Urban inland inundation and traffic risk monitoring and early warning system based on multi-modal data

Abstract

The invention relates to the technical field of urban inland inundation and traffic risk analysis, in particular to a multi-modal data urban inland inundation and traffic risk monitoring and early warning system, which comprises a data acquisition and transmission module, an information processing and analysis module, an urban inland inundation point detection module and an urban safety early warning response module, a city multi-source heterogeneous data set is obtained through the data acquisition and transmission module; the information processing and analyzing module receives and processes the urban multi-source heterogeneous data set, and processes the urban multi-source heterogeneous data set through a data fusion model to obtain an urban multi-modal database; an urban inland inundation and traffic risk assessment mechanism is constructed in the urban inland inundation point detection module, and inland inundation prediction results and traffic analysis conditions of different inland inundation points in the urban system are obtained based on the assessment mechanism and the multi-modal database; and the urban safety early warning response module receives the waterlogging prediction result, the traffic analysis condition and the urban multi-modal database to perform early warning evaluation and intelligent regulation and control on different areas of the urban system.

Description

Technical Field

[0001] The present invention relates to the technical field of urban waterlogging and traffic risk analysis, and specifically to a monitoring and early warning system for urban waterlogging and traffic risks based on multi-modal data. Background Art

[0002] With the rapid advancement of climate change and urbanization, the problems of urban waterlogging and traffic risks have become increasingly prominent, becoming important factors restricting the sustainable development of cities. Urban waterlogging refers to the phenomenon of waterlogging disasters in low-lying areas of cities due to heavy rainfall or continuous precipitation exceeding the urban drainage capacity. Traffic risks are closely related to the waterlogging situation. Waterlogging not only affects the travel of citizens but also leads to traffic paralysis, putting pressure on urban operations.

[0003] With climate change and urbanization, traditional urban drainage and traffic management systems are unable to effectively handle complex multi-source heterogeneous data, difficult to integrate multiple types of urban data, and have significant deficiencies in warning of waterlogging and traffic risks under extreme weather conditions. At the same time, they cannot conduct waterlogging early warning and traffic command according to the actual situation of the urban system. Especially under extreme weather conditions such as heavy rain, the urban drainage system is prone to overloading operation, and the traffic management department is also difficult to quickly obtain road condition information. With the continuous development of technologies such as the Internet of Things, cloud computing, and artificial intelligence, it is necessary to further improve and optimize the monitoring and early warning system for urban waterlogging and traffic risks, and it is expected to achieve more accurate and efficient monitoring and early warning functions to provide support for urban flood control and traffic management. Summary of the Invention

[0004] In view of the deficiencies of existing methods and the requirements of practical applications, to overcome the deficiencies of current urban waterlogging and traffic risk monitoring methods, meet the actual early warning requirements of urban systems, promote the effective integration of multi-modal data, achieve comprehensive and effective monitoring of urban waterlogging risks and traffic conditions in various regions of the urban system, provide timely and accurate decision-making basis for urban management departments, and further improve the emergency response ability and management level of urban systems. On the one hand, the present invention provides a multi-modal data-based urban waterlogging and traffic risk monitoring and early warning system, which includes: a data collection and transmission module, an information processing and analysis module, an urban waterlogging point detection module, and an urban safety early warning response module; obtaining a set of urban multi-source heterogeneous data through the data collection and transmission module; the information processing and analysis module receiving and processing the set of urban multi-source heterogeneous data, and processing the set of urban multi-source heterogeneous data through a multi-source heterogeneous data fusion model to obtain an urban multi-modal database; constructing an urban waterlogging and traffic risk assessment mechanism in the urban waterlogging point detection module, and obtaining the waterlogging prediction results and traffic analysis situations of different flood-prone points in the urban system based on the urban waterlogging and traffic risk assessment mechanism and the urban multi-modal database; the urban safety early warning response module receiving the waterlogging prediction results, the traffic analysis situations, and the urban multi-modal database to conduct early warning assessment and intelligent regulation on different regions of the urban system.

[0005] The data collection and transmission module, information processing and analysis module, urban waterlogging point detection module, and urban safety early warning response module of the present invention work together to achieve comprehensive collection, processing, assessment, intelligent early warning and regulation of urban multi-source heterogeneous data, improve the accuracy and efficiency of urban waterlogging and traffic risk monitoring and early warning, and help improve the emergency response ability of urban systems.

[0006] Optionally, obtaining a set of urban multi-source heterogeneous data through the data collection and transmission module includes: setting an information perception layer, a preprocessing layer, and a transmission layer in the data collection and transmission module; obtaining initial urban multi-source heterogeneous data through the information perception layer; establishing a data similarity analysis function in the preprocessing layer, and the preprocessing layer using the data similarity analysis function to process the initial urban multi-source heterogeneous data to obtain a similarity analysis result of the initial urban multi-source heterogeneous data; establishing a data condition constraint function in the transmission layer, and the transmission layer combining the data condition constraint function and the similarity analysis result to judge the initial urban multi-source heterogeneous data, and conducting system data transmission according to the condition judgment situation to obtain a set of urban multi-source heterogeneous data. The present invention processes and judges the initial urban multi-source heterogeneous data through the information perception layer, preprocessing layer, and transmission layer, can achieve the collection and transmission of urban multi-source heterogeneous data, and helps improve the overall level of the urban waterlogging and traffic risk monitoring and early warning system.

[0007] Optionally, a data similarity analysis function is established in the preprocessing layer. The preprocessing layer processes the initial urban multi-source heterogeneous data by using the data similarity analysis function, and the obtained similarity analysis result of the initial urban multi-source heterogeneous data includes: establishing a data similarity analysis function based on the distribution and attribute characteristics of the initial urban multi-source heterogeneous data; processing the initial urban multi-source heterogeneous data by using the data similarity analysis function to obtain the similarity analysis result of the initial urban multi-source heterogeneous data; the data similarity analysis function satisfies the following relationship: Wherein, represents the similarity result of any two sets of data in the urban multi-source heterogeneous data, represents a set of data in the urban multi-source heterogeneous data, represents another set of data in the urban multi-source heterogeneous data, represents the total number of feature categories of the original multi-source heterogeneous data, represents the corresponding weight coefficient, represents the corresponding weight coefficient, represents the multi-source heterogeneous data 's attribute characteristics, represents the multi-source heterogeneous data 's attribute characteristics.

[0008] The data similarity analysis function of the preprocessing layer of the present invention can quickly analyze the similarity degree of different data, providing technical support for the urban waterlogging and traffic risk monitoring and early warning system.

[0009] Optionally, a data condition constraint function is established in the transmission layer. The transmission layer combines the data condition constraint function and the similarity analysis result to judge the initial urban multi-source heterogeneous data, and performs system data transmission according to the condition judgment situation to obtain a set of urban multi-source heterogeneous data, including: setting a similarity threshold for the urban multi-source heterogeneous data based on the historical data of the urban system and the multi-source heterogeneous data; establishing a data condition constraint function based on the similarity threshold and the similarity analysis result; judging the initial urban multi-source heterogeneous data through the data condition constraint function and the similarity analysis result, and performing system data transmission according to the condition judgment situation to obtain a set of urban multi-source heterogeneous data; The data condition constraint function satisfies the following relationship: Wherein, represents the constraint analysis result of any two sets of data in the urban multi-source heterogeneous data, represents the similarity threshold of the multi-source heterogeneous data, Indicates the similarity result of any two sets of data in the urban multi-source heterogeneous data, Indicates the multi-source heterogeneous data And the multi-source heterogeneous data Are not relevant, Indicates the multi-source heterogeneous data And the multi-source heterogeneous data Are relevant.

[0010] The present invention establishes a data condition constraint function based on the similarity threshold and the similarity analysis result, enabling the system to screen and filter the initial urban multi-source heterogeneous data, and further ensuring the feasibility and effectiveness of the transmitted data set.

[0011] Optionally, the information processing and analysis module receives and processes the urban multi-source heterogeneous data set, and processes the urban multi-source heterogeneous data set through a multi-source heterogeneous data fusion model to obtain an urban multi-modal database, including: setting a multi-source heterogeneous data fusion model based on the actual operation status of the data acquisition and transmission module; processing the urban multi-source heterogeneous data set based on the multi-source heterogeneous data fusion model to obtain an urban multi-modal database; The multi-source heterogeneous data fusion model satisfies the following relationship: Wherein, Represents the multi-source heterogeneous data fusion model, Represents the parameters related to the fusion model, Represents the number of data groups of the original multi-source heterogeneous data, Represents the multi-source heterogeneous data Of the reference mean value, Represents the multi-source heterogeneous data in the urban multi-source heterogeneous data set , Represents the multi-source heterogeneous data Of the reference mean value, Represents the multi-source heterogeneous data in the urban multi-source heterogeneous data set , Represents the generalization parameter of the fusion model.

[0012] The multi-source heterogeneous data fusion model of the present invention can efficiently integrate data from different sources, maximize the mining of valuable information, and improve the efficiency and decision-making ability of the urban waterlogging and traffic risk monitoring system.

[0013] Optionally, constructing an urban waterlogging and traffic risk assessment mechanism in the urban waterlogging point detection module includes: analyzing urban detection images in the urban multi-modal database based on the urban waterlogging and traffic risk assessment mechanism, and obtaining the phase value analysis result and the spatio-temporal baseline threshold of the urban detection images; the urban waterlogging and traffic risk assessment mechanism analyzes the easily waterlogged points of the urban system in combination with the phase value analysis result and the spatio-temporal baseline threshold, and obtains the analysis result of the easily waterlogged points of the urban system; The phase value analysis result satisfies the following relationship: Wherein, represents the phase value of the urban detection image, represents time the deformation along the radar line of sight, represents time the deformation along the radar line of sight, represents the wavelength of the urban system detection radar.

[0014] The present invention analyzes the urban detection images in the urban multi-modal database, can quickly analyze and timely discover the precursors of waterlogging, and makes the emergency response more rapid and effective.

[0015] Optionally, constructing an urban waterlogging and traffic risk assessment mechanism in the urban waterlogging point detection module, obtaining the waterlogging prediction results and traffic analysis situations of different easily waterlogged points in the urban system based on the urban waterlogging and traffic risk assessment mechanism and the urban multi-modal database includes: obtaining the infiltration rate, the external water volume and the surface water accumulation volume of different easily waterlogged points in the urban system based on the urban waterlogging and traffic risk assessment mechanism and the easily waterlogged point analysis result; obtaining the waterlogging prediction results of different easily waterlogged points in the urban system according to the infiltration rate, the external water volume and the surface water accumulation volume; The infiltration rate satisfies the following relationship: Wherein, represents the infiltration rate of different easily waterlogged points in the urban system at different times, represents the original initial infiltration rate of the urban system, represents the stable infiltration rate of different easily waterlogged points in the urban system, represents the average infiltration decreasing rate of the urban system, represents different times; The external water volume satisfies the following relationship: Wherein, represents the external infiltration water volume of different easily waterlogged points in the urban system, Represents the total water output at the end of the drainage pipe network at different flood-prone points in the urban system, Represents the theoretical water quality index concentration of the urban system drainage pipe network, Represents the actual water quality index concentration of the urban system drainage pipe network, Represents the water quality index concentration of the infiltrated water from outside the drainage pipe network at different flood-prone points in the urban system, Represents the infiltration rate at different flood-prone points in the urban system at different times, Represents the start sampling time of the water volume at different flood-prone points in the urban system, Represents the start sampling time of the water volume at different flood-prone points in the urban system; The surface water accumulation volume satisfies the following relationship: Among them, Represents the surface water accumulation volume at different flood-prone points in the urban system, Represents the net rainfall intensity at different flood-prone points in the urban system, Represents the catchment area at different flood-prone points in the urban system, Represents the catchment area width at different flood-prone points in the urban system, Represents the actual water depth at different flood-prone points in the urban system, Represents the threshold of the ground water storage depth of the urban system, Represents the catchment area slope at different flood-prone points in the urban system, Represents the Manning roughness coefficient.

