Method and system for acquiring influence degree of urban element on waterlogging disaster loss

Abstract

The application provides a method and system for obtaining the influence degree of urban elements on the loss of waterlogging disasters, wherein the method comprises the following steps: constructing a hierarchical Bayesian network structure at the city and block scales, and corresponding the structure in the mechanism ontology model of the influence of urban elements on waterlogging disasters to the hierarchical Bayesian network structure; using a parameter learning algorithm to perform parameter learning training on the hierarchical Bayesian network structure to obtain the conditional probability of each node; and performing sensitivity analysis on each node in the hierarchical Bayesian network structure according to the conditional probability to quantify the influence degree of urban elements on the loss of waterlogging disasters. The technical scheme provided by the application can quantify the influence degree of urban elements on the loss of waterlogging disasters, which can enable people to intuitively observe the influence degree of each urban element on the loss of waterlogging disasters, and can provide a quantitative evaluation basis when evaluating the loss degree of waterlogging disasters.

Description

Technical Field

[0001] This invention belongs to the technical field of research on the impact of urban elements on the losses caused by urban flooding disasters, and specifically relates to a method and system for obtaining the impact of urban elements on the losses caused by urban flooding disasters. Background Technology

[0002] Floods are a common natural disaster. Urban floods differ significantly from watershed floods in their characteristics, and urban flooding often causes greater losses. This is because while dikes built around cities can improve their flood resistance, the increased density of urban roads and buildings reduces the permeable surface area and permeability. Furthermore, due to historical reasons, many underground pipes and drainage systems cannot meet the current drainage and flood discharge needs of cities, prolonging the duration of floods and exacerbating their severity. The large population and highly concentrated number of buildings and equipment in cities mean that once a flood occurs, it can cause substantial economic losses and even casualties.

[0003] Relevant studies have found that the underlying surface conditions, pipeline network distribution, and building distribution in urban areas have a significant impact on runoff generation and distribution. Furthermore, the losses from urban flooding are positively correlated with the density of urban population and economic distribution. These characteristics can provide optimization directions for urban flooding risk management. However, as a complex mega-system coupling natural and human phenomena, cities are characterized by a high concentration of population and infrastructure resources, complex spatial structures, and rapid dynamic evolution of physical objects and social behaviors. Their organizational structure is the form and mode of interaction and relationship among the various components of the city. Therefore, the process and characteristics of urban flooding disasters are more complex, involving a wide range of factors and types. Simply selecting relevant urban components based on the physical mechanisms of urban flooding formation and analyzing their impact on flooding disasters cannot comprehensively compare the degree of influence between various factors, nor can it provide an effective reference for urban risk management decisions.

[0004] The composition of urban elements is complex and diverse. Researching their impact mechanisms on urban flooding requires the collection and processing of data on various elements and disaster losses. Big data-based theories and methods can effectively utilize multi-source heterogeneous data on urban elements and disaster losses to deeply analyze the degree of impact of urban elements on flooding. Addressing the problem of integrating and utilizing multi-source heterogeneous data on urban elements and disaster losses, ontology theory, based on semantic description, can integrate and manage data from various urban management departments related to urban components and disaster losses, effectively improving the accuracy of quantifying the degree of impact of urban elements on urban flooding.

[0005] Although significant progress has been made in the research on urban factors and their impact on flood disasters, providing some guidance and suggestions for urban flood disaster risk management, the research still cannot specifically reflect the degree of impact of urban factors on urban flooding disasters. Therefore, its guiding role in urban flood prevention and disaster relief is limited, and it cannot provide sufficient support for the decision-making process of prioritizing urban factors in the process of urban flood disaster risk management.

[0006] Therefore, the relationship between urban elements and the degree of damage caused by urban flooding disasters obtained from existing technologies cannot provide sufficient support for guiding urban flood prevention and disaster relief. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for obtaining the degree of impact of urban elements on urban flooding disaster losses, so as to at least solve the problem that the relationship between urban elements and the degree of impact of urban flooding disaster losses obtained in the prior art cannot provide sufficient support for guiding urban flood prevention and disaster relief.

