Water event risk prediction method, device, computer equipment, and storage medium

Through the analysis of water event knowledge graph, the problem of insufficient analysis of the impact of local precipitation in traditional water event analysis methods is solved, accurate risk prediction and dynamic update are achieved, and real-time emergency management is supported.

CN118821928BActive Publication Date: 2024-11-22CHINA INST OF WATER RESOURCES & HYDROPOWER RES
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411306033.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-19
Publication Date
2024-11-22
Estimated Expiration
2044-09-19

AI Technical Summary

Technical Problem

The traditional water event analysis method is based on a large-scale precipitation object, ignoring the nuances of local areas, resulting in insufficient accurate analysis of the impact of precipitation. The existing methods lack dynamic and real-time correlation analysis, and cannot effectively support emergency management and real-time scheduling. .

Method used

Provide a water event risk prediction method, by obtaining pre-constructed water event knowledge maps, extracting water event historical data from target areas, generating structured data, determining risk factors related to spatial data and topographic data, and predicting risk types and levels. And analyze the effects of precipitation through relationship maps.

Benefits of technology

It realizes accurate analysis of the impact of precipitation in local areas, dynamically updates hydrological information, supports real-time risk prediction and emergency management, and improves the comprehensiveness and accuracy of the analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118821928B_ABST
    Figure CN118821928B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a method, device, computer equipment, and storage medium for predicting water event risks. The method includes: obtaining a pre-constructed water event knowledge graph; the water event knowledge graph includes: extracting data based on at least historical water event data of the target area to obtain triples that match pre-set prompt words; converting triples into structured data in a graph database; generating a water event knowledge graph based on at least structured data in the graph database; determining spatial data and terrain data of the predicted area in the target area, and determining the risk type of the predicted area based on the spatial data, terrain data, and risk factors related to the spatial data and terrain data of the predicted area; predicting the risk level of the predicted area based on the risk type of the predicted area and the data of the predicted area in the water event knowledge graph. The present method can accurately predict the risk level of the predicted area.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method, device, computer equipment, and storage medium for predicting water event risks. Background Art

[0002] With the continuous development of big data, people usually use various data to analyze and predict the disaster level or impact of impending natural disasters. However, traditional water event analysis usually divides a large range of precipitation objects based on the time dimension, such as the north and south tributaries of a river basin. Although this method can grasp the overall precipitation situation at a macro level, it has the problem of weak granularity in analyzing the subsequent impact after precipitation in local areas. Summary of the invention

[0003] Based on this, it is necessary to provide a water event risk prediction method, device, computer equipment, and storage medium to address the above technical issues.

[0004] In a first aspect, the present disclosure provides a method for predicting water event risks. The method comprises:

[0005] Obtaining a pre-constructed water event knowledge graph; the water event knowledge graph comprises: extracting data based on at least historical water event data in a target area to obtain triples matching pre-set prompt words; converting the triples into structured data in a graph database; generating a water event knowledge graph based at least on the structured data in the graph database;

[0006] Determine spatial data and topographic data of a prediction area in the target area, and determine a risk type of the prediction area based on the spatial data, topographic data of the prediction area and risk factors associated with the spatial data and topographic data;

[0007] Based on the risk type of the predicted area and the data of the predicted area in the water event knowledge graph, the risk level of the predicted area is predicted.

[0008] In one embodiment, the method further comprises:

[0009] Determine precipitation objects in the target area based on historical precipitation data of the target area, a preset precipitation threshold, and a preset area threshold;

[0010] Generate a relationship graph based on the precipitation object, the geospatial information of the target area and the water event knowledge graph;

[0011] Determining a current precipitation intensity in the target area, and determining an associated step number based on the current precipitation intensity;

[0012] The precipitation impact of the current precipitation intensity in the target area is determined based on the associated step number and the relationship map.

[0013] In one embodiment, determining the precipitation object in the target area according to the historical precipitation data of the target area, the preset precipitation threshold and the preset area threshold includes:

[0014] Acquire historical precipitation data of the target area, and determine a precipitation threshold value according to the historical precipitation data;

[0015] Determining a first area in the target region where historical precipitation data exceeds the precipitation threshold;

[0016] Based on the area of ​​the first area and a preset area threshold, determining a second area in the first area that is smaller than the area threshold;

[0017] A precipitation object is determined based on the second area, and attribute information of the precipitation object is set, where the attribute information includes: precipitation amount, area, and precipitation time.

[0018] In one embodiment, generating a relationship graph based on the precipitation object, the geospatial information of the target area and the water event knowledge graph includes:

[0019] Determining the geographic spatial information of the target area, superimposing the precipitation object with the geographic spatial information to obtain superimposed comprehensive information;

[0020] Performing entity recognition and data matching on the superimposed comprehensive information to determine entities in the superimposed comprehensive information;

[0021] A relationship graph is generated based on entities in the superimposed comprehensive information and the precipitation objects present in the water event knowledge graph.