[0016] The present invention analyzes the waterlogging parameters in sequence based on physical processes and parameter models, making the analysis results corresponding to the urban waterlogging evaluation parameters more accurate and reliable.

[0017] Optionally, constructing an urban waterlogging and traffic risk assessment mechanism in the urban waterlogging point detection module, and obtaining the waterlogging prediction results and traffic analysis situations at different flood-prone points in the urban system based on the urban waterlogging and traffic risk assessment mechanism and the urban multi-modal database includes: obtaining the traffic density, traffic saturation, and traffic driving speed at different flood-prone points in the urban system based on the urban waterlogging and traffic risk assessment mechanism and the flood-prone point analysis results; obtaining the traffic analysis situations at different flood-prone points in the urban system according to the traffic density, the traffic saturation, and the traffic driving speed; The traffic density satisfies the following relationship: Among them, Represents the traffic positioning data density in the area of different flood-prone points in the urban system, Represents the maximum distance between the mobile device and the base station within a certain time interval in the area of different flood-prone points, Indicates the length of the monitored traffic sections at different waterlogging-prone points in the urban system; The traffic saturation satisfies the following relationship: Wherein, The traffic saturation within different waterlogging-prone areas in the urban system, Indicates the average traffic volume within different waterlogging-prone areas over a certain time interval, Indicates the error coefficient of the traffic fitting center point of the road traffic, Indicates the traffic capacity limit threshold of the urban system; The traffic driving speed satisfies the following relationship: Wherein, Indicates the traffic driving speed within different waterlogging-prone areas in the urban system, Indicates the traffic situation index, Indicates the vehicle driving distance within different waterlogging-prone areas in the urban system, Indicates the time measurement error coefficient, Indicates the average time for a vehicle to drive out of different waterlogging-prone areas, The average time for a vehicle to drive into different waterlogging-prone areas.

[0018] The traffic density of the present invention can understand the vehicle distribution in different waterlogging-prone areas; the traffic saturation reflects the ratio of the actual traffic volume to the traffic capacity limit threshold; the traffic driving speed directly reflects the driving efficiency of vehicles in different waterlogging-prone areas, which is conducive to comprehensively evaluating the traffic conditions of the urban system.

[0019] Optionally, the urban waterlogging and traffic risk monitoring and early warning system for the multi-modal data further includes: setting an evaluation parameter membership analysis model in the urban safety early warning response module; performing membership analysis on the waterlogging prediction parameters in the waterlogging prediction results and the traffic analysis parameters in the traffic analysis situation through the evaluation parameter membership analysis model to obtain the membership analysis results of different evaluation parameters; The evaluation parameter membership analysis model satisfies the following relationship: Wherein, Indicates the evaluation results of different risk evaluation parameters after weighting, Indicates the weight vector, Indicates the membership of different traffic evaluation indicators to urban waterlogging and traffic, Indicates the original index matrix.

[0020] The membership analysis model of the present invention can quantify the influence degree of different evaluation indexes on urban waterlogging and traffic risks, making the evaluation process more objective and accurate, reducing the subjectivity of human judgment, and improving the reliability and consistency of the evaluation results.

[0021] Optionally, the urban safety early warning response module receives the waterlogging prediction result, the traffic analysis situation, and the urban multimodal database to conduct early warning evaluation and intelligent regulation on different regions of the urban system, including: the urban safety early warning response module combines the membership analysis result, the waterlogging prediction result, the traffic analysis situation, and the urban multimodal database to conduct early warning evaluation on different regions of the urban system, and obtains the urban waterlogging and traffic prediction results of different regions in the urban system; the urban safety early warning response module realizes the safety early warning and intelligent regulation of the urban system based on the urban waterlogging and traffic prediction results. The present invention combines the membership analysis result, the waterlogging prediction result, the traffic analysis situation, and the urban multimodal database, enabling the urban safety early warning response module to comprehensively and integrally evaluate the waterlogging and traffic risks of different regions in the urban system, making the evaluation results more accurate and being able to more truly reflect the actual risk situation and actual environmental conditions of the urban system. Brief Description of the Drawings

[0022] Figure 1 It is a flow chart of the urban waterlogging and traffic risk monitoring and early warning system with multimodal data of the present invention; Figure 2 It is a schematic diagram of the operation mechanism and process of different modules in the urban waterlogging and traffic risk monitoring and early warning system with multimodal data of the present invention; Figure 3 It is a structural diagram of the urban waterlogging and traffic risk monitoring and early warning system with multimodal data of the present invention. Detailed Embodiments

[0023] Specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described here are only for illustrative purposes and are not used to limit the present invention. In the following description, in order to provide a thorough understanding of the present invention, a large number of specific details are elaborated. However, it is obvious to those of ordinary skill in the art that the present invention does not have to adopt these specific details. In other instances, well-known circuits, software, or methods are not specifically described to avoid obscuring the present invention.

[0024] Throughout the specification, references to "one embodiment", "an embodiment", "one example" or "an example" mean that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Thus, the phrases "in one embodiment", "in an embodiment", "one example" or "an example" that appear throughout the specification do not necessarily all refer to the same embodiment or example. In addition, the particular features, structures, or characteristics may be combined in any suitable combination and / or sub-combination in one or more embodiments or examples. Further, those of ordinary skill in the art should understand that the diagrams provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0025] Please refer to Figure 1 , through the fusion processing of three levels of information perception, preprocessing, and transmission, the present invention integrates different data sources, evaluates urban waterlogging and traffic risks, and provides efficient and accurate decision-making support. The present invention provides a system for monitoring and warning urban waterlogging and traffic risks with multi-modal data. The above system includes the following steps: In the system for monitoring and warning urban waterlogging and traffic risks with multi-modal data, a data acquisition and transmission module, an information processing and analysis module, an urban waterlogging point detection module, and an urban safety warning response module are provided.

[0026] S1. Obtain a set of urban multi-source heterogeneous data through the data acquisition and transmission module. The specific implementation steps and related content are as follows: In order to improve the acquisition efficiency and accuracy of urban multi-source heterogeneous data, optimize the data transmission quality and reliability, and enhance the overall performance and flexibility of the data acquisition and transmission module, an information perception layer, a preprocessing layer, and a transmission layer are provided in the data acquisition and transmission module.

[0027] First, initial urban multi-source heterogeneous data can be obtained through the information perception layer.

[0028] The multi-modal data acquisition module includes a multi-modal data acquisition and information perception layer. The above perception layer is composed of different sensors and detection devices distributed throughout the city. The sensors include, but are not limited to, rain sensors, water level sensors, water accumulation depth sensors, flow velocity sensors, flow sensors, water quality sensors, gas sensors, and video monitoring devices, etc. Based on this, key information such as rainfall conditions, water conditions, water accumulation status, traffic flow, and speed status in various regions of the city can be captured in real time and accurately. Then, through the data acquisition and transmission module, initial urban multi-source heterogeneous data can be quickly obtained.

[0029] The information perception layer, based on various sensors and data acquisition devices, can comprehensively and deeply collect multi-dimensional information in the urban system. The above information mainly covers multiple aspects such as temperature, humidity, light intensity, sound, location, and video images, providing a data basis for subsequent urban waterlogging and traffic risk analysis and prediction. On the other hand, the information perception layer can convert the physical information of the urban system into digital signals or data streams, facilitating the smooth circulation and information interaction between the data acquisition and transmission module and the information processing and analysis module.

[0030] The information perception layer can monitor and collect data in real time, quickly capture the change information of the urban system, and collect and obtain relevant information in a timely manner, providing a basis for rapid decision-making and prompt action in the urban system, enabling the data acquisition and transmission module to respond more flexibly and efficiently to various complex situations.

[0031] Then, a data similarity analysis function is established in the preprocessing layer. The preprocessing layer uses the data similarity analysis function to process the initial urban multi-source heterogeneous data, obtaining the similarity analysis result of the initial urban multi-source heterogeneous data.

[0032] The preprocessing layer in the data acquisition and transmission module can receive and back up in real time the initial urban multi-source heterogeneous data transported by the information perception layer. At the same time, a data similarity analysis function is established based on the distribution and attribute characteristics of the initial urban multi-source heterogeneous data. Based on the distribution situation and attribute characteristics of the initial urban multi-source heterogeneous data, the internal relationship and information difference between the initial urban multi-source heterogeneous data are further analyzed. Based on this, a data similarity analysis function is designed and constructed. This function fully considers factors such as the distribution law, attribute type, value range of urban monitoring data, and the correlation between different data, and comprehensively uses knowledge and methods in fields such as statistics and machine learning. It can perform similarity analysis on different data groups in the initial urban multi-source heterogeneous data, revealing the similarities and differences between different data groups, providing technical support for initial data processing, mining, and application.

[0033] Based on the initial urban multi-source heterogeneous data from different sensor devices, in the embodiment, it is classified and managed in the form of data groups, and it is set that the initial urban multi-source heterogeneous data constitutes a data set , which contains n data combinations, that is, the data set satisfies the following relationship . Each data group contains specific urban detection information, and different data groups are different in structure and content. At the same time, based on the multi-source heterogeneous data set, the corresponding attributes of different data groups are analyzed, and the attribute set is denoted as , and its attribute set satisfies the following relationship , and the attribute describes the data group The feature or dimension information therein. Based on this, for the multi-source heterogeneous data in it, similarity analysis is carried out to reveal the internal connections and differences between different data groups, providing information basis and support for subsequent urban data processing and applications.

[0034] In the embodiment, the Manhattan distance algorithm is also introduced. Based on the distribution situation, attribute characteristics of the initial urban multi-source heterogeneous data and the Manhattan distance algorithm, a data similarity analysis function is established. Based on the analysis function, similarity analysis is carried out on the urban multi-source heterogeneous data, which can comprehensively and objectively evaluate the difference degree between data, providing a reliable similarity measurement basis for data comparative analysis.

[0035] The above data similarity analysis function satisfies the following relationship: Among them, represents the similarity result of any two groups of data in the urban multi-source heterogeneous data, represents a group of data in the urban multi-source heterogeneous data, represents another group of data in the urban multi-source heterogeneous data, represents the total number of feature categories of the original multi-source heterogeneous data, represents the corresponding weight coefficient, represents the corresponding weight coefficient, represents the multi-source heterogeneous data 's attribute characteristics, represents the multi-source heterogeneous data 's attribute characteristics.

[0036] The data similarity analysis function measures the similarity of different data groups through attribute characteristics and weighted differences, making the data screening result more accurate and improving the data quality.

[0037] The similarity result between any two groups of data in the urban multi-source heterogeneous data can quantify the similarity situation between different data. The smaller the value, the more similar any two groups of data are, and the larger the value, the greater the difference between the two groups of data.

[0038] Among them, represents any group of data in the urban multi-source heterogeneous data, represents another group of data in the urban multi-source heterogeneous data. x and y respectively represent any non-repeated data groups in the urban multi-source heterogeneous data, all from the data set in it, and .

[0039] The total number of feature categories of the initial multi-source heterogeneous data determines the number of attribute features to be considered in similarity calculation.

[0040] represents the weight coefficient corresponding to data group x, which can be used to adjust the importance of each attribute feature in data group x during similarity calculation; represents the weight coefficient corresponding to data group y, which is also used to adjust the importance of each attribute feature in data group y during similarity calculation. The weight coefficients of different data groups can be set according to actual analysis requirements and the situation of the urban system to reflect the influence degree of different attribute features on similarity calculation.

[0041] represents the attribute feature of multi-source heterogeneous data x, which is the value corresponding to data group x on the feature attribute; represents the attribute feature of multi-source heterogeneous data y, which is the value corresponding to data group y on the feature attribute. The above attribute feature values all belong to and can describe the specific numerical values or categories of different data group features, which is the information basis for similarity calculation.

[0042] To sum up, the data similarity analysis function measures the similarity between different data groups through the sum of weighted differences of any data group on each attribute feature, and flexibly controls the contribution degree of different attribute features in similarity calculation through the data group weight coefficient, so as to meet different similarity measurement requirements.