[0008] To achieve the above objectives, on the one hand, the present invention provides a method for obtaining the degree of influence of urban elements on urban flooding disaster losses, comprising: constructing a hierarchical Bayesian network structure at the city and street scales, mapping the structures in the ontology model of the mechanism of urban elements' influence on urban flooding disasters to the hierarchical Bayesian network structure respectively; using a parameter learning algorithm to perform parameter learning training on the hierarchical Bayesian network structure to obtain the conditional probability of each node therein; and performing sensitivity analysis on each node in the hierarchical Bayesian network structure based on the conditional probability of each node to quantify the degree of influence of urban elements on urban flooding disaster losses.

[0009] According to one embodiment of the present invention, mapping the sets in the mechanism ontology model of the impact of urban elements on urban flooding disasters to the hierarchical Bayesian network structure includes: setting the nodes in the hierarchical Bayesian network structure according to the set of concepts in the mechanism ontology model; obtaining the corresponding causal relationships according to the semantic and temporal relationships of the concepts in the mechanism ontology model; and setting the edges of each node in the hierarchical Bayesian network structure according to the causal relationships.

[0010] Furthermore, according to another embodiment of the present invention, the conceptual semantic relationship includes whole / part relationship, positive influence relationship, negative influence relationship, and relationship with both positive and negative influence, and the temporal relationship includes earlier than and later than.

[0011] According to another embodiment of the present invention, the parameter learning algorithm is an expectation maximization algorithm, comprising: calculating the expectation of the log-likelihood function of the nodes in the hierarchical Bayesian network structure, and calculating parameters to maximize the full likelihood expectation.

[0012] According to another embodiment of the present invention, the sensitivity analysis of each node in the hierarchical Bayesian network structure based on the conditional probability of each node includes: obtaining the sensitivity of the urban flooding loss index node to the urban element index node, and obtaining the degree of influence of the urban element index node on the urban flooding disaster loss based on the sensitivity.

[0013] On the other hand, the present invention also provides a system for obtaining the degree of impact of urban elements on urban flooding disaster losses, including a processor and a memory. The memory stores computer program instructions for execution on the processor. When the processor executes the computer program instructions, it implements the following method for obtaining the degree of impact of urban elements on urban flooding disaster losses: constructing a hierarchical Bayesian network structure at the city and street scales, mapping the structures in the ontology model of the mechanism of urban elements' impact on urban flooding disasters to the hierarchical Bayesian network structure; using a parameter learning algorithm to train the hierarchical Bayesian network structure to obtain the conditional probability of each node; and performing sensitivity analysis on the conditional probabilities of each node in the hierarchical Bayesian network structure to quantify the degree of impact of urban elements on urban flooding disaster losses.

[0014] According to one embodiment of the present invention, mapping the sets in the mechanism ontology model of the impact of urban elements on urban flooding disasters to the hierarchical Bayesian network structure includes: setting the nodes in the hierarchical Bayesian network structure according to the set of concepts in the mechanism ontology model; obtaining the corresponding causal relationships according to the semantic and temporal relationships of the concepts in the mechanism ontology model; and setting the edges of each node in the hierarchical Bayesian network structure according to the causal relationships.

[0015] Furthermore, according to another embodiment of the present invention, the conceptual semantic relationship includes whole / part relationship, positive influence relationship, negative influence relationship, and relationship with both positive and negative influence, and the temporal relationship includes earlier than and later than.

[0016] According to another embodiment of the present invention, the parameter learning algorithm is an expectation maximization algorithm, comprising: calculating the expectation of the log-likelihood function of the nodes in the hierarchical Bayesian network structure, and calculating parameters to maximize the full likelihood expectation.

