Urban flood disaster information model construction method and application

Abstract

The invention belongs to the technical field of disaster data storage, and discloses an urban flood disaster information model construction method and application, and the method comprises the steps: constructing an urban flood disaster information model through employing urban flood disaster data; wherein entities and relationships in the urban flood disaster information model correspond to entities and relationships in the urban flood disaster data; entities in the urban flood disaster data comprise basic entities and attribute entities, and the attribute entities are attribute features of the basic entities; the relationship in the urban flood disaster data comprises a spatial relationship and a semantic relationship between the basic entities. The invention further correspondingly provides related application based on the urban flood disaster information model. According to the method, urban flood disaster information modeling is constructed, and multi-source heterogeneous data is fused in combination with time sequence data and a spatial relationship, so that the comprehensiveness of urban flood disaster knowledge expression can be improved, and multi-element integration and analysis management of urban flood disaster data can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of disaster data storage, and more specifically, relates to a method for constructing an urban flood disaster information model and its application. Background Art

[0002] Flood prevention and control is a comprehensive issue that involves multiple dimensions of a city, including society, nature, environment, and management. The element data in different dimensions affect each other. In this context, the logical construction and storage management method of urban flood disaster data based on information models came into being, providing a new idea for the prevention and response of urban flood disasters.

[0003] Compared with other urban disasters (such as earthquakes), the causative factors of urban flood disasters are mainly geographical elements and meteorological elements. Geographical elements include geographical terms, geographical names, geographical distribution and other knowledge that reflects the external characteristics and connections of geographical objects, as well as geographical models that describe geographical evolution laws, geographical prediction laws and other geographical time and space transformations. They are highly professional and targeted knowledge, and other existing urban disaster information models cannot be directly adopted.

[0004] At present, the research on information models of urban flood disaster data is mainly concentrated in the construction stage, usually focusing on the extraction of entity elements and semantic relationships based on urban flood disaster event data to construct the corresponding information model. However, the information model constructed based only on urban flood disaster event data is relatively simple, and the knowledge that can be expressed is relatively one-sided, which also makes the granularity of the subsequent urban flood disaster data chain deduction results based on this information model relatively coarse, and the effect of flood disaster impact prediction results is relatively poor. Summary of the invention

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a method and application for constructing an urban flood disaster information model, the purpose of which is to improve the comprehensiveness of the expression of urban flood disaster knowledge.

[0006] To achieve the above object, according to a first aspect of the present invention, a method for constructing an urban flood disaster information model is provided, comprising:

[0007] An urban flood disaster information model is constructed using urban flood disaster data; wherein the entities and relationships in the urban flood disaster information model correspond to the entities and relationships in the urban flood disaster data;

[0008] The entities in the urban flood disaster data include basic entities and attribute entities, wherein the attribute entities are attribute characteristics of the basic entities; the basic entities include disaster factor data, meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, GDP data, disaster data and danger data in the urban flood disaster data;

[0009] The relationships in the urban flood disaster data include spatial relationships and semantic relationships between basic entities; among which, spatial relationships include hierarchical relationships, parallel relationships, connection relationships and distance relationships; there are hierarchical relationships between block data and boundary and political district data, between building facility data and block data, between block data and land use type data, between block data and meteorological data, between block data and population data, and between block data and GDP data; there are parallel relationships between block data; there are connection relationships between block data and well point data, between well point data and pipeline data, and between building facility data and well point data; there is a distance relationship between river system data and block data; there are semantic relationships between meteorological data and disaster factor data, between disaster factor data and dangerous situation data, between disaster situation and disaster situation data, and between disaster situation data and well point data.

[0010] Furthermore, the urban flood disaster data includes urban flood disaster event data and urban geographic information data;

[0011] The entities in the urban flood disaster data include:

[0012] Respectively extracting the basic entities and attribute entities corresponding to the urban flood disaster event data and the urban geographic information data;

[0013] The basic entities of the urban flood disaster event data and the urban geographic information data are integrated to obtain the basic entities in the urban flood disaster data; the attribute entities of the urban flood disaster event data and the urban geographic information data are knowledge-completed to obtain the attribute entities in the urban flood disaster data.