[0043] Using the above data similarity analysis function to process the initial urban multi-source heterogeneous data, the similarity analysis result of the initial urban multi-source heterogeneous data is obtained, and the similarity between different data groups is further analyzed based on the similarity results of any two groups of data in the urban multi-source heterogeneous data.

[0044] Applying the data similarity analysis function to analyze the initial urban multi-source heterogeneous data, the similarity analysis results between different data groups are obtained. The above results reflect the similarity or difference degree between different data groups. Further, analyzing the similarity results, the greater the distance between the data, that is, the higher the similarity measurement value, the smaller the similarity between the two groups of data, and vice versa.

[0045] Through data similarity analysis, the similarity and difference between urban multi-source heterogeneous data can be understood more accurately, providing a basis for subsequent urban monitoring data processing, mining and analysis. At the same time, the similarity analysis results can be used as the basis for data fusion and integration, which is conducive to the effective fusion of urban multi-source heterogeneous data in the future and improves the overall quality and usability of multi-source heterogeneous data.

[0046] On the other hand, understanding the similarity between multi-source heterogeneous data in cities helps to make more intelligent decisions and plans in the fields of urban planning, traffic management, environmental monitoring, etc., improve the intelligent level and management efficiency of urban waterlogging disaster prevention and traffic control, and better understand the distribution and utilization of urban resources, providing a basis for the optimal allocation and efficient utilization of urban resources.

[0047] Immediately afterwards, a data condition constraint function is established in the transport layer. The above transport layer combines the data condition constraint function and the similarity analysis result to judge the initial urban multi-source heterogeneous data, and conducts system data transmission according to the condition judgment situation, so as to obtain a set of urban multi-source heterogeneous data.

[0048] Set the similarity threshold of urban multi-source heterogeneous data based on the historical data and multi-source heterogeneous data of the urban system. In the embodiment, based on the analysis and comprehensive consideration of the historical data of the urban system and the characteristics of multi-source heterogeneous data, the similarity threshold of urban multi-source heterogeneous data is reasonably set and denoted as The above threshold not only reflects the regularity and change trend of historical data, but also takes into account the diversity and complexity of current monitoring data. At the same time, it reflects the similarity level between urban data, providing an accurate reference standard for subsequent constraint condition judgment and analysis.

[0049] Establish a data condition constraint function based on the similarity threshold and the similarity analysis result.

[0050] In the embodiment, according to the similarity threshold and the similarity analysis result, a data condition constraint function is constructed in the transport layer. This function not only considers the similarity characteristics between multi-source heterogeneous data, but also combines the actual needs and data characteristics of the urban system. At the same time, based on the similarity of multi-source heterogeneous data, a data threshold of the urban correlation coefficient is designed. This threshold will be used as an important basis and reference standard for screening the initial monitoring data of the city, providing a reference and judgment basis for the intelligent management of urban multi-source heterogeneous data.

[0051] The above data condition constraint function satisfies the following relationship: Among them, represents the constraint analysis result of any two groups of data in the urban multi-source heterogeneous data, represents the similarity threshold of the multi-source heterogeneous data, represents the similarity result of any two groups of data in the urban multi-source heterogeneous data, represents the multi-source heterogeneous data is not related to the multi-source heterogeneous data , represents the multi-source heterogeneous data is not related to the multi-source heterogeneous data Related

[0052] The above data condition constraint function is mainly used to judge the correlation between any two sets of data in urban multi-source heterogeneous data, that is, after determining whether specific similarity constraint conditions are met, it indicates that the information of this data set meets the requirements of urban risk judgment and can be used as the early warning basis and information foundation for waterlogging and traffic risks; if there is no correlation between any two sets of data in urban multi-source heterogeneous data, then the two sets of data need to be adjusted, optimized, or re-collected to meet the conditions and requirements of waterlogging and traffic risk early warning.

[0053] The constraint analysis result of any two sets of data in urban multi-source heterogeneous data is a binary value (0 or 1), which is used to indicate whether specific threshold conditions are met between different data sets and whether it can be used as the information basis and judgment foundation for urban waterlogging and traffic risk monitoring and early warning.

[0054] The similarity threshold of multi-source heterogeneous data is a preset value, which serves as the boundary for judging whether the data similarity reaches the relevant standard, that is, the basic information regulations and requirements for urban waterlogging and traffic risk monitoring and early warning.

[0055] The similarity result of any two sets of data in urban multi-source heterogeneous data reflects the degree of proximity of different data sets in the feature space of urban multi-source heterogeneous data.

[0056] The constraint analysis result of 0 indicates that there is no compliance with the threshold conditions between any two sets of data in urban multi-source heterogeneous data, that is, the similarity of any two sets of data is lower than the threshold, and it cannot be used as the information basis and judgment foundation for urban waterlogging and traffic risk monitoring and early warning.

[0057] The constraint analysis result of 1 indicates that there is compliance with the threshold conditions between any two sets of data in urban multi-source heterogeneous data, that is, the similarity of any two sets of data meets the threshold requirements and can be used as the information basis and judgment foundation for urban waterlogging and traffic risk monitoring and early warning.

[0058] In summary, based on the data condition constraint function, compare the similarity result of any two sets of data with the similarity threshold to judge whether any two sets of data meet the preset correlation conditions of the urban system and whether they are related to the monitoring characteristics of urban waterlogging and traffic risks. When the similarity result is less than or equal to the threshold, the data is considered relevant, and the relevant data information can be used as the information basis for urban waterlogging and traffic risk prediction; when the similarity result is greater than the threshold, the data is considered irrelevant, and the relevant data information does not meet the information judgment conditions for urban waterlogging and traffic risk prediction. The data condition constraint function provides a tool for the screening, fusion, and analysis of urban multi-source heterogeneous data.

[0059] Judging the initial urban multi-source heterogeneous data through the above data condition constraint function and similarity analysis results, and performing system data transmission according to the condition judgment situation to obtain an urban multi-source heterogeneous data set. The optimized urban multi-source data set, that is, the urban multi-source heterogeneous data set of the embodiment is obtained and satisfies the following relationship: Among them, represents the urban multi-source heterogeneous data set, represents the sub-set of data groups after conditional constraint screening represents the sub-set of data groups after conditional constraint screening represents the sub-set of data groups after conditional constraint screening

[0060] Taking the data condition constraint function as a judgment tool, by comparing the similarity results of any two groups of data in the urban multi-source heterogeneous data with the similarity threshold, to evaluate whether the initial data meets the preset conditions of the urban system for relevance, especially the relevance with the characteristics of urban waterlogging and traffic risk monitoring. When the similarity result of the data is less than or equal to the threshold, it is determined that the relevant data is relevant, so it can be used as the information basis for urban waterlogging and traffic risk prediction. On the contrary, if the similarity result of the data is greater than the threshold, it is considered that the relevant data does not match the prediction conditions of urban waterlogging and traffic risks, that is, the data is not relevant. The above data condition constraint function provides strong technical support for the screening, fusion and analysis process of urban multi-source heterogeneous data.

[0061] Furthermore, the processing method and analysis model of urban multi-source heterogeneous data in this embodiment are only an optional condition of the present invention. In one or some other embodiments, according to the prediction requirements of urban waterlogging and traffic risks and the actual situation of urban multi-source heterogeneous data, the data processing method and module analysis model can be optimized through specific prediction requirements and data characteristics, so as to more accurately capture the key information in the data, thereby improving the accuracy of the prediction results of urban waterlogging and traffic risks, and further enhancing the adaptability of the multi-modal data urban waterlogging and traffic risk monitoring and early warning system to different urban environments.

[0062] S2. The information processing and analysis module receives and processes the urban multi-source heterogeneous data set, and processes the urban multi-source heterogeneous data set through the multi-source heterogeneous data fusion model to obtain an urban multi-modal database. The specific implementation content is as follows: First, set the multi-source heterogeneous data fusion model based on the actual operation status of the data acquisition and transmission module.

[0063] ​​​In the embodiment, a multi-source heterogeneous data fusion model is designed based on the actual operating conditions of the data acquisition and transmission module. The above model not only considers the real-time and accuracy of data acquisition, but also fully takes into account the stability and reliability of data transmission, ensuring that the urban multi-source heterogeneous data set transported by the data acquisition and transmission module can be received efficiently and comprehensively.

[0064] After receiving the urban multi-source heterogeneous data set, the data acquisition and transmission module immediately starts a series of data processing processes, including but not limited to cleaning, verifying, integrating, and analyzing the urban multi-source heterogeneous data set. Through the cleaning step, the noise and outliers in the data are removed, ensuring the purity of the data; through the verification process, the accuracy and consistency of the data are verified, ensuring the reliability of the data; through the integration process, the heterogeneous data from different sources are organically integrated, which is beneficial to deeply analyze and explore the potential information and changing rules in the urban multi-source heterogeneous data, and further ensure the accurate monitoring and early warning of urban waterlogging and traffic risks.

[0065] The relevant content of the data integration process is as follows: In the embodiment, the data processing module can comprehensively integrate and fuse multi-source heterogeneous data. The data processing module obtains the urban multi-source heterogeneous data set of the data acquisition and transmission module. The above set includes n groups of multi-source heterogeneous data, and each group of data includes unique information and features. Since the above data comes from different sensors and there are differences and heterogeneity between different data groups, a practical value needs to be set for each group of data, denoted as , to accurately reflect the real situation it represents.

[0066] To effectively integrate the urban multi-source heterogeneous data set, an integration generalization parameter is introduced. This parameter can effectively adjust and optimize the fusion method between different data groups, ensuring that the fused data can not only retain the unique information of each group of data, but also form a consistent and accurate overall representation.

[0067] Based on the above settings and information, a fusion model of multi-source heterogeneous data is established in the data processing module. The model fully considers the practical value and generalization parameter of each group of data, and through specific algorithms and mechanisms, organically fuses the heterogeneous data from different sources to form a unified and reliable data set, providing a basis for the practical application and effective early warning of the urban waterlogging and traffic risk monitoring and early warning system for multi-modal data.

[0068] The above multi-source heterogeneous data fusion model satisfies the following relationship: Among them, represents the multi-source heterogeneous data fusion model, Represents the parameters related to the fusion model, Represents the number of data groups of the original multi-source heterogeneous data, Represents the multi-source heterogeneous data The reference mean value, Represents the multi-source heterogeneous data in the urban multi-source heterogeneous data set , Represents the multi-source heterogeneous data The reference mean value, Represents the multi-source heterogeneous data in the urban multi-source heterogeneous data set , Represents the generalization parameters of the fusion model.

[0069] The result of the multi-source heterogeneous data fusion model is obtained based on the differences between different data groups and the generalization parameters, and a data set with consistent data quality and format can be obtained quickly.

[0070] The parameters related to the fusion model can be used to adjust the weight or ratio of the fusion result. The above parameters can be dynamically adjusted and optimized according to data characteristics or fusion requirements to ensure the reliability and feasibility of the data fusion result.

[0071] The number of data groups of the original multi-source heterogeneous data refers to all relevant data groups that need to be considered during the fusion process to ensure the comprehensiveness and accuracy of the fusion result.

[0072] The reference mean values of different data groups in the multi-source heterogeneous data are mainly calculated based on historical data, experience, and existing technical formulas, and will be used as reference standards during the fusion process.

[0073] The actual values of different multi-source heterogeneous data groups in the urban multi-source heterogeneous data set refer to the specific data that needs to be processed and analyzed during the fusion process.

[0074] The generalization parameters of the fusion model are mainly used to adjust the tolerance or sensitivity of the fusion model to differences, that is, to control the degree of emphasis on individual data differences in the fusion result. A higher generalization parameter value can make the fusion result smoother; a lower generalization parameter value makes the fusion result more sensitive.

[0075] The multi-source heterogeneous data fusion model takes into account the differences between different data groups, the reference mean values, and the generalization parameters, and calculates the fusion result of the urban multi-source heterogeneous data set through weighted summation. This model can be used to process and analyze heterogeneous data from different sources to extract valuable information, which helps the urban waterlogging and traffic risk monitoring and early warning system to make more accurate predictions and monitoring results.