[0017] According to another embodiment of the present invention, the sensitivity analysis of each node in the hierarchical Bayesian network structure based on the conditional probability of each node includes: obtaining the sensitivity of the urban flooding loss index node to the urban element index node, and obtaining the degree of influence of the urban element index node on the urban flooding disaster loss based on the sensitivity.

[0018] The beneficial effects of this invention are as follows: The technical solution provided by this invention can quantify the impact of urban elements on urban flooding disaster losses. On the one hand, it allows people to intuitively observe the impact of various urban elements on urban flooding disaster losses; on the other hand, it provides a quantitative evaluation basis when assessing the extent of urban flooding disaster losses. Furthermore, the technical solution of this invention, employing ontology theory, can effectively establish the relationship between knowledge and data. It is an effective method for integrating multi-source heterogeneous data of urban elements and establishing the relationship between urban elements and urban flooding disasters. Moreover, the hierarchical Bayesian network structure at the city and street scales can be effectively integrated with the ontology model, making it an effective method for quantifying the impact of urban elements on urban flooding disasters using multi-source heterogeneous data. Attached Figure Description

[0019] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the invention are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein:

[0020] Figure 1 This is a flowchart of a method for obtaining the degree of impact of urban elements on urban flooding disaster losses according to an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of a system for obtaining the degree of impact of urban elements on the losses caused by urban flooding according to an embodiment of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Those skilled in the art should understand that the embodiments described below are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0023] Please see Figure 1 , Figure 1 The method described in this application for obtaining the degree of impact of urban elements on urban flooding disaster losses is shown below. This method utilizes a Bayesian network to quantify the degree of impact of urban elements on urban flooding disaster losses, thereby obtaining the degree of impact of urban elements on urban flooding disaster losses. Figure 1 The illustrated process provides a detailed description of the method used in this application to obtain the degree of impact of urban elements on the losses caused by urban flooding disasters.

[0024] like Figure 1 As shown, the method of this application for obtaining the degree of impact of urban elements on urban flooding disaster losses includes:

[0025] In step S1, a hierarchical Bayesian network structure is constructed at the city and street scales. In this embodiment, when constructing the hierarchical Bayesian network structure at the city and street scales, a mechanistic ontology model of the impact of urban elements on urban flooding disasters can be obtained first. Then, the structures in this mechanistic ontology model are mapped to the established hierarchical Bayesian network structure. During the mapping process, urban flooding disaster losses are used as the root node of the hierarchical Bayesian network structure at the city and street scales, and urban factors related to urban flooding disasters are used as the leaf nodes. Furthermore, parent and child nodes in the hierarchical Bayesian network structure at the city and street scales are set according to causal and temporal relationships. The hierarchical Bayesian network structure constructed through this embodiment has Bayesian nodes at the city scale, Bayesian network structure nodes at the street scale, and the same Bayesian nodes at both the city and street scales.

[0026] In this embodiment, it is set This represents the set of nodes in a Bayesian network structure at the city-level. This represents the set of nodes in a Bayesian network structure at the street scale. Let the set of identical Bayesian network structure nodes represent the city and street scales. Then the resulting Bayesian network is:

[0027] (1).

[0028] Furthermore, the mechanism ontology model of the impact of urban elements on urban flooding disasters in step S1 not only integrates the data integration process with the impact relationship of urban elements on urban flooding disaster losses, achieving efficient utilization of data resources, but also facilitates the quantification of the impact relationship of urban elements on urban flooding disaster losses. In this embodiment, the method for constructing the mechanism ontology model of the impact of urban elements on urban flooding disasters includes:

[0029] (1) Demand analysis: Based on the goal of establishing the impact relationship of urban elements on waterlogging disaster losses and data integration management, the construction requirement of the ontology model of the impact mechanism of urban elements on waterlogging disasters is determined. This is based on the integration management of urban elements and waterlogging disaster loss index data, and incorporates the impact relationship of urban elements on waterlogging disaster losses and its data relationship.