[0014] Furthermore, the urban flood disaster event data is merged with the basic entities of the urban geographic information data to obtain the basic entities in the urban flood disaster data, including:

[0015] Calculate the cosine value S of the angle between the word vectors of the mth attribute of the basic entity A and the basic entity B Am,Bm :

[0016]

[0017] Wherein, the basic entity A and the basic entity B correspond to any basic entity of the urban flood disaster event data and the urban geographic information data, and the m-th attribute of the basic entity A and the basic entity B has n participles in total; a i , b i The corresponding word vector is the i-th word segment in the m-th attribute of basic entity A and basic entity B;

[0018] Calculate the similarity S between basic entity A and basic entity B in terms of attributes A,B :

[0019]

[0020] Where k represents the total number of attributes of basic entity A and basic entity B;

[0021] The similarity S A,B Two basic entities whose values ​​are greater than a preset threshold are fused to obtain the basic entity in the urban flood disaster data.

[0022] Furthermore, the basic entities in the urban geographic information data include: meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, and GDP data;

[0023] The basic entities in the urban flood disaster event data include: disaster factor data, meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, GDP data, disaster data and dangerous data; among them, the disaster factor data include flood data, mudslide data, landslide data; the disaster data include weather data, economic loss data, population casualties data, house loss data, waterlogging data and pipe burst data.

[0024] Furthermore, the semantic relationship includes: hierarchical relationship, causal relationship, correlation relationship and temporal relationship between entities;

[0025] The disaster situation is hierarchical with weather data, economic loss data, population casualty data, housing loss data, waterlogging data and pipe burst data.

[0026] There is a causal relationship between flood data, debris flow data and landslide data respectively; there is a causal relationship between flood data, debris flow data and landslide data and dangerous situation data;

[0027] There is a correlation between the waterlogging points and the well point data;

[0028] There is a time relationship between disaster factor data, and between disaster factor data and its corresponding time attribute entity; among which, the time attribute entity corresponding to the disaster factor data is the occurrence time of the disaster factor.

[0029] Furthermore, the meteorological data includes historical precipitation data of the city;

[0030] The urban flood disaster information model construction method further includes:

[0031] The historical precipitation data of the city is input into a trained large model to perform precipitation prediction with fixed time resolution and spatial resolution to obtain fitted precipitation prediction data; and the fitted precipitation prediction data is used as the meteorological data.

[0032] According to a second aspect of the present invention, a method for storing urban flood disaster data is provided, comprising:

[0033] Extract entities and relationships from urban flood disaster data;

[0034] The entities and relationships in the extracted urban flood disaster data are stored correspondingly as entities and relationships in the urban flood disaster information model to realize the storage of urban flood disaster data; wherein, the urban flood disaster information model is constructed by the urban flood disaster information model construction method described in any one of the first aspects.

[0035] According to a third aspect of the present invention, a flood disaster spatiotemporal data link deduction method is provided, comprising:

[0036] Query the relevant entity information of the designated flood disaster event from the urban flood disaster information model; and use the spatial relationship and semantic relationship in the urban flood disaster information model to perform disaster chain deduction;

[0037] Wherein, the urban flood disaster information model is constructed by the urban flood disaster information model construction method described in any one of the first aspects.

[0038] According to a fourth aspect of the present invention, a method for predicting the impact of urban flood disasters is provided, comprising:

[0039] Inputting the disaster chain deduced by the flood disaster spatiotemporal data chain deduction method described in the third aspect and the historical urban flood disaster data into the trained secondary disaster factor prediction model to predict the type of secondary disaster subsequently caused by the disaster chain, predict the time and location of the secondary disaster, and / or predict the impact range of the secondary disaster;

[0040] Among them, the secondary disasters include disasters caused by urban floods.

[0041] According to a fifth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method for constructing an urban flood disaster information model as described in any one of the first aspects, the method for storing urban flood disaster data as described in the second aspect, the method for deducing a spatiotemporal data chain of flood disasters as described in the third aspect, and / or the method for predicting the impact of urban flood disasters as described in the fourth aspect is implemented.