[0076] By processing the urban multi-source heterogeneous data set based on the multi-source heterogeneous data fusion model in the information processing and analysis module, an urban multi-modal database can be obtained quickly.

[0077] In the embodiment, the multi-source heterogeneous data fusion model is used to comprehensively and systematically process the urban multi-source heterogeneous data set, and an urban multi-modal database is successfully obtained. This database not only integrates data from different sources and in different formats, but also improves the accuracy and consistency of the urban multi-source heterogeneous data through the fusion model.

[0078] Furthermore, an evaluation mechanism for the urban multi-source heterogeneous data set is also set in the information processing and analysis module.

[0079] After the above series of optimization steps and integration processes, a successfully integrated urban multi-source heterogeneous data set is obtained. The above data set not only integrates data from different sources and in different formats, but also improves the accuracy and consistency of urban data through the fusion model. To more comprehensively evaluate the quality of the integrated data set, the information processing and analysis module needs to evaluate the system data set. In the embodiment, the multi-modal data fusion time delay result and the quality coefficient of the fused multi-modal data are selected as two evaluation indicators for the data set.

[0080] The multi-modal data fusion time delay result reflects the efficiency in the fusion processing process. By calculating the time required from data reception to fusion completion, the real-time performance and response speed of the fusion processing are further evaluated. A shorter time delay means better processing efficiency and can better meet the needs of real-time data analysis and practical applications.

[0081] The multi-modal data fusion time delay result needs to satisfy the following relationship: Where, represents the delay result of the multi-modal data fusion time, represents the multi-modal data fusion time, represents the original transmission time of the multi-modal data.

[0082] The multi-modal data fusion time delay result refers to the time difference required for the multi-modal data from original transmission to completion of fusion processing.

[0083] The multi-modal data fusion time refers to the total time required from the start of multi-modal data fusion processing to the completion of fusion processing, including but not limited to the time required for various links such as data preprocessing, feature extraction, fusion algorithm execution, and result generation. It reflects the overall efficiency of the fusion processing process. A shorter fusion time means higher processing speed and real-time performance.

[0084] The original delivery time of multimodal data refers to the time required for multimodal data to reach the fusion processing system from the source, that is, the time from data generation to being received by the system. It mainly includes data transmission, storage, retrieval, etc. time, which is the premise of data fusion processing. A longer original delivery time may increase the overall delay of data fusion.

[0085] The time delay result of multimodal data fusion is the difference between the multimodal data fusion time and the original delivery time of multimodal data.

[0086] When the delay result is a positive number, it indicates that the fusion processing time exceeds the original delivery time, and there is a fusion delay.

[0087] When the delay result is a negative number, it means that the fusion processing time is shorter than the original delivery time, indicating that the fusion processing is more efficient or there is a significant delay in the original delivery process.

[0088] By optimizing and it is possible to reduce , thereby improving the real-time performance and efficiency of multimodal data fusion. The above time delay result of multimodal data fusion is an important indicator to measure the efficiency of multimodal data fusion processing. By optimizing relevant parameters, the real-time performance and accuracy of data fusion can be effectively improved to meet the operation requirements of the urban waterlogging and traffic risk monitoring and early warning system.

[0089] The quality coefficient of the fused multimodal data is mainly used to measure the quality of the data after fusion processing. In the embodiment, according to multiple dimensions such as the accuracy, integrity, and consistency of the data, a comprehensive evaluation index is designed to evaluate the quality of multimodal data. A higher data quality coefficient indicates that the data after fusion processing is more reliable and accurate, and can better support subsequent urban waterlogging and traffic risk analysis and prediction.

[0090] The quality coefficient of multimodal data needs to satisfy the following relationship: Among them, represents the quality coefficient of the fused multimodal data as an evaluation index, represents the number of feature information in the fused data, represents the total number of information in the fused data, represents the quality evaluation parameter.

[0091] The quality coefficient of multimodal data is an index used to evaluate the quality of the fused multimodal data. It reflects the proportion of the richness of the feature information of the fused data relative to the total information and is adjusted by the quality evaluation parameter.

[0092] The number of characteristic information in the fused data refers to the number of representative characteristic information in the fused data that can reflect urban waterlogging and traffic conditions. The above characteristic information is the key to urban monitoring and risk prediction, and directly determines the effectiveness and practicality of the urban waterlogging and traffic risk monitoring and early warning system.

[0093] The total number of all information in the fused data refers to the total amount of all information contained in the fused data, including but not limited to characteristic information, noise information, redundant information, etc. It further reflects the overall scale and information content of the fused data, and is the information basis for the operation of the urban waterlogging and traffic risk monitoring and early warning system.

[0094] The quality evaluation parameter is a parameter used to adjust the data quality evaluation standard. It can be set and optimized according to the application scenarios, waterlogging conditions and traffic conditions of specific cities. The introduction of the evaluation parameter makes the data quality coefficient more flexible and scientific, and can be adjusted according to different actual application scenarios and early warning requirements. The larger the above evaluation parameter value, the smaller the data quality coefficient, indicating that the higher the requirement for data quality.

[0095] The multi-modal data quality coefficient is obtained by taking the ratio of the number of characteristic information in the fused data to the total number of all information in the fused data and then adjusting it by the square of the quality evaluation parameter. The larger the value of the multi-modal data quality coefficient, the higher the proportion of the characteristic information in the fused data relative to all information, and the better the data quality.

[0096] In this embodiment, an optimized fusion algorithm is also introduced to improve the data processing accuracy. The number of characteristic information can be increased, and the number of information can be reduced after removing noise and redundant information, so as to further improve the evaluation index value and enhance the quality of the fused multi-modal data.

[0097] The above multi-modal data quality coefficient is an index that comprehensively reflects the quality of the fused data. By reasonably setting and adjusting relevant parameters, the data quality can be evaluated more accurately, providing strong support for data analysis and decision-making.

[0098] By calculating the above two evaluation indexes, the data set after fusion processing is evaluated, so that the information processing and analysis module can output a high-quality urban multi-source heterogeneous data set, providing technical support and information basis for the intelligent construction and risk management of the city.

[0099] In order to further improve the accuracy of the urban waterlogging prediction results and the actual traffic situation, in the embodiments, the analysis techniques and mathematical models of the data collection and transmission module and the information processing and analysis module are fully utilized, which can efficiently process a large amount of urban multi-source heterogeneous data, analyze and extract the hidden information rules and transformation patterns from it, and provide support for urban waterlogging prediction and traffic warning. The calculation models in different modules can automatically learn and optimize the relevant model algorithms, continuously improve the accuracy and reliability of the urban multi-modal database, help to more deeply understand the mechanism of urban waterlogging occurrence and the operation of urban traffic, and thus provide an important information basis and reference for urban planning and management, and help the city better cope with the possible urban waterlogging risks and traffic restrictions in the future.

[0100] Furthermore, the method for obtaining the urban multi-modal database in this embodiment is only an optional condition of the present invention. In one or some other embodiments, the method for obtaining the urban multi-modal database can be optimized according to the characteristic attributes of the urban multi-source heterogeneous data and the intelligent control requirements of different modules. There are differences in the multi-source heterogeneous data in different regions of the city. By optimizing the acquisition method, it can be ensured that the urban multi-modal database is more in line with the actual data characteristics, improve the adaptability of the multi-modal database, and thus improve the accuracy and effectiveness of the subsequent analysis and prediction results.

[0101] S3. Construct an urban waterlogging and traffic risk assessment mechanism in the urban waterlogging point detection module, and obtain the urban waterlogging prediction results and traffic analysis situations of different flood-prone points in the urban system based on the urban waterlogging and traffic risk assessment mechanism and the urban multi-modal database. The specific implementation content is as follows: The urban waterlogging and traffic risk monitoring and warning system for multi-modal data further includes: analyzing the urban detection images in the urban multi-modal database based on the urban waterlogging and traffic risk assessment mechanism, and obtaining the phase value analysis results and spatio-temporal baseline thresholds of the urban detection images.

[0102] The urban waterlogging and traffic risk monitoring and warning system not only has basic data optimization and screening functions, but also can analyze the urban detection images in the urban multi-modal database. The system uses advanced image processing techniques and algorithms to perform phase value analysis on the urban detection images to reveal the hidden spatio-temporal change characteristics and rules in the images. At the same time, in combination with historical data and real-time monitoring data, the spatio-temporal baseline threshold is set and calculated as an important basis for judging urban flood-prone points.

[0103] The urban waterlogging point detection module can analyze the phase values of urban detection images, accurately identify the areas prone to waterlogging in the city, and then analyze the traffic congestion or accident occurrence in the relevant areas. The analysis results of the phase values and the spatio-temporal baseline threshold provide intuitive and quantitative evaluation indicators. When the urban detection image exceeds the set spatio-temporal baseline threshold, the system will automatically trigger the waterlogging point marking mechanism, timely update and release the urban waterlogging point information, providing guarantee for subsequent emergency response and decision-making support.

[0104] In addition, the urban waterlogging point detection module also has the ability of continuous learning and optimization. It can dynamically adjust and improve the analysis results of the phase values and the spatio-temporal baseline threshold according to the continuously accumulated historical data and experience cases, so as to improve the accuracy of the analysis results of urban waterlogging points.

[0105] The above analysis results of the phase values satisfy the following relationship: Among them, represents the phase value of the urban detection image, represents time the deformation along the radar line of sight, represents time the deformation along the radar line of sight, represents the wavelength of the urban system detection radar.

[0106] The analysis results of the phase values of the urban detection image are calculated by comparing the deformations along the radar line of sight at different time points, which reflect the changes of the urban surface or buildings during the radar monitoring period.

[0107] The deformations along the radar line of sight at different times refer to the deformation amounts of the urban surface or buildings along the radar line of sight at different time points. It is one of the key parameters for calculating the phase value and reflects the state changes of the monitoring object at a specific time point. By comparing the deformation amounts at different time points, the change situation of the phase value can be calculated.

[0108] The wavelength of the urban system detection radar refers to the wavelength of the electromagnetic wave emitted by the urban system detection radar. Among them, the wavelength is an important factor affecting the radar monitoring accuracy and the phase value calculation, which determines the sensitivity of the radar to the deformation of the urban surface or buildings. The shorter the above wavelength, the higher the sensitivity of the radar to the deformation, and the smaller changes on the urban surface can be detected.

[0109] The above urban waterlogging and traffic risk assessment mechanism analyzes the waterlogging-prone points of the urban system by combining the phase value analysis results and the spatio-temporal baseline threshold, and obtains the analysis results of the waterlogging-prone points of the urban system. By comparing the deformations in the direction of the radar line of sight at different time points, it is used to reflect the deformations of the urban surface or buildings, providing key information for the urban waterlogging and traffic risk monitoring and early warning system. To sum up, the urban waterlogging and traffic risk monitoring and early warning system can obtain the analysis results of the waterlogging-prone points of the urban system by analyzing the phase values of urban detection images and combining spatio-temporal baseline thresholds, effectively enhancing the system's ability to respond to urban waterlogging and traffic risks.

[0110] The urban waterlogging point detection module can visually manage the analysis results of urban waterlogging-prone points and related information.

[0111] The urban waterlogging point detection module has a visualization management function. It can not only intuitively display the analysis results of urban waterlogging-prone points, including key information such as the locations, scopes, and preliminary risk judgments of waterlogging-prone points, but also dynamically update relevant data. This module also matches a graphical interface and an interactive design, enabling relevant personnel to easily browse, query, and analyze the data of urban system waterlogging-prone points, realizing instant visual monitoring of the data. In addition, it supports multiple data display methods, such as map marking, chart statistics, trend analysis, etc., to meet the needs of different personnel for data visualization. At the same time, this module also has the functions of data export and report generation, and can share the analysis results in an intuitive and easy-to-understand form with relevant departments and decision-makers, providing support for the prevention and control of urban waterlogging and emergency response.