[0030] (2) Data analysis and collection: Collect data and related information on urban elements and waterlogging disaster loss indicators at the city and street scales.

[0031] (3) Extracting information, extracting concepts and attributes: Based on the identification results and indicator system of urban elements associated with urban waterlogging disasters, analyze the concepts and data attributes (such as time, space, etc.) of urban elements and waterlogging disaster losses, and define the semantic concept system and concept attributes of the ontology model;

[0032] (4) Determine the classification and relationship between concepts: Based on the composition of urban elements and the impact mechanism of urban elements on waterlogging disaster losses, construct the hierarchical structure of ontology model concepts; establish the impact relationship of urban elements on waterlogging disaster losses at the city and street scales, and determine the semantic relationship of ontology model; determine the data relationship between urban elements and waterlogging disaster losses, and determine the temporal and spatial relationships of ontology model data.

[0033] (5) Ontology Structure Verification: The ontology construction criteria are used to verify whether the ontology of the impact mechanism of urban elements on urban flooding disasters meets the standards, including whether the concepts are clear, whether they are consistent, whether they are scalable, whether they meet the principle of minimum coding preference, and the minimal ontology convention. If they do not meet the standards, return to the concept and attribute extraction stage until the requirements are met.

[0034] (6) Ontology model construction: Use model building software (such as Protégé) to construct an ontology model of the impact mechanism of urban elements on urban flooding disaster.

[0035] (7) Ontology model saving: Archive the ontology for easy reuse later.

[0036] In step S2, a parameter learning algorithm is used to train the constructed hierarchical Bayesian network structure at the city and street scales to obtain the conditional probabilities of each node. Since the hierarchical Bayesian network structure at the city and street scales constructed in step S1 only represents causal relationships among its nodes, but it has not yet been trained and its parameters are not yet fixed, a parameter learning algorithm is used in step S2 to train its parameters, solidify them, and thus obtain the conditional probabilities of each node.

[0037] In step S3, sensitivity analysis is performed on the conditional probabilities of each node in the hierarchical Bayesian network structure at the city and street scales to quantify the impact of urban elements on urban flooding losses. Sensitivity analysis of the hierarchical Bayesian network structure at the city and street scales refers to analyzing the impact of changes in the parameters of the network structure on its output. In this embodiment, the impact of urban elements on urban flooding losses can be quantified by the sensitivity of the root node to its parent node in the hierarchical Bayesian network structure at the city and street scales. A higher sensitivity of a particular urban element to urban flooding losses indicates a greater impact. In the sensitivity analysis of the internal network model, the sensitivity of variables can be quantified by calculating and determining a sensitivity function, thereby quantifying the impact of urban elements on urban flooding losses.

[0038] In summary, the technical solution provided in this application can quantify the impact of urban elements on urban flooding disaster losses. On the one hand, it allows people to intuitively observe the impact of various urban elements on urban flooding disaster losses; on the other hand, it provides a quantitative evaluation basis when assessing the extent of urban flooding disaster losses. Furthermore, the use of ontology theory can effectively establish the relationship between knowledge and data, and is an effective method for integrating multi-source heterogeneous data of urban elements and establishing the relationship between urban elements and urban flooding disasters. Moreover, the hierarchical Bayesian network structure at the city and street scales can be effectively integrated with the ontology model, making it an effective method for quantifying the impact of urban elements on urban flooding disasters using multi-source heterogeneous data.

[0039] The preceding text provides an overall overview of the method for obtaining the impact of urban elements on urban flooding disaster losses according to the present invention. The following text, in conjunction with specific application scenarios, describes the method of mapping the concept set in the mechanistic ontology model to the established hierarchical Bayesian network structure at the city and street scales, respectively. It is understood that the description of the method below is exemplary and not restrictive. Therefore, the description above of mapping the concept set in the mechanistic ontology model to the established hierarchical Bayesian network structure at the city and street scales also applies to the description of the method below.

