration model HiPIMS is utilized to realize high-resolution water depth grid dynamic simulation, changes of road traffic flow, rate and traffic capacity under a flood situation are accurately captured, and the reliability of a result is verified through verification indexes. On the basis of road grade service characteristics and flood disaster loss rates, vulnerability curves of different grades of roads are deduced, and rate verification is carried out by using mobile phone signaling data based on a mobile base station, so that quantitative analysis and grade division of road network disaster vulnerability are innovatively realized, and the accuracy and timeliness of road network risk quantification are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban road network traffic flood disaster risk assessment, and in particular to an urban road network traffic flood disaster vulnerability assessment method based on multi-source multi-modal big data technology. Background Art

[0002] At present, urban flood disasters occur frequently, which has a serious impact on the traffic capacity and transportation function of urban road networks. As one of the main disaster-bearing bodies of flood disasters, urban roads often face problems such as infrastructure damage, traffic congestion and interruption in flood events. Traffic accidents, economic losses and casualties caused by extreme rainfall are particularly serious. However, existing flood risk assessment methods have significant limitations, mainly manifested in the low assessment accuracy and timeliness due to lack of data or insufficient integration capabilities.

[0003] Traditional research relies heavily on meteorological and hydrological data, ignoring comprehensive factors such as surface impervious area and drainage system capacity, resulting in limited accuracy of results and response efficiency; existing hydrological and hydrodynamic models have low computational efficiency and are difficult to simulate complex dynamic flood processes. In road network vulnerability assessment, macro studies usually adopt a binary judgment method of "unobstructed-interrupted", ignoring the complex topological structure of the road network and the changes in the dynamic capacity of roads; micro studies use fixed water depth thresholds to judge the road traffic status, but fail to reflect the actual traffic characteristics. For example, some flooded roads still have a certain capacity, while unflooded slippery roads may also have reduced capacity due to a decrease in the friction coefficient. In addition, existing studies generally lack effective flood verification data, making it difficult to accurately assess the true impact of complex dynamic floods on road networks. Summary of the invention

[0004] In view of the above technical problems in the related art, the present invention provides a method for assessing the vulnerability of urban road network traffic floods based on multi-source and multi-modal data, which can solve the above problems.

[0005] To achieve the above technical objectives, the technical solution of the present invention is implemented as follows:

[0006] A method for assessing the vulnerability of urban road network traffic flood disasters based on multi-source and multi-modal data comprises the following steps:

[0007] S100, collecting and preprocessing multi-source and multi-modal traffic disaster data;

[0008] S200, constructing a vulnerability index system for the pre-processed disaster data, including index extraction and index quantification, wherein the extracted indicators include flood characteristic indicators, vulnerability indicators and traffic flow data. The index quantification is to use the convolutional neural network method to extract flood depth and range characteristics from different types of data, and to fuse the features extracted from multi-source and multi-modal data through a weighted fusion method;

[0009] S300, constructing a road network disaster vulnerability model, analyzing the road network disaster vulnerability through the road network disaster vulnerability model, and classifying the road network flood vulnerability;

[0010] S310, classify roads into different levels according to their socio-economic functional attributes, road network connectivity and spatial distribution geometric characteristics;

[0011] S320, using the curve fitting toolbox in Matlab, a quadratic function is used to fit the water depth-safe speed curve under existing similar urban conditions;

[0012] S330, based on the collected original maximum safe speed limit, water depth data, road capacity and traffic flow data of different road grades, using a curve regression fitting method to derive the specific relationship between the safe driving speed and water depth under different road grades;

[0013] S340, using the high-performance hydrodynamic integrated model HiPIMS to simulate and obtain the real-time water depth distribution of road network traffic;

[0014] S350, obtaining / estimating the road section rate based on mobile phone signaling data;

[0015] S360, input road grade, flood depth, and traffic flow data into the GIS platform, and use spatial analysis tools to classify the flood vulnerability of roads based on the maximum safe speed of the road and the reduction in the response to floods. The road vulnerability is divided into five levels: low, secondary, medium, high, and very high, and a flood vulnerability level map is drawn for each road section.

[0016] Furthermore, the multi-source and multi-modal traffic disaster data in step S100 includes social media data, traffic data, socioeconomic data, land use types, DEM, and also includes flood-related road network disaster data acquired in real time through the Internet of Things and remote sensing.