[0112] Then, based on the urban waterlogging and traffic risk assessment mechanism and the above analysis results, the infiltration rates, external water inflows, and surface water accumulations of different waterlogging-prone points in the urban system are obtained; and the waterlogging prediction results of different waterlogging-prone points in the urban system are obtained based on the infiltration rates, external water inflows, and surface water accumulations.

[0113] The infiltration rates of different waterlogging-prone points in the urban system will change dynamically with the system detection time. It is closely related to the urban waterlogging and traffic risk assessment mechanism. By analyzing based on the time dimension and relevant parameter information of urban infiltration, it can be found that there is an internal connection and change relationship between the infiltration rate and time. The above relationship reveals how the infiltration rate changes with time, thus providing a scientific basis for the prediction, prevention and control, and emergency management of urban waterlogging situations.

[0114] The above infiltration rate analysis needs to satisfy the following relationship: Among them, represents the infiltration rate of different waterlogging-prone points in the urban system at different times, represents the original initial infiltration rate of the urban system, Represents the steady infiltration rate of different waterlogging - prone points in the urban system. Represents the average infiltration decreasing rate of the urban system. Represents different moments. The infiltration rate of different waterlogging - prone points in the urban system at different moments refers to the amount of infiltration water per unit area passing through the surface of different waterlogging - prone points in the urban system per unit time. It is a key parameter for evaluating the surface water infiltration capacity and surface runoff of waterlogging - prone points in the study of urban waterlogging.

[0115] The steady infiltration rate of different waterlogging - prone points in the urban system, that is, the value to which the infiltration rate of different waterlogging - prone points gradually tends as time goes by. The above - mentioned steady infiltration rate reflects a relatively stable infiltration state reached after long - term infiltration on the surfaces of different waterlogging - prone points, and it mainly depends on factors such as the soil pore condition, texture, structure, and soil fissures of different waterlogging - prone points.

[0116] The original initial infiltration rate of the urban system, that is, the infiltration rate at the beginning of the infiltration behavior. The above - mentioned initial infiltration rate is the infiltration rate in the initial stage of water infiltration into the soil body. At this time, due to the large suction gradient of the soil wetting front at the waterlogging - prone points, the infiltration intensity is usually large. The size of the initial infiltration rate is affected by factors such as the surface water absorption capacity, soil physical characteristics, biological characteristics, water content, and precipitation characteristics of different waterlogging - prone points in the urban system.

[0117] The average infiltration decreasing rate of the urban system refers to the parameter of the attenuation rate of the infiltration rate with time. During the infiltration process of the urban system, as the urban surface and soil pores are gradually filled with water, the infiltration rate will gradually decrease. The average infiltration decreasing rate can quantify the attenuation process, and its size depends on the surface water absorption capacity, soil properties, and structural characteristics of different waterlogging - prone points in the urban system.

[0118] Different moments refer to the time variable. In the analysis of the infiltration rate, time is an important independent variable, which describes the change of the infiltration process with time. By changing the time variable, the infiltration rate at different moments can be calculated.

[0119] The above - mentioned infiltration rate analysis relationship describes the law of the change of the infiltration rate of different waterlogging - prone points in the urban system with time. Based on obtaining parameter information such as the initial infiltration rate, steady infiltration rate, and average infiltration decreasing rate from the urban multi - modal database, the infiltration rate at different moments in different regions of the city can be predicted, providing a scientific basis for the prevention, control, and emergency management of urban waterlogging.

[0120] In order to analyze the surface external water inflow at different waterlogging-prone points in the urban system, a quantitative analysis technique is introduced in the embodiment. Based on the urban multi-modal database output by the information processing and analysis module, the monitoring and survey data of the drainage pipes, meteorological data, regional water supply data and other multi-source information at different waterlogging-prone points in the urban system can be quickly obtained. By comparing and analyzing the water consumption, the discharged sewage volume and the water volume balance between the upstream and downstream of the pipe network, the dynamic change law of the water volume at different waterlogging-prone points in the urban system can be deeply analyzed. An external water inflow analysis function is established based on the above relevant data, and this function can scientifically characterize the internal relationship between the external water inflow and various influencing factors.

[0121] The above external water inflow analysis function satisfies the following relationship: Among them, represents the external infiltration water inflow at different waterlogging-prone points in the urban system, represents the total water outflow at the end of the drainage pipe network at different waterlogging-prone points in the urban system, represents the theoretical water quality index concentration of the urban drainage pipe network, represents the actual water quality index concentration of the urban drainage pipe network, represents the water quality index concentration of the external infiltration water in the drainage pipe network at different waterlogging-prone points in the urban system, represents the infiltration rate at different waterlogging-prone points in the urban system at different times, represents the start sampling time of the water volume at different waterlogging-prone points in the urban system, represents the start sampling time of the water volume at different waterlogging-prone points in the urban system; The external infiltration water inflow at different waterlogging-prone points in the urban system refers to the external water volume that infiltrates into a specific waterlogging-prone point through the urban drainage pipe network within a certain period of time. The above external water volume can come from groundwater, rainwater, river water or other external water sources, and has an important impact on urban waterlogging and the operation of the drainage system.

[0122] The total water outflow at the end of the drainage pipe network at different waterlogging-prone points in the urban system refers to the total water volume flowing out from the end of the drainage pipe network at a specific waterlogging-prone point within a certain period of time, which includes the normal drainage volume and the external infiltration water volume in the area of this waterlogging-prone point.

[0123] The theoretical water quality index concentration of the urban drainage pipe network refers to the expected concentration of the water quality index in the urban drainage pipe network in the absence of external infiltration water. The above concentration can be used as a reference standard to evaluate the difference between the actual water quality and the theoretical water quality.

[0124] The actual water quality index concentration of the urban drainage pipe network refers to the actual concentration of the water quality index in the urban drainage pipe network in actual measurement. This concentration may be affected by various factors such as external infiltration water, sewage treatment efficiency, and the condition of the drainage system.

[0125] The concentration of water quality indicators of the infiltrated water from external sources in the drainage pipe network at different waterlogging-prone points in the urban system refers to the concentration of water quality indicators of the external water infiltrating into a specific waterlogging-prone point, which varies depending on the source, composition, and nature of the external water and has an important impact on the water quality and treatment efficiency of the drainage system.

[0126] The infiltration rate at different times at different waterlogging-prone points in the urban system refers to the change of the infiltration rate at a specific waterlogging-prone point over time within a certain period. The magnitude and variation law of the above infiltration rate are of great significance for evaluating the infiltration speed and total amount of external water volume.

[0127] The start sampling time and end sampling time of the water volume at different waterlogging-prone points in the urban system define the time range for water volume sampling and analysis. Within this time range, data such as the water volume and water quality of the drainage pipe network can be collected and analyzed, which helps to calculate the external infiltrated water volume and evaluate the operation status of the urban drainage system.

[0128] The above external water volume analysis function comprehensively considers the water volume, water quality, and time factors of the drainage pipe network, and can accurately measure the external infiltrated water volume at different waterlogging-prone points in the urban system, providing a scientific basis for urban waterlogging prevention and control and drainage system management.

[0129] Based on the infiltration rate characteristics of each waterlogging-prone point in the urban system, it is possible to comprehensively evaluate the water infiltration capacity of the surface of different waterlogging-prone point areas. At the same time, combined with meteorological data, geological conditions, and the status of the drainage system, using the infiltration rate as the core index, further analyze and quantify the water volume changes at each waterlogging-prone point in different time periods. This analysis process not only considers the physical properties and structural characteristics of each waterlogging-prone point, but also fully incorporates the dynamic characteristics of the urban hydrological cycle, which can more accurately reveal the internal connection and influence relationship between the water volume and infiltration rate at different waterlogging-prone points, thus providing more scientific and accurate decision-making support for urban waterlogging early warning, drainage system optimization, and water resource management.

[0130] In this embodiment, a surface water accumulation analysis function is established based on the infiltration rate information of each waterlogging-prone point, and it satisfies the following relationship: Among them, represents the surface water accumulation volume at different waterlogging-prone points in the urban system, represents the net rainfall intensity at different waterlogging-prone points in the urban system, represents the catchment area at different waterlogging-prone points in the urban system, represents the catchment area width at different waterlogging-prone points in the urban system, represents the actual water depth at different waterlogging-prone points in the urban system, represents the ground water storage depth threshold of the urban system, Represents the slope of the catchment area of different waterlogging - prone points in the urban system. Represents the Manning roughness coefficient.

[0131] The surface water accumulation volume at different waterlogging - prone points in the urban system refers to the volume of water accumulated on the surface of a specific waterlogging - prone point in the urban system. The above - mentioned surface water accumulation volume changes over time and is an important indicator for evaluating urban waterlogging conditions.

[0132] The change rate of the surface water accumulation volume at different waterlogging - prone points in the urban system refers to the increase or decrease in the surface water accumulation volume per unit time, which reflects the dynamic change process of the water accumulation volume.

[0133] The net rainfall intensity at different waterlogging - prone points in the urban system refers to the effective rainfall amount that falls on a specific waterlogging - prone point per unit time after deducting losses such as evaporation and infiltration, which affects the formation and growth of surface water accumulation.

[0134] The catchment area of different waterlogging - prone points in the urban system refers to the area of the region where rainwater flows towards a specific waterlogging - prone point. It determines the amount of water collected at different waterlogging - prone points and thus affects the final surface water accumulation volume.

[0135] The catchment area width of different waterlogging - prone points in the urban system refers to the width of the catchment area in the horizontal direction. It is related to the shape and size of the catchment area and affects the water flow path and speed at different waterlogging - prone points.

[0136] The actual water depth at different waterlogging - prone points in the urban system refers to the depth of the water accumulated on the surface of a specific waterlogging - prone point, which directly reflects the degree of water accumulation and the risk degree of waterlogging at different waterlogging - prone points.

[0137] The ground water storage depth threshold of the urban system refers to the maximum water storage depth that the urban ground can bear. Exceeding this depth poses a risk of waterlogging and is an important parameter for evaluating the capacity of the urban drainage system and the risk of waterlogging.

[0138] The slope of the catchment area of different waterlogging - prone points in the urban system refers to the ground slope of the catchment area of different waterlogging - prone points. It affects the water flow speed and direction at different waterlogging - prone points, and thus affects the formation and distribution of surface water accumulation at different waterlogging - prone points.

[0139] The Manning roughness coefficient is an empirical coefficient mainly used to describe the resistance of water flow through the surface. It affects the water flow velocity and flow rate at different waterlogging - prone points, and thus affects the change of the surface water accumulation volume.

[0140] The surface water accumulation analysis function comprehensively considers multiple factors such as the net rainfall intensity, catchment area, width, actual water depth, ground water storage depth threshold, catchment area slope, and Manning roughness coefficient of different flood-prone points. It can accurately describe the dynamic change process of surface water accumulation at different flood-prone points in the urban system, providing a scientific basis for urban waterlogging early warning and drainage system optimization.

[0141] Furthermore, based on the infiltration rate, external water volume, and surface water accumulation of the above different flood-prone points, the waterlogging prediction results of different flood-prone points in the urban system are obtained.

[0142] By obtaining the infiltration rate, external water volume, and surface water accumulation of different flood-prone points, the occurrence, development, and recession processes of urban waterlogging can be predicted more accurately. The above analysis results provide key inputs for the waterlogging early warning system, helping to take preventive measures in advance. The data of infiltration rate, external water volume, and surface water accumulation reveal the actual operation status of the drainage system, providing a scientific basis for the daily management and maintenance of the drainage system, and helping to timely discover and solve problems at different flood-prone points.

[0143] Based on the data analysis of the infiltration rate, external water volume, and surface water accumulation of each flood-prone point in the urban system, the waterlogging prediction results of different flood-prone points are further obtained, providing a scientific basis for urban management and waterlogging control decision-making, and helping to build a safer and more sustainable urban environment.