[0042] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0043] (1) The method for constructing an urban flood disaster information model of the present invention, in addition to considering disaster factor data, meteorological data, disaster data and dangerous data, also considers river system data, land use type data, boundary and political district data, building facility data, block data, well data, pipeline data, population data and GDP data when constructing an urban flood disaster information model. These data can reflect the element characteristics of flood disaster event data (such as disaster-prone environment) and the element characteristics of urban geographic information data. Based on these data that can reflect the characteristics of urban flood disasters themselves, spatial relationships including hierarchical relationships, parallel relationships, connection relationships and distance relationships are used to associate entities that can simultaneously reflect the element characteristics of flood disaster event data and urban geographic information data, and semantic relationships are used to associate meteorological data, disaster factor data, disaster data and well data. The organizational structure of the urban flood disaster information model constructed in this way is more reasonable, and can fully reflect the spatial relationship and semantic relationship between flood disaster event data and urban geographic information data, thereby improving the comprehensiveness of urban flood disaster knowledge expression.

[0044] (2) Furthermore, considering the impact of urban geographic information data on urban flood disasters, urban geographic information data is used as part of urban flood disaster data, and the spatiotemporal relationship of flood disaster events is combined with the spatial relationship of urban geographic information. The resulting information model integrates urban flood disaster knowledge and urban geographic information, and can fully reflect the characteristics of urban flood disaster data. In addition, when subsequent information utilization is carried out based on this information model, urban flood disaster data resources can be fully utilized, and it is convenient for urban flood disaster data asset management.

[0045] (3) Furthermore, the urban flood disaster information model constructed in the present invention includes the time attribute characteristics (time attribute entities) of the disaster factor data, as well as the time relationship of the disaster factor data, which can fully reflect the temporal relationship of the flood disaster event data, and organically combine the temporal relationship of the flood disaster event data with the spatial relationship of the urban geographic information data, thereby enriching the knowledge expression ability of the urban flood disaster information model.

[0046] (4) As a preference, using precipitation forecast data as meteorological data can make the meteorological data more uniform in time segments.

[0047] (5) Furthermore, the urban flood disaster information model in the present invention combines the spatiotemporal relationship, integrates multi-source heterogeneous data, and performs the spatiotemporal disaster chain deduction of urban flood disasters based on the constructed urban flood disaster information model, which can conduct in-depth analysis of the coupled complex relationship in urban flood disaster data, and then predict secondary disaster factors, and realize the multi-factor integration and analysis management of urban flood disaster data. Moreover, the results of the urban flood disaster data chain deduction and flood disaster impact prediction based on the urban flood disaster information model also have finer granularity and better effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Schematic diagram of a method for constructing an urban flood disaster information model in an embodiment of the present invention.

[0049] Figure 2 The present invention provides a flowchart for constructing an urban flood disaster information model in an embodiment of the present invention.

[0050] Figure 3 Schematic diagram of the knowledge expression structure of urban flood disaster data in an embodiment of the present invention.

[0051] Figure 4 This is an urban flood disaster information model constructed in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] Example 1

[0054] like Figure 1-Figure 3 As shown, an embodiment of the present invention provides a method for constructing an urban flood disaster information model, comprising:

[0055] An urban flood disaster information model is constructed using urban flood disaster data; wherein the entities and relationships in the urban flood disaster information model correspond to the entities and relationships in the urban flood disaster data.

[0056] The entities in urban flood disaster data include basic entities and attribute entities, among which attribute entities are the attribute characteristics of basic entities; basic entities include disaster factor data, meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, GDP data, disaster data and dangerous data in urban flood disaster data; each data is regarded as an entity.

[0057] The relationships in urban flood disaster data include spatial relationships and semantic relationships between basic entities; among them, spatial relationships include: hierarchical relationships, parallel relationships, connection relationships and distance relationships between basic entities. There are hierarchical relationships between block data and boundary and political district data, between building and facility data and block data, between block data and land use type data, between block data and meteorological data, between block data and population data, and between block data and GDP data. There is a parallel relationship between block data and block data. There is a connection relationship between block data and well point data, between well point data and pipeline data, and between building and facility data and well point data. There is a distance relationship between river system data and block data.