[0017] Furthermore, the preprocessing of the collected multimodal traffic disaster data includes: image cleaning and labeling of image data; word segmentation and syntactic analysis of text data; and reclassification and grading of traffic data.

[0018] Furthermore, the flood characteristic indicators in step S200 are obtained through HiPIMS model simulation, including real-time water depth data, and the damage is evaluated by calculating the reduction in vehicle speed through the flood inundation map; the traffic flow data is based on the personnel movement trajectory information in the mobile phone signaling data of the mobile base station.

[0019] Furthermore, the fusion of features extracted from multi-source multi-modal data in step S200 specifically includes: first, evaluating the reliability according to the historical performance, authority, coverage and update frequency of the data source, and evaluating the accuracy of the data through error analysis and time extension, and assigning a corresponding weight to each data feature based on the evaluation results; then, performing a weighted average calculation on the feature values ​​from different data sources in the same indicator to generate the final fused feature value.

[0020] Furthermore, in step S320, the quadratic function is v0=f(h)=ah 2 +bh+c, where v0 represents the maximum safe speed that can be accommodated when the flooding depth is h, and a, b, and c are function parameters related to the initial maximum speed of the road; the specific relationship in step S330 is given by the formula v0=f(v i , h) indicates that v i represents the original maximum safe speed limit of roads of different levels, and h is the flood depth; the calculation formula for the flood vulnerability of the road in step S360 is: Among them C r represents the vulnerability of the road network, v max It indicates the vehicle capacity under different road grades. It is an adjustment coefficient used to consider the impact of different road grades and service levels on vulnerability.

[0021] Furthermore, step S300 also includes: S370, optimizing the model parameters using a gradient ascent method.

[0022] Furthermore, it also includes: a verification method for the road network disaster vulnerability model based on keyword indexing of social media disaster big data, through flood disaster and disaster degree information mining and keyword quantification, combining precision, recall rate and F-score to verify the accuracy of the model, and a real-time verification and acquisition method of road traffic rate based on mobile phone signaling data, using mobile phone signaling data and road weights to estimate real-time traffic rate, analyze traffic flow conditions, and calibrate the model through actual flood and traffic rate data to improve the accuracy and reliability of path algorithms and parameters.

[0023] Beneficial effects of the present invention: This application uses multi-source data fusion technology to combine hydrology, water conservancy, land use, DEM, road network attributes, traffic flow, social economy, social media and mobile phone signaling and other multi-source data to build a high-precision, near-real-time road network disaster vulnerability assessment model. The weighted fusion method is coupled with the HiPIMS model to achieve high-resolution water depth grid dynamic simulation, accurately capture the changes in road section traffic flow, rate and capacity under flood scenarios, and verify the reliability of the results through verification indicators (precision, recall rate and F-score). Based on the road grade service characteristics and flood loss rate, the vulnerability curves of roads of different grades are derived, and the rate verification is carried out using mobile phone signaling data based on mobile base stations, which innovatively realizes accurate assessment of road sections. The invention significantly improves the accuracy and timeliness of flood risk assessment, provides scientific support for flood risk assessment, traffic emergency response and disaster prevention and relief, and effectively reduces the impact of floods on the transportation system and social economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] The present invention is further described in detail below with reference to the accompanying drawings.

[0026] Figure 1 It is a flow chart of a method for assessing urban road network traffic flood vulnerability based on multi-source multi-modal data according to an embodiment of the present invention;

[0027] Figure 2 It is the information composition of social media data in a disaster situation described in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0029] like Figure 1-2 As shown, according to the present invention, a method for assessing the vulnerability of urban road network traffic flood disasters based on multi-source multi-modal data is disclosed, comprising the following steps:

[0030] Collect multi-source and multi-modal road network traffic data, which includes social media data, traffic data, socio-economic data, land use type, DEM, road network attributes and road construction information data. It also includes flood-related data obtained in real time through the Internet of Things and remote sensing. After vectorization, interpolation and normalization, it provides standardized data support for model calculation.

[0031] A vulnerability index system is constructed. The index system is constructed from aspects such as road network capacity, flood inundation characteristics, and dynamic response behavior. It mainly includes index extraction and index quantification. The extracted indicators include flood characteristic indicators, vulnerability indicators, and traffic flow data. The index quantification covers two aspects: feature recognition and fusion analysis.