[0144] Immediately afterwards, based on the urban waterlogging and traffic risk assessment mechanism and the analysis results of urban flood-prone points, the traffic density, traffic saturation, and traffic speed of different flood-prone points in the urban system are obtained; at the same time, based on the traffic density, traffic saturation, and traffic speed, the traffic analysis situation of different flood-prone points in the urban system is obtained.

[0145] The traffic density of the road sections at different flood-prone points in the urban system refers to the average maximum distance that mobile devices such as vehicles and pedestrians can cover per unit length of the road section within a specific time period. This characteristic index can effectively measure the density of mobile positioning data within a certain airspace unit, reflecting the distribution and congestion degree of traffic flow on the road section.

[0146] The calculation formula of the above traffic density satisfies the following relationship: Among them, represents the traffic positioning data density within the areas of different flood-prone points in the urban system, represents the maximum distance between mobile devices and base stations within a certain time interval in the areas of different flood-prone points, represents the length of the monitored traffic road sections at different flood-prone points in the urban system.

[0147] The traffic positioning data density in different flood-prone areas of the urban system is an indicator that measures the number of positioning data generated by mobile devices such as vehicles and pedestrian smart devices on a unit road section within a specific flood-prone area. The higher the positioning data density, the more frequent the traffic activities, the more mobile devices, or the higher the collection frequency of positioning data in that flood-prone area, which is of great significance for evaluating the traffic congestion level of the road section, predicting traffic flow changes, and conducting traffic planning and management.

[0148] The maximum distance between a mobile device and a base station within a certain time interval in different flood-prone areas reflects the maximum distance between the mobile device and the base station or positioning reference point when the mobile device moves within the flood-prone area during a specific time period. The above distance can be used as an indicator to measure the activity range or mobility of the mobile device. If the value of the maximum distance is large, it means that the mobile device has a wider activity range or relatively stronger mobility in this area, which is of great importance for understanding the dynamic changes of traffic flow in different flood-prone areas, evaluating road traffic capacity, and optimizing traffic signal control.

[0149] The length of the monitored traffic road section at different flood-prone points in the urban system refers to the road section length within a specific flood-prone area when calculating traffic density. The above road section length is the basis for calculating traffic density, which determines the spatial range used when calculating the coverage distance of mobile devices on a unit length road section. The length of the monitored traffic road section can be determined or optimized according to the actual road layout, traffic monitoring requirements, and data analysis requirements. By reasonably setting the length of the monitored traffic road section, the accuracy and effectiveness of traffic density calculation can be ensured.

[0150] The above parameters are important parameters for calculating the traffic density of road sections at different flood-prone points in the urban system. They respectively reflect the traffic positioning data density, the activity range of mobile devices, and the length of the monitored road section, which are of great significance for evaluating the traffic conditions of road sections, optimizing traffic management strategies, and improving the operation efficiency of the urban traffic system.

[0151] The traffic density calculation formula can accurately analyze the traffic density at different flood-prone points, providing an important reference for urban traffic management and flood emergency response. By analyzing and comparing the traffic density of road sections at different flood-prone points, the urban traffic conditions at different flood-prone points can be better understood, providing strong support for formulating scientific and reasonable traffic diversion plans and emergency measures.

[0152] The traffic saturation of different flood-prone points in the urban system is a fundamental indicator for measuring the traffic busyness and congestion status of different regions during a specific period. This indicator reflects the traffic status and congestion level of different flood-prone areas by quantifying the proportional relationship between the traffic flow and the road capacity on the road section within the predicted time. The greater the traffic saturation, the closer the traffic flow on this road section is to or exceeds its capacity, indicating that the traffic status is more congested.

[0153] Based on the relevant data information in the urban multi-modal database and traffic operation characteristics, in the embodiment, an analysis function of traffic saturation for different flood-prone points in the urban system is established, and it satisfies the following relationship: Among them, The traffic saturation in different flood-prone areas of the urban system, represents the average value of traffic volume within a certain time interval in different flood-prone areas, represents the error coefficient of the fitting center point of the road traffic, represents the traffic capacity limit threshold of the urban system; The traffic saturation in different flood-prone areas of the urban system is a reference indicator for measuring the traffic busyness and congestion status of a specific flood-prone area within a period of time. The higher the traffic saturation, the closer the traffic flow in this flood-prone area is to or exceeds its capacity.

[0154] The average value of traffic volume within a certain time interval in different flood-prone areas refers to the average value of the traffic flow passing through a certain flood-prone area during a specific period, which reflects the traffic busyness of this flood-prone area during this period. The greater the above-mentioned average traffic volume, the greater the traffic flow in this area, and the corresponding congestion level and traffic jam risk are also higher.

[0155] The error coefficient of the fitting center point of the road traffic reflects the deviation degree between the traffic volume data and the fitting center point or expected value. The smaller the error coefficient, the closer the actual traffic volume is to the expected value, and the more consistent the prediction of the traffic flow is with the actual situation. The above error coefficient is affected by various factors, such as the weather, traffic accidents, road construction, etc. in different flood-prone areas.

[0156] The traffic capacity limit threshold of the urban system refers to the maximum traffic flow that the urban traffic system can bear. When the actual traffic volume exceeds the above threshold, traffic congestion will occur. The traffic capacity limit threshold is affected by various factors such as the urban road layout, traffic facility conditions, and traffic management policies.

[0157] The saturation analysis function comprehensively considers the average traffic volume, error coefficient, and traffic capacity limit threshold to analyze the traffic saturation of different flood-prone points in the urban system, which helps traffic managers quickly and accurately understand the traffic conditions in different flood-prone areas and provides a reference basis for formulating traffic diversion plans and optimizing traffic management policies. At the same time, by comparing the traffic saturations of different flood-prone points, traffic bottleneck problems can also be discovered, providing strong support for the improvement and optimization of urban traffic.

[0158] In the urban system, by combining the specific road section names and driving direction information of different flood-prone points, the driving data of vehicles passing through different flood-prone area sections can be comprehensively analyzed, and then the average driving speed of vehicles on the road sections in different flood-prone areas can be determined. The average driving speed index is an important parameter for measuring the traffic efficiency and traffic fluency of road sections, and plays a crucial role in evaluating traffic conditions, predicting traffic congestion, and optimizing traffic management strategies.

[0159] In this embodiment, a traffic driving speed analysis function is established based on the road section information and vehicle driving data of different flood-prone points, and satisfies the following relationship: Among them, represents the traffic driving speed in different flood-prone areas of the urban system, represents the traffic situation index, represents the driving distance of vehicles in different flood-prone areas of the urban system, represents the time measurement error coefficient, represents the average time for vehicles to drive out of different flood-prone areas, The average time for vehicles to drive into different flood-prone areas.

[0160] The traffic driving speed in different flood-prone areas of the urban system is an index to measure the driving speed of vehicles in a specific flood-prone area. The higher the traffic driving speed, the higher the traffic efficiency and the better the traffic fluency in this area.

[0161] The traffic situation index is a comprehensive index that can reflect the overall traffic situation, which takes into account various factors such as traffic flow, congestion degree, accident incidence rate, etc. The higher the above traffic situation index, the more complex the traffic situation or the higher the congestion degree, and the greater the impact on the vehicle driving speed may be.

[0162] The above traffic situation index satisfies the following relationship; Among them, represents the traffic situation index, represents the number of different flood-prone points divided in the urban system, Represents the attenuation coefficient of the traffic situation model. Represents the fitting coefficient of the traffic situation model.

[0163] The traffic situation index can serve as an important reference for traffic management departments, urban planners, and the public to understand the urban traffic conditions, and contribute to formulating scientific traffic management strategies and travel plans.

[0164] The number of different flood - prone points divided in the urban system refers to the number of different flood - prone points in the urban flood - prone point detection module. It is used to evaluate the flood risk of different regions. By dividing different flood - prone points, the urban flood risk can be evaluated more accurately and provide basic data for subsequent traffic situation analysis.

[0165] The attenuation coefficient of the traffic situation model can describe the quantity of the attenuation rate of energy or signal intensity with the propagation distance or time, and further represent the attenuation degree of the traffic situation with time or space. It reflects the dynamic change characteristics of traffic conditions with time and space.

[0166] The fitting coefficient of the traffic situation model is mainly a parameter used to describe the fitting degree between the mathematical model and the actual data. It reflects the closeness between the model prediction result and the actual traffic conditions. The magnitude of the above - mentioned fitting coefficient can evaluate the accuracy and reliability of the traffic situation model. A higher fitting coefficient means that the model can better predict the actual traffic conditions.

[0167] The driving distance of vehicles in different flood - prone point areas in the urban system refers to the total distance that vehicles travel in a specific flood - prone point area. The longer the above - mentioned driving distance, the more time the vehicle may spend in this area. However, the driving speed is also affected by other factors, such as traffic congestion and road conditions in different flood - prone point areas.

[0168] The time measurement error coefficient reflects the inaccuracy or deviation in the time measurement process. The larger the time measurement error coefficient, the greater the difference between the actual time and the measured time, which in turn affects the calculation and analysis of the vehicle driving speed.

[0169] The average time for vehicles to leave different flood - prone point areas refers to the average time for vehicles to leave a specific flood - prone point area. The longer the above - mentioned leaving time means that there is traffic congestion or poor road conditions in this area, resulting in a slowdown in the vehicle driving speed.

[0170] The average time for vehicles to enter different flood - prone point areas refers to the average time for vehicles to enter a specific flood - prone point area. The length of the above - mentioned entering time is affected by various factors such as traffic flow, signal control, and road layout in different flood - prone point areas.

[0171] By comprehensively considering the traffic situation index, the driving distance of vehicles, the time measurement error coefficient, and the average time for vehicles to enter and exit waterlogging-prone areas, the traffic driving speed in different waterlogging-prone areas of the urban system is calculated, which helps to understand the traffic flow conditions in waterlogging-prone areas and evaluate the differences in traffic efficiency between different sections. At the same time, by comparing the traffic driving speeds in different waterlogging-prone areas, potential traffic bottlenecks and congestion points can be discovered, providing support for traffic decision-making, improvement, and optimization in different waterlogging-prone areas of the urban system.

[0172] Finally, based on the above urban waterlogging and traffic risk assessment mechanism and the urban multi-modal database, the traffic analysis situation of different waterlogging-prone points in the urban system is obtained.

[0173] Through the urban waterlogging and traffic risk assessment mechanism and based on the data resources in the urban multi-modal database, a comprehensive and in-depth analysis of the traffic conditions at different waterlogging-prone points in the urban system is carried out. Combining the analysis results of waterlogging-prone points, key indicators such as traffic density, traffic saturation, and traffic driving speed at each waterlogging-prone point are calculated. These indicators not only reflect the traffic busyness, congestion status, and traffic efficiency in waterlogging-prone areas but also provide an information basis for the operation of the urban traffic system. By comparing and analyzing the above traffic reference indicators, the traffic analysis situation of different waterlogging-prone points can be comprehensively understood, providing a scientific basis for formulating targeted traffic management strategies, optimizing the layout of traffic facilities, and improving the overall disaster resistance ability of the urban traffic system.

[0174] In the urban waterlogging point detection module, an urban waterlogging and traffic risk assessment mechanism is established and combined with the urban multi-modal database to obtain the waterlogging prediction results and traffic analysis situation of different waterlogging-prone points in the urban system.

[0175] The multi-modal database in the information processing and analysis module covers various types such as remote sensing images, meteorological data, hydrological data, and geographical information of the urban system, which can comprehensively reflect various influencing factors of urban waterlogging and traffic risks. The urban waterlogging point detection module accurately identifies the waterlogging risk points of the urban system based on the above multi-source information and evaluates the waterlogging prediction results and traffic analysis situation of each waterlogging risk point (waterlogging-prone point). The urban waterlogging point detection module of the system can accurately reflect the latest dynamics of urban waterlogging and traffic risks, which helps the urban safety early warning response module adjust prevention measures and early warning strategies based on the situations of different waterlogging-prone points, thereby improving the pertinence and timeliness of the urban waterlogging and traffic risk monitoring and early warning system with multi-modal data.