[0040] In one embodiment, step S1 above, which maps the concept set in the mechanistic ontology model to the established hierarchical Bayesian network structure at the city and street scales, includes: setting the nodes in the hierarchical Bayesian network structure according to the concept set in the mechanistic ontology model; firstly, obtaining the causal relationships between nodes in the hierarchical Bayesian network structure at the city and street scales based on the semantic and temporal relationships of the concepts in the mechanistic ontology model; and then setting the edges of each node in the hierarchical Bayesian network structure at the city and street scales based on the causal relationships between the nodes. The aforementioned semantic relationships refer to the relationships between concepts in the ontology model, such as the causal relationship between a city element and a flood disaster; the aforementioned temporal relationships refer to the chronological order of events. When setting the nodes in the hierarchical Bayesian network structure at the city and street scales based on the causal relationships between the nodes, if one node is the cause of another node, then that node is the parent node of the other node. The configuration method in this embodiment allows the ontological model of the mechanism of urban elements' impact on urban flooding disasters to be mapped to a hierarchical Bayesian network structure at the city and street scales, so that the hierarchical Bayesian network structure at the city and street scales can represent the relationship between urban elements and urban flooding disaster losses.

[0041] Furthermore, in another embodiment of this invention, the conceptual semantic relationship system of the above-mentioned ontology model of the impact mechanism of urban elements on urban flooding disasters includes: whole / part relationship, positive impact relationship, negative impact relationship, and relationship with both positive and negative impacts. The established conceptual semantic relationship system constructs unidirectional relationships between urban elements and urban flooding disaster losses. This unidirectional relationship can be understood as a causal relationship where urban elements are the cause and urban flooding disaster losses are the effect. Therefore, the conceptual semantics of the ontology model of urban elements on urban flooding disasters can be converted into causal relationships of corresponding nodes in a Bayesian network. For example, the whole is the parent node of the part; if one node has a positive impact relationship with another node, then that node is the parent node of the other node; conversely, if one node has a negative impact relationship with another node, then that node is the child node of the other node. In addition, the time relationship of the index data of the urban elements on the impact mechanism ontology model of the above-mentioned impact mechanism includes "earlier" and "later". The causal relationship of the corresponding nodes in the hierarchical Bayesian network structure at the city and block scale is obtained by using "earlier" and "later". For example, if the time of one node is earlier than the time of another node, then the above-mentioned node is regarded as the parent node of the other node. Conversely, if the time of one node is later than the time of another node, then the above-mentioned node is regarded as the child node of the other node.

[0042] The parameter learning algorithm in step S2 above can be the maximum likelihood estimation algorithm, the gradient ascent algorithm, or the expectation maximization algorithm. The following text uses the expectation maximization algorithm as an example to introduce in detail the method of parameter learning for the hierarchical Bayesian network structure at the city and street scales. It should be understood that the following introduction of the method is exemplary and not restrictive.

[0043] In one embodiment, the method for parameter learning of the constructed hierarchical Bayesian network structure at the city and street scales using the expectation-maximization algorithm in step S2 above includes: calculating the expectation of the log-likelihood function of the nodes in the hierarchical Bayesian network structure at the city and street scales, and calculating the parameters that maximize the complete expectation of the above likelihood. In this embodiment, let the likelihood function of a node B sample in the hierarchical Bayesian network structure at the city and street scales be... Then the log-likelihood function of the likelihood function is: ,in Let D represent the set of parameters for node B, and let D represent the sample data for node B. Then, calculate the expected value of the log-likelihood function for node B. The formula used at that time is:

[0044] (2)

[0045] in , Let m be the k-th parameter in node B, where k and m are both positive integers.