[0032] A road network disaster vulnerability model is constructed to analyze the impact of flood disasters on the functions of urban road networks, study the damage of flood characteristics (water depth) to road network service performance (traffic speed, capacity), extract dynamic features through the high-performance hydrodynamic integrated model HiPIMS, construct the safety speed-water depth relationship curve using curve regression, analyze vulnerability indicators based on the GIS platform, iteratively optimize model parameters through the gradient ascent method to improve evaluation accuracy, and finally complete the road network disaster vulnerability analysis and grading.

[0033] The road network disaster vulnerability model is verified. The model verification is a method of verifying the road network disaster vulnerability model based on keyword indexing of social media disaster big data. The model accuracy is verified by mining flood disaster and disaster severity information and quantifying keywords, combining precision, recall rate and F-score. At the same time, the real-time verification and acquisition method of road traffic rate based on mobile phone signaling data uses mobile phone signaling data and road weights to estimate the real-time traffic rate, analyze traffic flow conditions, and calibrate the model through actual flood and traffic rate data to improve the accuracy and reliability of path algorithms and parameters.

[0034] Embodiment 1:

[0035] The social media data in this application comes from platforms such as Weibo and Twitter, and contains images, text, geographic location information and timestamps, which are obtained through crawlers or APIs and stored in structured or unstructured data forms. Image processing technology is combined with annotation tools (Labelme) to annotate the target area (roads, waterlogging) and extract feature information; in terms of text processing, after key semantic extraction, character cleaning and stop word filtering, sentiment analysis is carried out to quantify the disaster situation, and the time and location formats are unified to achieve data alignment and efficient analysis.

[0036] Furthermore, the present application obtains vectorized urban road network information, including roads, road sections and intersections, through an electronic map (OpenStreetMap). At the same time, relevant data such as DEM, rainfall, land use type, Manning coefficient and other hydrological parameters and urban drainage capacity are collected. A high-performance hydrodynamic integrated model (HiPIMS) is used to simulate typical rainfall events, and after initializing the parameters, it is run on a GPU accelerator to obtain the real-time maximum flooding depth. The road network information is segmented and processed through GIS software, and the road section distribution information is extracted; the DEM data is optimized according to the resolution, the rainfall data is interpolated through the Thiessen polygon method, the Manning coefficient is calibrated according to the land use, and the drainage capacity is based on the generalized drainage node rate to meet the model input requirements.

[0037] Embodiment 2:

[0038] The indicators extracted in this application include flood characteristic indicators (water depth), vulnerability indicators (vehicle speed reduction) and traffic flow data. The flood characteristic indicators are obtained through HiPIMS model simulation and include real-time water depth data. The vulnerability indicators are the vehicle speed reduction calculated through the flood inundation map; the traffic flow data is the personnel movement trajectory information in the mobile phone signaling data based on the mobile base station.

[0039] In this application, the quantification of indicators covers two aspects: feature recognition and fusion analysis. Convolutional neural networks (CNNs) are used to extract flood depth and range features from image data, including: cleaning and segmenting social media images, extracting waterlogging areas and estimating water depths. These image features provide key support for the quantification of flood disaster range and water depth. Fusion analysis achieves the integration and quantification of flood disaster information by weighting the features of multi-source and multi-modal data. The specific process includes: first, assigning weights to each data feature based on the reliability and accuracy of the data source. Remote sensing images have higher weights than social media images or text descriptions due to their higher precision. Then, the feature values ​​of the same indicator (water depth) from different data sources are weighted averaged according to the weights to generate the final fused feature value. In order to ensure the scientific nature of the fusion, the setting of weights can be completed by verifying the accuracy of historical data or assigning expert experience. Weighted fusion effectively integrates the advantages of data from different sources, reduces the deviations that may be caused by a single data source, and provides high-quality input data for the comprehensive assessment and model optimization of subsequent flood disasters.