[0058] Meteorological data and disaster factor data, disaster factor data and dangerous situation data, disaster situation data and disaster situation data, and disaster situation data and well point data are connected through semantic relationships.

[0059] As a specific implementation method, urban flood disaster data includes urban flood disaster event data and urban geographic information data; wherein, entities extracted from urban flood disaster data include:

[0060] Entities in urban flood disaster event data and entities in urban geographic information data are extracted, and the basic entities in the extracted entities are merged to obtain the basic entities in the urban flood disaster data; the attribute entities in the extracted entities are supplemented with knowledge to obtain the attribute entities in the urban flood disaster data.

[0061] Relationships extracted from urban flood disaster data include:

[0062] Extract the relationships between entities in flood disaster event data and the relationships between entities in urban geographic information data; align the extracted relationships to obtain the relationships in the extracted urban flood disaster data.

[0063] In the embodiment of the present invention, with reference to the regional natural disaster system theory, the Flood Control Regulations of the People's Republic of China, the statistics of natural disasters and other relevant national standards, and combined with the characteristics of urban flood disasters, the attributes of urban flood disaster event data are described from the perspectives of disaster factors, disaster-prone environment, disaster situation and dangerous situation. Urban flood disaster disaster disaster factors include primary disaster factors and secondary disaster factors. In the embodiment of the present invention, the primary disaster factors are urban flood disasters, and the secondary disaster factors refer to mudslides and landslides caused by urban flood disasters. Disaster-prone environment data include meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data and GDP data. Disaster situation data include weather data, economic loss data, population casualties data, house loss data, waterlogging point data and pipe burst data. That is, in the embodiment of the present invention, the basic entities in the flood disaster event data include: disaster factor data, disaster-prone environment data, disaster situation data and dangerous situation data. Among them, disaster-causing factor data, disaster-prone environment data, disaster situation data and risk data are first-level entities, and their corresponding second-level entities and attribute entities are shown in Table 1-3.

[0064] Table 1 Classification of disaster factor data and corresponding attribute entities

[0065]

[0066] Table 2 Disaster-prone environmental data and corresponding attribute entities

[0067]

[0068] Table 3 Specific classification of disasters and dangerous situations and corresponding attribute entities

[0069]

[0070] In the embodiment of the present invention, the flood disaster event data is text data, and the acquisition method includes: data related to flood disaster events released by the government and the Internet, and historical flood event related materials. Specifically, it includes semi-structured data including data related to flood disaster events released by the municipal people's government, etc. (such as flood disaster accident reports) and data related to flood disaster events published on encyclopedias, as well as historical flood event related materials, unstructured data related to flood disaster events on Weibo and web news.

[0071] In an embodiment of the present invention, the basic entities in the urban geographic information data include vector data including river system data, boundary and political district data, building facility data, block data, well point data, pipeline data, and raster data including land use type data, population data, GDP data and meteorological data.

[0072] The relationship between the entities corresponding to urban flood disaster event data and urban geographic information data includes two categories: spatial relationship and semantic relationship. Among them, the relationship between the entities corresponding to urban geographic information data is mainly spatial relationship. Spatial relationship includes: hierarchical relationship, parallel relationship, connection relationship and distance relationship between basic entities. Semantic relationship includes superior-subordinate relationship, causal relationship, correlation relationship and time relationship between basic entities. Among them, the relationship between the first-level disaster situation and its corresponding second-level basic entity is a superior-subordinate relationship. For example, the disaster situation data includes weather data, economic loss data, population casualty data, house loss data, waterlogging point data and pipe burst data. The disaster situation data and weather data, economic loss data, population casualty data, house loss data, waterlogging point data and pipe burst data are all superior-subordinate relationships. There is a causal relationship between the primary hazard factors and the secondary hazard factors in the hazard factor data, between the hazard factor data and the dangerous situation data, and between the meteorological data and the hazard factor data. For example, floods cause mudslides, floods cause landslides, meteorology causes floods, meteorology causes mudslides, and meteorology causes landslides. There is a causal relationship between flood data, mudslide data, and landslide data and dangerous situation data, for example, floods cause dangerous situations. There is a correlation relationship between the waterlogging points and well point data in the disaster situation data. The hazard factor data and the hazard factor data and their corresponding time attribute entities are connected through a time relationship. For example, there is a time relationship between floods, mudslides, and landslides and their corresponding time attribute entities. The details are shown in Table 4 below.