[0046] In the aforementioned expectation-maximization algorithm, when calculating the parameters that maximize the full likelihood expectation, we first assume that there exists... This can make the following formula applicable to Established:

[0047] (3)

[0048] Where C is the set, the parameter value sequence is generated by iterating continuously according to the following formula (4). , and thus Until convergence, thus obtaining the space of values ​​for the parameters. Maximum likelihood function :

[0049] (4)

[0050] Through the above steps, we can obtain the conditional probability of the city area and the level Bayesian node connected to the given sample D as follows:

[0051] (5).

[0052] Based on the above method, the probability of child nodes under different states of the parent node can be obtained, and the probability of different disaster loss levels occurring under different urban factor indicators can be used to complete the parameter learning and training of hierarchical Bayesian network structures at the city scale and street scale.

[0053] The previous section provided a detailed introduction to the method of using the expectation-maximization algorithm to learn the parameters of the constructed hierarchical Bayesian network structure at the city and street scales. The following section, in conjunction with a specific application scenario, provides a detailed explanation of the method of performing sensitivity analysis on each node in the hierarchical Bayesian network structure at the city and street scales based on the conditional probabilities of each node.

[0054] In one embodiment, a method for sensitivity analysis of nodes in a hierarchical Bayesian network structure at the city and street scales based on their conditional probabilities includes: first, obtaining the sensitivity of node parameters; and then, based on the sensitivity of node parameters, determining the degree of influence between nodes. In this embodiment, the hierarchical Bayesian network structure at the city and street scales is configured with nodes representing urban flooding disaster loss indicators. At the urban element indicator node The probability of taking the value under the condition is Where 'a' is the node representing the waterlogging disaster loss index. The value of is , where b is the node of the urban element indicator. The value of the city element indicator node The parameters are Where c represents any city element indicator node. The value of π is taken as the node of the urban element indicator. The sensitivity of parameter ψ is determined by its combined value with that of its parent node:

[0055] (6)

[0056] in for and The function, and

[0057] (7)

[0058] In the above formula, m1, m2, m3, and m4 are fixed coefficients.

[0059] As the parameter ψ changes, the urban element indicator nodes Other values ​​constitute the parameters The value of is changed accordingly to ensure that the sum of the probabilities of all possible values ​​is 1.

[0060] Set up urban element indicator nodes Relative to the index node of waterlogging disaster loss The degree of influence is That is, the indicator node of waterlogging disaster loss. Urban element indicator nodes The sensitivity is ,but:

[0061] (8)

[0062] In the formula: r and s represent the nodes of the urban flooding disaster loss index, respectively. and urban element indicator nodes The number of possible values, Indicator nodes for waterlogging disaster losses The i-th parameter and city element indicator node The j-th parameter, Indicator nodes representing losses from urban flooding disasters Urban element indicator nodes The sensitivity, with values ​​ranging from [0,1], The larger the value, the more important it is for urban element indicator nodes. Indicators of losses from urban flooding disasters The greater the degree of influence.

[0063] According to another aspect of this application, this application also provides a system for obtaining the degree of impact of urban elements on urban flooding disaster losses, such as... Figure 2As shown, the system includes a processor, an inserter, a communication interface, and a communication bus. The processor, inserter, and communication interface communicate with each other via the communication bus. The processor provides computing and control capabilities. The inserter includes a non-volatile insert medium and an internal inserter. The non-volatile insert medium contains an operating system and computer program instructions. The internal inserter provides an environment for the operation of the operating system and computer program instructions in the non-volatile insert medium. The communication interface of the above device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through WIFI, carrier networks, NFC (Near Field Communication), or other technologies. The system provided in this embodiment for obtaining the degree of impact of urban elements on urban flooding disaster losses has an inserter for inserting computer program instructions, which cause the processor to execute the above-described method for obtaining the degree of impact of urban elements on urban flooding disaster losses and its various embodiments.