[0040] Embodiment three:

[0041] The core of the road network disaster vulnerability model construction in this application is to derive the relationship between safe speed and water depth under different road grades through regression analysis, and further calculate the vulnerability of the road network. The following are the specific implementation steps of the model construction method:

[0042] Through the curve fitting toolbox in Matlab, the following quadratic function is used to fit the existing water depth-safe speed curve. The specific formula is as follows:

[0043] v0=f(h)=ah 2 +bh+c;

[0044] Where v0 represents the maximum safe speed that can be accommodated under the flood depth h, a, b, c are function parameters related to the initial maximum speed of the road. Further considering the original maximum safe speed limit v of different road grades i , so v0 is about v i The binary function of and h is as follows:

[0045] v0=f(v i , h)=f a (v i , h)+f b (v i , h)+f c (v i , h);

[0046]

[0047] After collecting the original maximum safe speed limit v of different road levels i Based on the data of main roads, secondary roads, branch roads, water depth, road capacity and traffic flow, the relationship between safe driving speed and water depth under different road grades is derived by using curve regression fitting method. The specific relationship is given by the following formula:

[0048]

[0049] The road network vulnerability C is calculated based on the safe driving speed under flooding conditions and the original maximum safe speed limit of different road grades. r , the specific formula is as follows:

[0050]

[0051] in is the adjustment factor used to consider the impact of different road grades and service levels on vulnerability, v max Indicates the vehicle's traffic capacity at different road levels and the maximum safe speed v specified by different road levels i The value is directly related to the actual road network function level. The safe speed v0 is related to the actual flooding depth, which can be obtained by comparing the water depth and the corresponding safe speed relationship curve;

[0052] According to the functional attributes and importance of roads, roads are divided into different levels such as main roads, secondary roads, and branch roads. When calculating vulnerability, the differences in service performance of different road levels are taken into account. Main roads usually carry higher traffic flows, and their flood vulnerability will be more affected, while branch roads or secondary roads are less affected. Combined with water depth data, the proportion of traffic capacity reduction of each road section is calculated, and the vulnerability is determined by comparing the difference between the safe driving speed under flood conditions and the original maximum safe speed limit;

[0053] Input data such as road grade, flood depth, and traffic flow into the GIS platform for spatial analysis to identify which roads are most severely affected and which areas may cause traffic interruption. The flood vulnerability of roads is graded according to the maximum safe speed of the road and the reduction in the response to floods. Road vulnerability is divided into five levels: low, secondary, medium, high, and very high. A flood vulnerability level map of each road section is drawn to clarify the vulnerability level of each road section, provide a decision-making basis for urban traffic management and emergency response, and preliminarily establish a road network disaster vulnerability model.

[0054] Furthermore, the gradient ascent method is used to optimize the model parameters. First, the maximum eigenvalue of the pairwise comparison matrix that passes the consistency test and its corresponding eigenvector are calculated by the hierarchical analysis method as the initial value of the model optimization. Subsequently, the gradient ascent method is used for iterative optimization to continuously adjust the weight combination to gradually approach the maximum value of the model accuracy index (area under the ROC curve AUC). The main steps of the gradient ascent algorithm are as follows:

[0055] Initial weight setting: Initial weights are set based on historical data and expert experience;

[0056] Difference calculation: Calculate the difference (error) between the model prediction result and the true value under the current weight;

[0057] Gradient calculation and update: Calculate the gradient of the model error relative to each parameter, and update the weight parameters by the gradient ascent method. The specific formula is as follows:

[0058]

[0059] Among them, θ is the model parameter, α is the learning rate, and J(θ) is the loss function;

[0060] Iterative optimization: Continuously iterate and optimize, and gradually make the model accuracy approach the maximum value by adjusting parameters.

[0061] In summary, when using the road network disaster vulnerability model to assess the vulnerability of the road network, the reduction in road safety speed is first taken as the core assessment indicator to establish a corresponding assessment model. Subsequently, the natural break point method based on ArcGIS is used for grading. This method ensures that the differences between categories are significant and the differences within categories are minimal by calculating the variance and judging the classification effect. In the process of model application, it is necessary to extract the flood depth based on the flood simulation model and quantify the reduction in road safety speed under different water depth conditions, and then grade the road network based on this, so as to achieve accurate road network disaster vulnerability assessment.

[0062] Embodiment 4:

[0063] The verification principle of the road network vulnerability assessment model in this application is to use big data mining technology to obtain actual disaster information (the relationship between speed and water depth), and to evaluate the accuracy of the model by comparing the road network vulnerability level predicted by the model with the actual disaster level, using indicators such as precision, recall and F-score.