[0073] Table 4 Relationship categories

[0074]

[0075] The notation in Table 4: <entity 1, relationship A, entity 2> means that entity 1 points to entity 2 and there is a relationship A between the two.

[0076] As one of the implementation methods, the ERNIE-UIE model is used to extract basic entities, attribute entities and relationships from urban flood disaster event data, including:

[0077] Preprocessing of urban flood disaster event data includes: data format conversion, jieba filtering stop words, regular expression library eliminating invalid data; fine-tuning the trained model provided by ERNIE-UIE; using the fine-tuned ERNIE-UIE model to extract basic entities, attribute entities and relations from the processed urban flood disaster event data, so as to obtain the basic entities, attribute entities and relations corresponding to the flood event data.

[0078] Preferably, for the disaster factor data in the urban flood disaster event data, the corresponding attribute entities include a time attribute entity and a location attribute entity. The time attribute entity is the time when the disaster occurred, and the location attribute entity is the location where the disaster occurred. Extracting the time attribute entity includes: matching the string with the time expression form in the pre-processed urban flood disaster event data by regular expression, thereby extracting the time information, and standardizing it, and unifying the time expressions of different formats by using python to obtain the time attribute entity. In the embodiment of the present invention, the unified time expression is "YYYYMMDDHH.mm".

[0079] After extracting the time attribute entity, the time relationship between the disaster factor data is obtained, including: arranging the urban flood disaster events in chronological order, that is, arranging the time expressions with unified formats in chronological order, and obtaining the time series information of the urban flood disaster events, which is used to reflect the time relationship between the disaster factor entities.

[0080] The urban geographic information data is extracted using the Python programming language embedded in GIS to extract basic entities, attribute entities and relationships, including:

[0081] First, the urban geographic information data is preprocessed, including: using GIS to read shp files (vector data of urban river systems, boundaries and political districts, building facilities, blocks, well points and pipelines) and tif data (raster data of population distribution, GDP, land use type, and meteorology), obtain the number, name, longitude and latitude of meteorological and river system data, and unify the coordinates, projection transformation, and elevation benchmark of various urban geographic information data;

[0082] Then, the GIS hydrological analysis and terrain analysis modules are used to analyze the geographic attributes of the preprocessed data, such as "terrain relief", "terrain humidity index", and "terrain roughness index", to obtain the basic entities and attribute entities corresponding to the urban geographic information data;

[0083] Finally, Python programming language is used to analyze the spatial relationships of the extracted entities.

[0084] As one of the implementation methods, after extracting the entities and relationships of urban flood disaster event data and urban geographic information data, the knowledge of multi-source heterogeneous data sources is integrated by using entity alignment, relationship alignment, and knowledge completion. Among them, entity alignment uses entity similarity calculation, including:

[0085] Calculate the cosine value S of the angle between the word vectors of the mth attribute of the basic entity A and the basic entity B Am,Bm :

[0086]

[0087] Among them, basic entity A and basic entity B correspond to any basic entity in urban flood disaster event data and urban geographic information data; and the m-th attribute of basic entity A and basic entity B has n participles in total; a i 、b i The corresponding word vector is the i-th word vector in the m-th attribute of basic entity A and basic entity B. The word vector of the i-th word vector is obtained using the GLOVE algorithm. m and B m Represents the mth attribute vector in basic entity A and basic entity B respectively; |A m | and |B m | respectively represent vector A m and B m The absolute value of the basic entity. Here, the attribute of the basic entity is the attribute entity corresponding to the basic entity.

[0088] Calculate the similarity S between basic entity A and basic entity B in terms of attributes A,B :

[0089]

[0090] Where k represents the total number of attributes of basic entity A and basic entity B.

[0091] The similarity S A,B The two basic entities with values ​​greater than a preset threshold are fused to obtain the basic entities in the urban flood disaster data.