[0064] Based on the above description in this specification, those skilled in the art will also understand that terms used, such as "front," "rear," "left," "right," "top," and "bottom," which indicate orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings of this specification. They are only for the purpose of facilitating the explanation of the present invention and simplifying the description, and do not explicitly or implicitly suggest that the device or element involved must have the specific orientation, or be constructed and operated in a specific orientation. Therefore, the above-mentioned orientation or positional relationship terms should not be understood or interpreted as limitations on the present invention.

[0065] Furthermore, the terms "first" or "second," etc., used in this specification to refer to numbers or ordinal numbers are for descriptive purposes only and should not be construed as indicating relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this specification, "a plurality of" means at least two, such as two, three, or more, unless otherwise expressly and specifically defined.

[0066] While various embodiments of the invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of the invention. The appended claims are intended to define the scope of the invention and therefore cover any modular compositions, equivalents, or alternatives within the scope of these claims.

Claims

1. A method for obtaining the degree of impact of urban elements on urban flooding disaster losses, characterized in that, include: A hierarchical Bayesian network structure is constructed at the city and street scales, and the structures in the ontology model of the mechanism of urban elements' impact on urban flooding disasters are respectively mapped to the hierarchical Bayesian network structure. The hierarchical Bayesian network structure is trained using a parameter learning algorithm to obtain the conditional probability of each node. Sensitivity analysis was performed on each node in the hierarchical Bayesian network structure to quantify the impact of urban elements on urban flooding disaster losses. The sensitivity analysis includes: setting up nodes for urban flooding disaster loss indicators in a hierarchical Bayesian network structure at the city and street scales. At the urban element indicator node The probability of taking the value under the condition is Where 'a' is the node representing the waterlogging disaster loss index. The value of is , where b is the node of the urban element indicator. The value of the city element indicator node The parameters are Where c represents any city element indicator node. The value of π is taken as the node of the urban element indicator. The sensitivity of parameter ψ is determined by its combined value with that of its parent node: （6） in for and The function, and （7） In the above formula, m1, m2, m3, and m4 are fixed coefficients; Quantifying the impact of urban elements on urban flooding disaster losses includes: obtaining the impact of urban element indicator nodes on urban flooding disaster losses based on the aforementioned sensitivity; and, as the parameter ψ changes, the urban element indicator nodes... Other values ​​constitute the parameters The value of is changed accordingly to ensure that the sum of the probabilities of all possible values ​​is 1; Set up urban element indicator nodes Relative to the index node of waterlogging disaster loss The degree of influence is That is, the indicator node of waterlogging disaster loss. Urban element indicator nodes The sensitivity is ,but: （8） In the formula: r and s represent the nodes of the urban flooding disaster loss index, respectively. and urban element indicator nodes The number of possible values, Indicator nodes for waterlogging disaster losses The i-th parameter and city element indicator node The j-th parameter, Indicator nodes representing losses from urban flooding disasters Urban element indicator nodes The sensitivity, with values ​​ranging from [0,1], The larger the value, the more important it is for urban element indicator nodes. Indicators of losses from urban flooding disasters The greater the degree of influence.

2. The method for obtaining the degree of impact of urban elements on urban flooding disaster losses according to claim 1, characterized in that, Mapping the sets in the ontology model of the mechanism by which urban elements affect urban flooding disasters to the hierarchical Bayesian network structure includes: Based on the set of concepts in the aforementioned mechanism ontology model, the nodes in the hierarchical Bayesian network structure are configured; Based on the conceptual semantic relationships and temporal relationships in the aforementioned mechanism ontology model, the corresponding causal relationships are obtained; The edges of each node in the hierarchical Bayesian network structure are set according to the causal relationship.

3. The method for obtaining the degree of impact of urban elements on urban flooding disaster losses according to claim 2, characterized in that, The conceptual semantic relationships include whole / part relationships, positive influence relationships, negative influence relationships, and relationships with both positive and negative influences. The temporal relationships include earlier than and later than.