[0176] Furthermore, in this embodiment, the method for analyzing waterlogging and traffic conditions at different waterlogging-prone points is only an optional condition of the present invention. In one or some other embodiments, the method for analyzing waterlogging and traffic conditions at different waterlogging-prone points can be optimized and adjusted according to the prediction requirements of urban waterlogging and traffic risks and the actual operating conditions of the system. Since the characteristics of urban waterlogging and traffic risks vary from city to city, by optimizing and adjusting the analysis method, the system can more accurately adapt to the specific needs of different cities, thereby improving the practicality of the monitoring and early warning system.

[0177] S4. The urban safety early warning response module receives the waterlogging prediction results, traffic analysis situation, and urban multimodal database to conduct early warning assessment and intelligent regulation on different regions of the urban system. The specific implementation content is as follows: The urban waterlogging and traffic risk monitoring and early warning system based on multimodal data further includes: setting an evaluation parameter membership degree analysis model in the urban safety early warning response module; through the evaluation parameter membership degree analysis model, conducting membership degree analysis on the waterlogging prediction parameters in the waterlogging prediction results and the traffic analysis parameters in the traffic analysis situation, and the membership degree analysis results of different evaluation parameters can be obtained.

[0178] An evaluation parameter membership degree analysis model is set in the urban safety early warning response module to effectively evaluate the waterlogging and traffic risks of different waterlogging-prone points in the urban system.

[0179] Based on the urban waterlogging and traffic risk assessment mechanism constructed in the urban waterlogging point detection module, the evaluation parameters in the embodiment mainly include two aspects: waterlogging prediction parameters and traffic analysis parameters. According to the above implementation content, the waterlogging prediction parameters mainly include the infiltration rate, external water volume, and surface water accumulation volume, and these related parameters directly reflect the waterlogging situation of the waterlogging-prone points; the traffic analysis parameters mainly include traffic density, traffic saturation, and traffic driving speed, and the above parameters comprehensively reflect the traffic conditions and congestion degree around the waterlogging-prone points.

[0180] In order to quantify the membership degree of each evaluation parameter to urban waterlogging and traffic risks, a membership degree matrix is constructed in the embodiment based on the historical characteristics of urban waterlogging and traffic risks and the waterlogging prediction results and traffic analysis situation of urban waterlogging-prone points. 。

[0181] The embodiment clarifies the dimension and element arrangement of the membership degree matrix (original index matrix) which contains multiple rows and columns, covering the index values of all evaluation parameters, and the specific composition form is as follows: Among them, represents the original index matrix, represents different evaluation parameters of the original evaluation parameter matrix, Indicates different vulnerable waterlogging points in the urban system. Indicates the membership degree of the i-th evaluation parameter to the j-th urban waterlogging and traffic risk evaluation level.

[0182] The membership degree of different traffic evaluation indicators to urban waterlogging and traffic refers to the elements in the original index matrix where Indicates the membership degree of the i-th evaluation parameter to the j-th urban waterlogging and traffic risk evaluation level, and further indicates the influence degree of different evaluation parameters on the urban waterlogging and traffic risk evaluation level. In this embodiment, the membership degree value is between 0 and 1, and the larger the value, the higher the membership degree between the two.

[0183] The rows in the original index matrix represent different evaluation parameters (infiltration rate, external water volume, surface water accumulation, traffic density, traffic saturation, traffic driving speed), and the columns represent different evaluation objects (different vulnerable waterlogging points in the urban system). Each element in the original index matrix is the original evaluation data, which can be used to describe the performance or state of a specific evaluation object (different vulnerable waterlogging points) under specific evaluation parameters (infiltration rate, external water volume, surface water accumulation, traffic density, traffic saturation, traffic driving speed).

[0184] In this embodiment, the evaluation indicators of different vulnerable waterlogging points in the urban system include six evaluation dimensions: infiltration rate, external water volume, surface water accumulation, traffic density, traffic saturation, and traffic driving speed. At this time .

[0185] The vulnerable waterlogging points in the urban system can be used to evaluate the waterlogging and traffic risks at the intersections of 12 urban vulnerable waterlogging points. At this time .

[0186] Considering the differences in the importance and influence of different evaluation parameters in the urban waterlogging and traffic risk evaluation level, a weight vector is introduced to reflect the relative importance and influence degree of each evaluation parameter.

[0187] Combining the membership degree matrix and the weight vector The final evaluation results of each evaluation parameter can be calculated, that is, the influence degree and action relationship of different risk evaluation parameters on urban waterlogging and traffic risks can be obtained, which helps to comprehensively reflect the waterlogging and traffic risk conditions of different vulnerable waterlogging points in the urban system and provide a basis for disaster prevention and early warning response.

[0188] The above evaluation parameter membership degree analysis model satisfies the following relationship: Among them, Indicates the evaluation results of different risk evaluation parameters after weighting. represents the weight vector, represents the membership degrees of different traffic evaluation indicators to urban waterlogging and traffic, represents the original index matrix.

[0189] The evaluation results of different risk evaluation parameters after weighting refer to the weighting process of the membership degrees of each evaluation parameter, which is conducive to comprehensively analyzing the waterlogging and traffic risk levels of different waterlogging-prone points in the city.

[0190] The weight vector is a vector whose elements represent the weights of each evaluation parameter. The weight reflects the importance or influence of different evaluation parameters in the overall risk evaluation of waterlogging and traffic. Among them, some evaluation parameters have a greater impact on waterlogging risk or traffic risk, so higher weight coefficients need to be assigned. The elements of the above weight vector can be determined by expert scoring, historical data analysis or statistical methods.

[0191] In summary, the evaluation parameter membership degree analysis model can quantify the membership degrees of different evaluation parameters to the risk evaluation level in the urban waterlogging and traffic risk monitoring and early warning system of multi-modal data, and conduct a comprehensive evaluation in combination with the weight vector, providing a more accurate reference basis for the monitoring and early warning of urban waterlogging and traffic risks. Based on this, it helps to improve the accuracy and timeliness of the prediction results and early warning results of urban waterlogging and traffic risks, and provides technical support for urban intelligent management.

[0192] Finally, the urban safety early warning response module combines the membership degree analysis results, waterlogging prediction results, traffic analysis situation and urban multi-modal database to conduct early warning evaluation on different regions of the urban system, and obtains the urban waterlogging and traffic prediction results of different regions in the urban system; thus, the urban safety early warning response module realizes the safety early warning and intelligent regulation of the urban system based on the urban waterlogging and traffic prediction results.

[0193] The urban safety early warning response module can receive and integrate the membership degree analysis results in a timely manner, and at the same time combine the multi-dimensional data of waterlogging prediction (key indicators such as infiltration rate, external water volume and surface water accumulation volume), as well as the traffic analysis situation (covering traffic density, traffic saturation and traffic speed parameters), and make full use of the information resources of the urban multi-modal database.

[0194] The urban safety early warning response module integrates and comprehensively analyzes the above-mentioned multivariate analysis results and heterogeneous data, and evaluates and warns against waterlogging and traffic risks in different regions of the urban system. During the monitoring and warning process, the module not only pays attention to the changes in various evaluation parameters, but also needs to analyze the interaction and influence relationship between different risk evaluation parameters, as well as the potential threats and impacts of different risk evaluation parameters on the overall safety of the urban system. Through algorithm models and intelligent analysis technologies, the module can accurately predict the possible waterlogging and traffic congestion conditions in different regions of the urban system in a future period of time, and generate targeted urban waterlogging and traffic prediction results.

[0195] Based on the risk prediction results of different waterlogging-prone areas, the urban safety early warning response module can quickly activate the early warning mechanism, timely release early warning information to relevant departments and the public, and provide strong support for the safety management and emergency response of the urban system. At the same time, the module can also intelligently adjust the operation strategy of the urban system according to the prediction results, including but not limited to adjusting the traffic signal control mode, controlling the operation state of the drainage system, and formulating waterlogging prevention and control measures, etc., to achieve the safety early warning and intelligent regulation of the urban system.

[0196] In an optional embodiment, based on the waterlogging and traffic risk prediction results of waterlogging-prone areas, the urban safety early warning response module can quickly activate the early warning mechanism and timely release early warning information to relevant departments and the public. The specific content is as follows: Multi-channel release of early warning information: The system of this embodiment uses multiple channels such as television, radio, mobile APP, and social media to release early warning information in a timely and extensive manner to ensure the coverage rate of early warning information. At the same time, an information sharing mechanism is established with departments such as meteorology and hydrology to ensure the accuracy and timeliness of early warning information.

[0197] Hierarchical early warning system: The early warning information of the urban system is divided into different levels (blue, yellow, orange, red, etc.). Different early warning colors correspond to different risk levels and emergency response measures. Furthermore, for early warning information at different levels, emergency plans are formulated to clarify the responsibilities and tasks of each department.

[0198] Update and cancellation of early warning information: According to the changes in real-time monitoring data and risk prediction results, the early warning information is updated in a timely manner to ensure the timeliness of urban monitoring and early warning information. When the risk is lifted, the early warning cancellation signal is released in a timely manner to restore the normal order of the urban system.

[0199] The measures for preventing and controlling urban waterlogging include: strengthening the construction and management of drainage pipe networks, regularly maintaining and inspecting the drainage pipe networks to ensure unobstructed drainage, upgrading the drainage pipe networks at easily waterlogged points in the city to improve the drainage capacity, and setting up emergency drainage equipment such as mobile pump trucks. At the same time, sponge cities can be built. Through measures such as green space areas, rain gardens, and permeable pavements, the rainwater absorption and retention performance of the urban system can be improved. It is also necessary to strengthen the monitoring and inspection of easily waterlogged areas, set up water level monitoring stations, video monitoring and other equipment in easily waterlogged areas, and monitor the water level changes in real time, which helps to promptly discover and handle problems such as poor drainage.

[0200] The traffic prevention and control measures include: optimizing traffic signal control, dynamically adjusting the timing plan of traffic signals according to real-time traffic flow and urban waterlogging prediction results to relieve traffic congestion, setting up temporary traffic signals or traffic control measures in easily waterlogged areas to guide vehicles to detour. At the same time, strengthen traffic information dissemination. Through various channels such as traffic radio, electronic displays, and mobile phone APPs, promptly release traffic conditions information to guide the public to travel reasonably. Furthermore, warning signs and reminder information can be set up in easily waterlogged areas to remind drivers to pay attention to safety, and formulate a traffic emergency evacuation plan for easily waterlogged areas, clarifying the evacuation routes and evacuation points. The urban safety early warning response module provides strong support for the safety management and emergency response of the urban system by activating the early warning mechanism, releasing early warning information, and implementing urban waterlogging and traffic prevention and control measures.

[0201] In summary, the urban safety early warning response module can make full use of multi-modal data resources, combine prediction analysis technology and intelligent regulation means, provide a comprehensive, accurate and efficient solution for the monitoring and early warning of urban waterlogging and traffic risks, and lay a foundation for building a safe, intelligent and resilient urban system.

[0202] In this embodiment, a monitoring and early warning system for urban waterlogging and traffic risks based on multi-modal data is constructed. The system consists of four core parts: a data collection and transmission module, an information processing and analysis module, an urban waterlogging point detection module, and an urban safety early warning response module.

[0203] First of all, the data collection and transmission module is responsible for collecting multi-source heterogeneous data in the city, forming a comprehensive set of urban multi-source heterogeneous data, which covers the monitoring data of all aspects of the urban system and provides an information basis for subsequent risk analysis and prediction.

[0204] Next, the information processing and analysis module receives and processes the multi-source heterogeneous data output by the data collection and transmission module. By using a multi-source heterogeneous data fusion model, the relevant data is integrated into an urban multi-modal database, providing a unified and standardized data basis for the risk assessment and early warning analysis of the system.

[0205] In the urban waterlogging point detection module, a perfect urban waterlogging and traffic risk assessment mechanism is constructed. This mechanism combines the data information in the urban multi-modal database, can accurately predict the waterlogging situation at different vulnerable waterlogging points in the urban system, and comprehensively analyze the surrounding traffic conditions, providing a reference basis for waterlogging and traffic risk early warning and intelligent control.