[0092] Preferably, the meteorological data includes historical precipitation data of the city, and also includes homogenizing the meteorological data, specifically including: inputting the historical precipitation data of the city into a trained large model (such as the Fuxi large model), performing precipitation prediction with fixed time resolution and spatial resolution, obtaining the precipitation prediction model with the best fitting effect, and then obtaining the precipitation prediction data with the best effect, and using the precipitation prediction data as meteorological data, which can make the meteorological data more uniform in time segmentation. In the embodiment of the present invention, precipitation prediction is performed on the historical precipitation data of the city with a time resolution of 6 hours and a spatial resolution of 0.25°.

[0093] Furthermore, the urban flood disaster information model constructed above also includes attribute characteristics of relationships. Among them, the attribute characteristics of hierarchical relationships, parallel relationships, connection relationships and distance relationships correspond to hierarchy, parallel, connection and distance. The attribute characteristics of hierarchical relationships, causal relationships, correlation relationships and time relationships correspond to hierarchical, causal, correlation and time.

[0094] The method for constructing an urban flood disaster information model of the present invention, in addition to considering disaster factor data, meteorological data, disaster data and dangerous data, also considers river system data, land use type data, boundary and political district data, building facility data, block data, well data, pipeline data, population data and GDP data when constructing an urban flood disaster information model. These data can reflect the element characteristics of flood disaster event data (such as disaster-prone environment) and the element characteristics of urban geographic information data. Based on these data that can reflect the characteristics of urban flood disasters themselves, spatial relationships including hierarchical relationships, parallel relationships, connection relationships and distance relationships are used to associate entities that can simultaneously reflect the element characteristics of flood disaster event data and urban geographic information data, and semantic relationships are used to associate meteorological data, disaster factor data, disaster data and well data, so that the organizational structure of the urban flood disaster information model constructed in this way is more reasonable, and can fully reflect the spatial relationship and semantic relationship between flood disaster event data and urban geographic information data, thereby improving the comprehensiveness of urban flood disaster knowledge expression. In addition, the subsequent urban flood disaster data chain simulation and flood disaster impact prediction are carried out based on the urban flood disaster information model, and the results also have finer granularity and better effects.

[0095] By converting urban flood disaster data into information model storage, the existing query statements can be used to quickly access and analyze urban flood disaster data, thereby improving the ability to query and make real-time decisions on urban flood disaster data. In addition, the information model constructed by the present invention defines different entities according to different types of urban flood disaster data. When the storage object changes or the storage structure needs to be adjusted, the entities, the relationships between entities, and the attributes of the relationships can be directly added, subtracted, or modified in the information model, such as Figure 4 shown.

[0096] The present invention constructs an urban flood disaster information model based on the inherent attributes of urban flood disaster data, and the urban flood disaster knowledge that can be expressed can be more abundant and comprehensive.

[0097] Example 2

[0098] An embodiment of the present invention provides a method for storing urban flood disaster data, comprising:

[0099] Extracting entities and relationships from urban flood disaster data; wherein, extracting entities and relationships from urban flood disaster data refers to the corresponding description in the above-mentioned embodiment 1;

[0100] The entities and relationships in the extracted urban flood disaster data are stored correspondingly as the entities and relationships in the urban flood disaster information model constructed in Example 1, so as to realize the storage of urban flood disaster data.

[0101] As a specific implementation manner, the storage method includes:

[0102] The meteorological data, river system data, land use type data, weather data, boundary and political district data, building facility data, block data, well point pipeline data, disaster data, and danger data in the urban flood disaster data are stored as basic entities of the urban flood disaster information model constructed in Example 1;

[0103] Use spatial and semantic relationships to associate basic entities;

[0104] Add attribute features of basic entities and relationships to realize the storage of urban flood disaster data. For example, the "occurrence time" of flood disasters, the "river name, distance to the river, river bottom area" of the river system, the "building facility ID, building facility type" of the building facilities, and the "block ID, block building density, average height of the block building" and other attribute data are written into the information model in batches; for another example, the "hourly rainfall, daily rainfall, cumulative rainfall" attribute data of the weather entity in the disaster risk situation are written into the information model in batches, and the relationships between entities are written into the information model in batches. In the embodiment of the present invention, the basic entities and attribute entities as well as the relationships between entities are written into the information model through python.