4. The method for obtaining the degree of impact of urban elements on urban flooding disaster losses according to claim 1, characterized in that, The parameter learning algorithm is an expectation-maximization algorithm, which includes: calculating the expectation of the log-likelihood function of the nodes in the hierarchical Bayesian network structure, and calculating parameters to maximize the full likelihood expectation.

5. A system for acquiring the degree of impact of urban elements on urban flooding disaster losses, characterized in that, The system includes a processor and a memory, the memory storing computer program instructions for execution on the processor, wherein when the processor executes the computer program instructions, it implements a method for obtaining the degree of impact of urban elements on urban flooding losses: A hierarchical Bayesian network structure is constructed at the city and street scales, and the structures in the ontology model of the mechanism of urban elements' impact on urban flooding disasters are respectively mapped to the hierarchical Bayesian network structure. The hierarchical Bayesian network structure is trained using a parameter learning algorithm to obtain the conditional probability of each node. Sensitivity analysis was performed on each node in the hierarchical Bayesian network structure to quantify the impact of urban elements on urban flooding disaster losses. The sensitivity analysis based on the conditional probabilities of each node in the hierarchical Bayesian network structure includes: first, obtaining the sensitivity of node parameters; then, based on the sensitivity of node parameters, determining the degree of influence between nodes. Let's assume that in the hierarchical Bayesian network structure at the city and street scales, the nodes representing the urban flooding disaster loss index... At the urban element indicator node The probability of taking the value under the condition is Where 'a' is the node representing the waterlogging disaster loss index. The value of is , where b is the node of the urban element indicator. The value of the city element indicator node The parameters are Where c represents any city element indicator node. The value of π is taken as the node of the urban element indicator. The sensitivity of parameter ψ is determined by its combined value with that of its parent node: （6） in for and The function, and （7） In the above formula, m1, m2, m3, and m4 are fixed coefficients; Then, based on the sensitivity, the degree of influence of urban element indicator nodes on the losses from urban flooding disasters is obtained. As the parameter ψ changes, the urban element indicator nodes... Other values ​​constitute the parameters The value of is changed accordingly to ensure that the sum of the probabilities of all possible values ​​is 1; Set up urban element indicator nodes Relative to the index node of waterlogging disaster loss The degree of influence is That is, the indicator node of waterlogging disaster loss. Urban element indicator nodes The sensitivity is ,but: （8） In the formula: r and s represent the nodes of the urban flooding disaster loss index, respectively. and urban element indicator nodes The number of possible values, Indicator nodes for waterlogging disaster losses The i-th parameter and city element indicator node The j-th parameter, Indicator nodes representing losses from urban flooding disasters Urban element indicator nodes The sensitivity, with values ​​ranging from [0,1], The larger the value, the more important it is for urban element indicator nodes. Indicators of losses from urban flooding disasters The greater the degree of influence.

6. The system for obtaining the degree of impact of urban elements on urban flooding disaster losses according to claim 5, characterized in that, Mapping the sets in the ontology model of the mechanism by which urban elements affect urban flooding disasters to the hierarchical Bayesian network structure includes: Based on the set of concepts in the aforementioned mechanism ontology model, the nodes in the hierarchical Bayesian network structure are configured; Based on the conceptual semantic relationships and temporal relationships in the aforementioned mechanism ontology model, the corresponding causal relationships are obtained; The edges of each node in the hierarchical Bayesian network structure are set according to the causal relationship.

7. The system for obtaining the degree of impact of urban elements on urban flooding disaster losses according to claim 6, characterized in that, The conceptual semantic relationships include whole / part relationships, positive influence relationships, negative influence relationships, and relationships with both positive and negative influences. The temporal relationships include earlier than and later than.

8. The system for obtaining the degree of impact of urban elements on urban flooding disaster losses according to claim 5, characterized in that, The parameter learning algorithm is an expectation-maximization algorithm, which includes: calculating the expectation of the log-likelihood function of the nodes in the hierarchical Bayesian network structure, and calculating parameters to maximize the full likelihood expectation.

Images