[0206] Finally, the urban safety early warning response module receives and comprehensively analyzes the relevant information output by other modules. Based on the waterlogging prediction results, traffic analysis situation, and urban multi-modal database, it conducts accurate early warning assessment on different regions of the urban system, and intelligently adjusts the operation strategy of the urban system according to the assessment results to achieve the safety early warning and intelligent control of the city.

[0207] Based on the operation mechanism construction and collaborative cooperation process of different component modules in the urban waterlogging and traffic risk monitoring and early warning system of this embodiment, please refer to Figure 2 .

[0208] In summary, the urban waterlogging and traffic risk monitoring and early warning system based on multi-modal data realizes the comprehensive, accurate, and real-time monitoring and early warning of urban waterlogging and traffic risks through the collaborative work of each module, providing a strong guarantee for the safety management and emergency response of the city.

[0209] Please refer to Figure 3 , in an alternative embodiment, the present invention also provides an urban waterlogging and traffic risk monitoring and early warning system based on multi-modal data. The above system includes a data acquisition and transmission module, an information processing and analysis module, an urban waterlogging point detection module, and an urban safety early warning response module. The above data acquisition and transmission module, information processing and analysis module, urban waterlogging point detection module, and urban safety early warning response module are interconnected to implement the specific steps of the relevant embodiments of the urban waterlogging and traffic risk monitoring and early warning system based on multi-modal data provided by the present invention. The urban waterlogging and traffic risk monitoring and early warning system based on multi-modal data of the present invention has a complete structure, is objective and stable, and improves the overall applicability and practical application ability of the present invention.

[0210] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope of the claims and the description of the present invention.

Claims

1. A multimodal data urban waterlogging and traffic risk monitoring and early warning system, characterized by: The multimodal data urban waterlogging and traffic risk monitoring and early warning system includes: a data acquisition and transmission module, an information processing and analysis module, an urban waterlogging point detection module and an urban safety early warning response module; Obtaining a multi-source heterogeneous urban data set through the data acquisition and transmission module; The information processing and analysis module receives and processes the urban multi-source heterogeneous data set, and processes the urban multi-source heterogeneous data set through a multi-source heterogeneous data fusion model to obtain an urban multimodal database; An urban waterlogging and traffic risk assessment mechanism is constructed in the urban waterlogging point detection module, and waterlogging prediction results and traffic analysis conditions of different waterlogging-prone points in the urban system are obtained based on the urban waterlogging and traffic risk assessment mechanism and the urban multimodal database; The city safety early warning response module receives the waterlogging prediction result, the traffic analysis situation and the city multimodal database to perform early warning assessment and intelligent regulation on different areas of the city system.

2. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 1 is characterized in that: The obtaining of a city multi-source heterogeneous data set by the data acquisition and transmission module comprises: An information perception layer, a preprocessing layer and a transmission layer are set in the data acquisition and transmission module; Obtaining initial city multi-source heterogeneous data through the information perception layer; Establishing a data similarity analysis function in the preprocessing layer, and using the data similarity analysis function in the preprocessing layer to process the initial city multi-source heterogeneous data to obtain a similarity analysis result of the initial city multi-source heterogeneous data; A data condition constraint function is established in the transmission layer. The transmission layer judges the initial urban multi-source heterogeneous data in combination with the data condition constraint function and the similarity analysis result, and performs system data transmission according to the condition judgment to obtain a set of urban multi-source heterogeneous data.

3. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 2 is characterized in that: The data similarity analysis function is established in the preprocessing layer, and the preprocessing layer processes the initial city multi-source heterogeneous data using the data similarity analysis function to obtain the similarity analysis result of the initial city multi-source heterogeneous data, including: Establish a data similarity analysis function based on the distribution and attribute characteristics of the initial city multi-source heterogeneous data; Using the data similarity analysis function to process the initial city multi-source heterogeneous data, to obtain a similarity analysis result of the initial city multi-source heterogeneous data; The data similarity analysis function satisfies the following relationship: in, Represents the similarity results of any two sets of data in urban multi-source heterogeneous data. Represents a set of data in urban multi-source heterogeneous data. Represents another set of data in the city's multi-source heterogeneous data. Represents the total number of feature categories of the original multi-source heterogeneous data, express The corresponding weight coefficient is, express The corresponding weight coefficient is, Representing multi-source heterogeneous data The attribute characteristics of Representing multi-source heterogeneous data Attribute characteristics.

4. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 2 is characterized in that: The data condition constraint function is established in the transmission layer, the transmission layer judges the initial urban multi-source heterogeneous data in combination with the data condition constraint function and the similarity analysis result, and the system data is transmitted according to the condition judgment to obtain the urban multi-source heterogeneous data set, which includes: The similarity threshold of urban multi-source heterogeneous data is set based on the historical data of the urban system and multi-source heterogeneous data; Establishing a data condition constraint function based on the similarity threshold and the similarity analysis result; The initial urban multi-source heterogeneous data are judged by the data condition constraint function and the similarity analysis result, and system data is transmitted according to the condition judgment to obtain the urban multi-source heterogeneous data set; The data condition constraint function satisfies the following relationship: in, Represents the constraint analysis results of any two sets of data in urban multi-source heterogeneous data. Represents the similarity threshold of multi-source heterogeneous data, Represents the similarity results of any two sets of data in urban multi-source heterogeneous data. Representing multi-source heterogeneous data Multi-source heterogeneous data Not relevant, Representing multi-source heterogeneous data Multi-source heterogeneous data Related.

5. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 1 is characterized in that: The information processing and analysis module receives and processes the urban multi-source heterogeneous data set, and processes the urban multi-source heterogeneous data set through a multi-source heterogeneous data fusion model to obtain an urban multimodal database including: Set up a multi-source heterogeneous data fusion model based on the actual operation status of the data acquisition and transmission module; Processing the urban multi-source heterogeneous data set based on a multi-source heterogeneous data fusion model to obtain an urban multimodal database; The multi-source heterogeneous data fusion model satisfies the following relationship: in, represents the multi-source heterogeneous data fusion model, Represents the parameters related to the fusion model, Represents the number of data groups of original multi-source heterogeneous data, Representing multi-source heterogeneous data The reference mean, Represents multi-source heterogeneous data in a city multi-source heterogeneous data set , Representing multi-source heterogeneous data The reference mean, Represents multi-source heterogeneous data in a city multi-source heterogeneous data set , represents the generalization parameter of the fusion model.

6. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 1 is characterized in that: The urban waterlogging and traffic risk assessment mechanism constructed in the urban waterlogging point detection module includes: Based on the urban waterlogging and traffic risk assessment mechanism, the urban detection images in the urban multimodal database are analyzed, and the phase value analysis results and spatiotemporal baseline thresholds of the urban detection images are obtained; The urban waterlogging and traffic risk assessment mechanism analyzes the waterlogging-prone points of the urban system in combination with the phase value analysis results and the spatiotemporal baseline threshold, and obtains the waterlogging-prone points analysis results of the urban system; The phase value analysis result satisfies the following relationship: in, Represents the phase value of the city detection image, Indicates time Deformation along the radar line of sight, Indicates time Deformation along the radar line of sight, Indicates the wavelength of the urban system detection radar.

7. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 6 is characterized in that: The urban waterlogging and traffic risk assessment mechanism is constructed in the urban waterlogging point detection module, and the waterlogging prediction results and traffic analysis of different waterlogging points in the urban system are obtained based on the urban waterlogging and traffic risk assessment mechanism and the urban multimodal database, including: Based on the urban waterlogging and traffic risk assessment mechanism and the flood-prone point analysis results, the infiltration rate, external water volume and surface water volume of different flood-prone points in the urban system are obtained; Obtaining waterlogging prediction results for different flood-prone points in the urban system based on the infiltration rate, the amount of external water and the amount of surface water accumulation; The infiltration rate satisfies the following relationship: in, represents the infiltration rate at different flood-prone points in the urban system at different times, represents the original initial infiltration rate of the urban system, represents the stable infiltration rate of different flood-prone points in the urban system, represents the average infiltration decline rate of the urban system, Indicates different moments; The amount of external water satisfies the following relationship: in, represents the amount of external infiltration water at different flood-prone points in the urban system, It represents the total water outflow at the end of the drainage network at different flood-prone points in the urban system. It represents the theoretical water quality index concentration of the urban system drainage network. Indicates the actual water quality index concentration of the urban system drainage network. It indicates the water quality index concentration of external infiltration water in the drainage network at different flood-prone points in the urban system. represents the infiltration rate at different flood-prone points in the urban system at different times, Indicates the start sampling time of water volume at different flood-prone points in the urban system, Indicates the start sampling time of water volume at different flood-prone points in the urban system; The surface water volume satisfies the following relationship: in, represents the amount of surface water at different flood-prone points in the urban system, represents the net rainfall intensity at different flood-prone points in the urban system, represents the catchment area of ​​different flood-prone points in the urban system, represents the width of the catchment area at different flood-prone points in the urban system, represents the actual water depth at different flood-prone points in the urban system, represents the ground water storage depth threshold of the urban system, represents the slope of the catchment area at different flood-prone points in the urban system, represents the Manning roughness coefficient.

8. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 6 is characterized in that: The urban waterlogging and traffic risk assessment mechanism is constructed in the urban waterlogging point detection module, and the waterlogging prediction results and traffic analysis of different waterlogging points in the urban system are obtained based on the urban waterlogging and traffic risk assessment mechanism and the urban multimodal database, including: Based on the urban waterlogging and traffic risk assessment mechanism and the waterlogging point analysis results, the traffic density, traffic saturation and traffic speed of different waterlogging points in the urban system are obtained; Obtaining traffic analysis conditions of different flood-prone points in the urban system according to the traffic density, the traffic saturation and the traffic travel speed; The traffic density satisfies the following relationship: in, represents the density of traffic location data in different flood-prone areas in the urban system, It indicates the maximum distance between the mobile device and the base station at a certain time interval in different flood-prone areas. Indicates the length of monitored traffic sections at different flood-prone points in the urban system; The traffic saturation satisfies the following relationship: in, Traffic saturation in different flood-prone areas in the urban system, It represents the average traffic volume in different flood-prone areas within a certain time interval. Represents the error coefficient of the center point of road traffic fitting, represents the traffic capacity limit threshold of the urban system; The traffic speed satisfies the following relationship: in, represents the traffic speed in different flood-prone areas in the urban system, represents the traffic situation index, represents the vehicle travel distance in different flood-prone areas in the urban system, represents the time measurement error coefficient, It represents the average time for vehicles to leave different flood-prone areas. The average time it takes for vehicles to enter different flood-prone areas.

9. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 6, characterized in that: The multimodal data urban waterlogging and traffic risk monitoring and early warning system also includes: Setting up the evaluation parameter membership analysis model in the city safety early warning response module; Performing membership analysis on the waterlogging prediction parameters in the waterlogging prediction result and the traffic analysis parameters in the traffic analysis situation through the evaluation parameter membership analysis model to obtain membership analysis results of different evaluation parameters; The evaluation parameter membership analysis model satisfies the following relationship: in, represents the evaluation results of different risk assessment parameters after weighting, represents the weight vector, It represents the degree of subordination of different traffic evaluation indicators to urban waterlogging and traffic. represents the original indicator matrix.

10. The multimodal data urban waterlogging and traffic risk monitoring and early warning system according to claim 9, characterized in that: The city safety early warning response module receives the waterlogging prediction result, the traffic analysis situation and the city multimodal database to perform early warning assessment and intelligent regulation on different areas of the city system, including: The urban safety early warning response module combines the membership analysis results, the waterlogging prediction results, the traffic analysis situation and the urban multimodal database to perform early warning assessments on different areas of the urban system, and obtains urban waterlogging and traffic prediction results for different areas of the urban system; The city safety warning response module realizes safety warning and intelligent regulation of the city system based on the urban waterlogging and traffic prediction results.

Images