[0105] Example 3

[0106] The embodiment of the present invention provides a method for performing spatiotemporal data chain deduction of flood disasters using the urban flood disaster information model constructed in Embodiment 1, including:

[0107] Query the relevant entity information of the specified flood disaster event from the urban flood disaster information model constructed in Example 1; for example, query the disaster entity (such as weather, economic losses, casualties, house losses, waterlogging points, pipe bursts, etc.), time information (such as the occurrence time of the disaster factor), and spatial information (such as longitude and latitude, blocks, boundaries and conditions, etc.).

[0108] Disaster chain deduction is carried out by using the spatial and semantic relationships in the urban flood disaster information model. For example, the causal relationship in the information model is used to analyze the causal relationship between disaster entities, and the transmission path and key nodes of the disaster chain are determined, including the starting entity that induces urban flood disasters, and the entity with a sharp increase in economic losses or housing losses caused by urban flood disasters. The temporal relationship between entities in the information model is used to analyze the time sequence of the occurrence of disaster entities, build a time series model of the disaster chain, and use graph algorithms to predict the development trend of disasters. The spatial relationship between entities in the information model is used to analyze the spatial relationship between disaster-prone environment entities, including distance relationships, hierarchical relationships, etc., to build a spatial propagation model of the disaster chain, and use graph algorithms to predict the impact range of disasters.

[0109] For related schemes, please refer to the corresponding description in Example 1, which will not be repeated here.

[0110] Example 4

[0111] The embodiment of the present invention provides a method for predicting the impact of urban flood disasters, including:

[0112] The spatiotemporal characteristics of the disaster chain obtained by the deduction method in Example 3 and the historical urban flood disaster data are input into the trained secondary disaster factor prediction model to predict the type of secondary disasters subsequently caused by the disaster chain, such as secondary disasters (mudslides, landslides, etc.) that may be caused by floods; and the time and location of the secondary disasters; and the scope of the impact of the secondary disasters, such as the wells and blocks that may be affected by floods. In the embodiment of the present invention, a graph algorithm is used to establish a secondary disaster factor prediction model.

[0113] For the relevant scheme, please refer to the corresponding description in Example 3, which will not be repeated here.

[0114] Example 5

[0115] The embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for constructing an urban flood disaster information model as in embodiment 1 is implemented, the method for storing urban flood disaster data as in embodiment 2 is implemented, the method for deducing a spatiotemporal data chain of flood disasters as in embodiment 3 is implemented, and / or the method for predicting the impact of urban flood disasters as in embodiment 4 is implemented. For related solutions, please refer to the description of the corresponding embodiments, which will not be repeated here.

[0116] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing an urban flood disaster information model, characterized in that: include: An urban flood disaster information model is constructed using urban flood disaster data; wherein the entities and relationships in the urban flood disaster information model correspond to the entities and relationships in the urban flood disaster data; The entities in the urban flood disaster data include basic entities and attribute entities, wherein the attribute entities are attribute characteristics of the basic entities; the basic entities include disaster factor data, meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, GDP data, disaster data and danger data in the urban flood disaster data; The relationships in the urban flood disaster data include spatial relationships and semantic relationships between basic entities; among which, spatial relationships include hierarchical relationships, parallel relationships, connection relationships and distance relationships; there are hierarchical relationships between block data and boundary and political district data, between building facility data and block data, between block data and land use type data, between block data and meteorological data, between block data and population data, and between block data and GDP data; there are parallel relationships between block data; there are connection relationships between block data and well point data, between well point data and pipeline data, and between building facility data and well point data; there is a distance relationship between river system data and block data; there are semantic relationships between meteorological data and disaster factor data, between disaster factor data and dangerous situation data, between disaster situation and disaster situation data, and between disaster situation data and well point data.

2. The method for constructing an urban flood disaster information model according to claim 1, characterized in that: The urban flood disaster data includes urban flood disaster event data and urban geographic information data; The entities in the urban flood disaster data include: Respectively extracting the basic entities and attribute entities corresponding to the urban flood disaster event data and the urban geographic information data; The basic entities of the urban flood disaster event data and the urban geographic information data are integrated to obtain the basic entities in the urban flood disaster data; the attribute entities of the urban flood disaster event data and the urban geographic information data are knowledge-completed to obtain the attribute entities in the urban flood disaster data.