[0064] Precision measures the accuracy of model predictions, indicating the proportion of true positive examples in the classification results to predicted positive examples. If the model evaluates u subsamples out of v samples, and w of them are correct results, the precision formula is:

[0065] Precision = u / w;

[0066] In road network vulnerability assessment, a positive case refers to a situation where the vulnerability level predicted by the model is consistent with the actual disaster level. Recall measures the model's ability to identify positive cases, indicating the proportion of correctly identified positive cases to all actual positive cases. The specific formula is as follows:

[0067] recall=u / v;

[0068] F-score is the harmonic mean of precision and recall. It takes both precision and recall into consideration and can evaluate the performance of the model more comprehensively. The precision, recall and F-score values ​​of the model under different rainfall scenarios are calculated as follows:

[0069] F-score=2 Precision*recall=(Precision+recall).

[0070] Furthermore, road traffic flow (road section speed) is an important indicator of urban traffic conditions. Mobile phone signaling data is used to collect user information and signal propagation data through the mobile communication system to estimate the average speed of the road section. The model is verified and calibrated using real flood disaster data and traffic flow data to ensure the accuracy and reliability of the model. To obtain real-time monitoring of road traffic speed, the following methods can be used:

[0071] Point matching: Based on map data or GIS information, a database containing the locations and attributes of all road sections is constructed. For offset points, the nearest neighbor algorithm is used to calculate the Euclidean distance between the offset point and each road section, and the road section with the shortest distance is selected as the matching result. The offset point can be accurately mapped to the corresponding road section, achieving efficient correction and accurate matching of the point.

[0072] Trajectory data generation: After point matching, the points have been accurately mapped to the corresponding road sections. Combined with the timestamp and longitude and latitude data of the points, the dynamic drawing and path estimation of the personnel movement routes are carried out along the electronic map navigation path (AutoNavi API), thereby generating complete user trajectory information and intuitively showing their dynamic behavior patterns.

[0073] Rate data generation: The trajectory data is processed and analyzed to extract the average rate of the road section. First, based on the trajectory data of each pedestrian, the total distance moved on the road section is calculated, and the total time spent on the road section is determined by the timestamp, so as to calculate the average rate of each pedestrian. The formula is: average rate = total distance / total time. When multiple pedestrians move on the same road section, the average rate of all pedestrians over a period of time can be summarized. Furthermore, in order to more accurately reflect the overall rate distribution of the road section, the contribution of each pedestrian's travel time on the road section to the average rate is considered, and the weighted average method is used to calculate the average rate of the road section. The formula is: average rate of the road section = ∑ (individual rate × travel time) ÷ ∑ (travel time).

[0074] The applicability and effectiveness of the model are verified using specific cities as cases. For example, Zhengzhou is used as the research object, and flood scenarios with different rainfall intensities are simulated to verify the consistency between the vulnerability distribution map predicted by the model and the actual disaster situation. The sources of errors are analyzed, and the model performance is optimized by adjusting the indicator weights, parameters and data accuracy to improve the accuracy and practicality of the assessment.

[0075] In summary: The application comprehensively considers the impact of multiple factors on the vulnerability of the road network, including road network structure, function, terrain, rainfall, drainage, etc., and the evaluation results are more accurate and comprehensive. Advanced technical means are used, such as GIS technology for data processing and analysis, HiPIMS model for high-precision flood simulation, and big data mining technology for model verification, which improves the scientificity and reliability of the model. The model verification method is based on keyword indexed big data mining technology, which can better reflect the accuracy of the model in practical applications and avoid the unreliable model problem caused by the lack of effective verification data in traditional verification methods. This application can be applied to urban planning and management departments to provide a scientific basis for urban flood control and drainage planning, road facility construction and maintenance, and traffic management. In terms of disaster warning and emergency response, the model can quickly assess the vulnerability of the road network, provide support for the formulation of reasonable evacuation routes and rescue plans, and help reduce the impact of heavy rain and waterlogging disasters on urban transportation and residents' lives.

[0076] 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 assessing the vulnerability of urban road network traffic flood disasters based on multi-source and multi-modal data, characterized in that: The steps include: S100, collecting and preprocessing multi-source and multi-modal traffic disaster data; S200, constructing a vulnerability index system for the pre-processed disaster data, including index extraction and index quantification, wherein the extracted indicators include flood characteristic indicators, vulnerability indicators and traffic flow data. The index quantification is to use the convolutional neural network method to extract flood depth and range characteristics from different types of data, and to fuse the features extracted from multi-source and multi-modal data through a weighted fusion method; S300, constructing a road network disaster vulnerability model, analyzing the road network disaster vulnerability through the road network disaster vulnerability model, and classifying the road network flood vulnerability; S310, classify roads into different levels according to their socio-economic functional attributes, road network connectivity and spatial distribution geometric characteristics; S320, using the curve fitting toolbox in Matlab, a quadratic function is used to fit the water depth-safe speed curve under existing similar urban conditions; S330, based on the collected original maximum safe speed limit, water depth data, road capacity and traffic flow data of different road grades, using a curve regression fitting method to derive the specific relationship between the safe driving speed and water depth under different road grades; S340, using the high-performance hydrodynamic integrated model HiPIMS to simulate and obtain the real-time water depth distribution of road network traffic; S350, obtaining / estimating the road section rate based on mobile phone signaling data; S360, input road grade, flood depth, and traffic flow data into the GIS platform, and use spatial analysis tools to classify the flood vulnerability of roads based on the maximum safe speed of the road and the reduction in the response to floods. The road vulnerability is divided into five levels: low, secondary, medium, high, and very high, and a flood vulnerability level map is drawn for each road section.

2. The method for assessing urban road network traffic flood vulnerability based on multi-source and multi-modal data according to claim 1 is characterized in that: The multi-source and multi-modal traffic disaster data in step S100 include social media data, traffic data, socioeconomic data, land use types, DEM, and also flood-related road network disaster data acquired in real time through the Internet of Things and remote sensing.

3. The method for assessing urban road network traffic flood vulnerability based on multi-source multi-modal data according to claim 2 is characterized in that: The preprocessing of the collected multimodal traffic disaster data includes: image cleaning and labeling of image data; word segmentation and syntactic analysis of text data; and reclassification and grading of traffic data.

4. The method for assessing urban road network traffic flood vulnerability based on multi-source multi-modal data according to claim 1, characterized in that: The flood characteristic indicators in step S200 are obtained through HiPIMS model simulation, including real-time water depth data. The damage is evaluated by calculating the reduction in vehicle speed through the flood inundation map; the traffic flow data is based on the personnel movement trajectory information in the mobile phone signaling data of the mobile base station.

5. The method for assessing urban road network traffic flood vulnerability based on multi-source and multi-modal data according to claim 1, characterized in that: The fusion of features extracted from multi-source multi-modal data in step S200 specifically includes: first, evaluating the reliability according to the historical performance, authority, coverage and update frequency of the data source, and evaluating the accuracy of the data through error analysis and time extension, and assigning a corresponding weight to each data feature based on the evaluation results; then, performing a weighted average calculation on the feature values ​​from different data sources in the same indicator to generate the final fused feature value.

6. The method for assessing urban road network traffic flood vulnerability based on multi-source and multi-modal data according to claim 1, characterized in that: In step S320, the quadratic function is v0=f(h)=ah 2 +bh+c, where v0 represents the maximum safe speed that can be accommodated when the flooding depth is h, and a, b, and c are function parameters related to the initial maximum speed of the road; the specific relationship in step S330 is given by the formula v0=f(v i , h) indicates that v i represents the original maximum safe speed limit of roads of different levels, and h is the flood depth; the calculation formula for the flood vulnerability of the road in step S360 is: Among them C r represents the vulnerability of the road network, v max It indicates the vehicle capacity under different road grades. It is an adjustment coefficient used to consider the impact of different road grades and service levels on vulnerability.

7. The method for assessing urban road network traffic flood vulnerability based on multi-source multi-modal data according to claim 1, characterized in that: Step S300 also includes: S370, optimizing the model parameters using the gradient ascent method.

8. The method for assessing urban road network traffic flood vulnerability based on multi-source and multi-modal data according to claim 1, characterized in that: It also includes: a method for verifying the road network disaster vulnerability model based on keyword indexing of social media disaster big data; mining flood disaster and disaster severity information and keyword quantification, combining precision, recall and F-score to verify the accuracy of the model; a real-time verification and acquisition method for road traffic speed based on mobile phone signaling data, using mobile phone signaling data and road weights to estimate real-time traffic speeds, analyze traffic flow conditions, and calibrate the model through actual flood and traffic speed data to improve the accuracy and reliability of path algorithms and parameters.

Images