3. The method for constructing an urban flood disaster information model according to claim 2, characterized in that: The urban flood disaster event data is merged with the basic entities of the urban geographic information data to obtain the basic entities in the urban flood disaster data, including: Calculate the cosine value S of the angle between the word vectors of the mth attribute of the basic entity A and the basic entity B Am,Bm : Wherein, the basic entity A and the basic entity B correspond to any basic entity of the urban flood disaster event data and the urban geographic information data, and the m-th attribute of the basic entity A and the basic entity B has n participles in total; a i 、b i The corresponding word vector is the i-th word segment in the m-th attribute of basic entity A and basic entity B; Calculate the similarity S between basic entity A and basic entity B in terms of attributes A,B : Where k represents the total number of attributes of basic entity A and basic entity B; The similarity S A,B Two basic entities whose values ​​are greater than a preset threshold are fused to obtain the basic entity in the urban flood disaster data.

4. The method for constructing an urban flood disaster information model according to claim 2 or 3, characterized in that: The basic entities in the urban geographic information data include: meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, and GDP data; The basic entities in the urban flood disaster event data include: disaster factor data, meteorological data, river system data, land use type data, boundary and political district data, building facility data, block data, well point data, pipeline data, population data, GDP data, disaster data and dangerous data; among them, the disaster factor data include flood data, mudslide data, landslide data; the disaster data include weather data, economic loss data, population casualties data, house loss data, waterlogging data and pipe burst data.

5. The method for constructing an urban flood disaster information model according to claim 4, characterized in that: The semantic relationship includes: hierarchical relationship, causal relationship, correlation relationship and temporal relationship between entities; The disaster situation is hierarchical with weather data, economic loss data, population casualty data, housing loss data, waterlogging data and pipe burst data. There is a causal relationship between flood data, debris flow data and landslide data respectively; there is a causal relationship between flood data, debris flow data and landslide data and dangerous situation data; There is a correlation between the waterlogging points and the well point data; There is a time relationship between disaster factor data, and between disaster factor data and its corresponding time attribute entity; among which, the time attribute entity corresponding to the disaster factor data is the occurrence time of the disaster factor.

6. The method for constructing an urban flood disaster information model according to claim 1, characterized in that: The meteorological data include historical precipitation data of the city; The urban flood disaster information model construction method further includes: The historical precipitation data of the city is input into a trained large model to perform precipitation prediction with fixed time resolution and spatial resolution to obtain fitted precipitation prediction data; and the fitted precipitation prediction data is used as the meteorological data.

7. A method for storing urban flood disaster data, characterized in that: include: Extract entities and relationships from urban flood disaster data; The entities and relationships in the extracted urban flood disaster data are stored correspondingly as entities and relationships in the urban flood disaster information model to realize the storage of urban flood disaster data; wherein, the urban flood disaster information model is constructed by the urban flood disaster information model construction method described in any one of claims 1-6.

8. A method for deducing spatiotemporal data links of flood disasters, characterized in that: include: Query relevant entity information of a specified flood disaster event from the urban flood disaster information model; And use the spatial and semantic relationships in the urban flood disaster information model to deduce the disaster chain; Wherein, the urban flood disaster information model is constructed by the urban flood disaster information model construction method described in any one of claims 1-6.

9. A method for predicting the impact of urban flood disasters, characterized in that: include: Input the disaster chain deduced by the flood disaster spatiotemporal data chain deduction method described in claim 8 and the historical urban flood disaster data into the trained secondary disaster factor prediction model to predict the type of secondary disaster subsequently caused by the disaster chain, predict the time and location of the secondary disaster, and / or predict the impact range of the secondary disaster; Among them, the secondary disasters include disasters caused by urban floods.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, the method for constructing an urban flood disaster information model as described in any one of claims 1 to 6 is implemented, the method for storing urban flood disaster data as described in claim 7 is implemented, the method for deducing a spatiotemporal data link of flood disasters as described in claim 8 is implemented, and / or the method for predicting the impact of urban flood disasters as described in claim 9 is implemented.

Images