[0022] In one embodiment, the method further comprises:

[0023] Determine the precipitation object in the target area, and determine the downstream water system object matching the precipitation object;

[0024] Tracking the impact path according to the downstream water system object matching the precipitation object to determine the downstream area affected by the precipitation object;

[0025] Updating the downstream area affected by the precipitation object in the water event knowledge graph;

[0026] Determine the flood peak propagation time, and mark the time windows affected by different entities in the knowledge graph based on the flood peak propagation time;

[0027] Overlay analysis is performed on the precipitation object to determine the cross-impact area, and the water event knowledge graph is updated based on the cross-impact area.

[0028] In one embodiment, the method further comprises:

[0029] Determining time series data of historical water events in the water event knowledge graph, and determining a periodic pattern of the time series data;

[0030] Identifying a change point in the time series, determining a time node based on the change point and the periodic pattern; and determining a category of the periodic pattern based on the time node;

[0031] Configuring a hydrological event perception chain, and building a water event impact chain based on the hydrological event perception chain and the temporal relationship in the water event knowledge graph;

[0032] A plurality of precipitation scenarios are set for the forecast area, simulation prediction is performed based on the water event knowledge graph and the plurality of precipitation scenarios for the forecast area, and the time when each entity in the water event knowledge graph is affected by the precipitation scenario is determined.

[0033] In one embodiment, determining the risk type of the predicted area based on the spatial data, topographic data of the predicted area and risk factors related to the spatial data and topographic data includes:

[0034] Superimposing the spatial data to obtain multi-layer spatial data;

[0035] Performing a first analysis on the multi-layer spatial data according to the levels to obtain a first analysis result;

[0036] Performing a second analysis on the terrain data to obtain a second analysis result;

[0037] Determining risk factors affecting risk based on the first analysis result and the second analysis result;

[0038] A risk index of the predicted area is calculated based on the risk factor, and a risk type of the predicted area is determined based on the risk index and a preset risk threshold interval.

[0039] In a second aspect, the present disclosure further provides a water event risk prediction device. The device comprises:

[0040] A knowledge graph acquisition module is used to acquire a pre-built water event knowledge graph; the water event knowledge graph includes: extracting data based on at least historical water event data in a target area to obtain triples matching pre-set prompt words; converting the triples into structured data in a graph database; generating a water event knowledge graph based on at least the structured data in the graph database;

[0041] a risk type determination module, configured to determine spatial data and terrain data of a predicted area in the target area, and determine the risk type of the predicted area based on the spatial data, terrain data of the predicted area and risk factors associated with the spatial data and terrain data;

[0042] The risk level prediction module is used to predict the risk level of the prediction area based on the risk type of the prediction area and the data of the prediction area in the water event knowledge graph.

[0043] In a third aspect, the present disclosure further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in any of the above method embodiments are implemented.

[0044] In a fourth aspect, the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0045] In a fifth aspect, the present disclosure further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0046] In the above embodiments, data extraction is performed based on at least the historical data of water events in the target area to obtain a triple that matches the preset prompt word; the triple is converted into structured data in the graph database; and a water event knowledge graph is generated based on at least the structured data in the graph database, which includes multiple dimensions such as water conservancy facilities, watershed information, and historical events. The water event knowledge graph can be dynamically updated and maintained to reflect the latest hydrological information and scheduling schemes. In this way, we can establish the influence relationship between the objects in the water event, associate the upstream and downstream water conditions and work conditions and other perception information, and form a systematic impact chain analysis. When it is necessary to perform risk analysis on the predicted area, the spatial data and terrain data of the predicted area in the target area can be determined, and the risk type of the predicted area can be determined based on the spatial data, terrain data and risk factors related to the spatial data and terrain data of the predicted area. The risk of the predicted area can be analyzed spatially, thereby determining the risk type of the predicted area at the spatial level. Combining the risk type and the data of the predicted area in the water event knowledge graph, the water event knowledge graph can be fully utilized to accurately predict the risk level of the predicted area. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the specific embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 A schematic diagram of an application environment of a water incident risk prediction method in an embodiment;

[0049] Figure 2 A schematic diagram of a process of predicting water event risk in an embodiment;

[0050] Figure 3 A schematic diagram of a process for determining precipitation impact in one embodiment;

[0051] Figure 4 Schematic diagram of the process of step S302 in one embodiment;

[0052] Figure 5 Schematic diagram of the process of step S304 in one embodiment;

[0053] Figure 6 A schematic diagram of a process for updating a water event knowledge graph in one embodiment;

[0054] Figure 7 A schematic diagram of a flow chart of time analysis in one embodiment;

[0055] Figure 8 Schematic diagram of the process of step S204 in one embodiment;

[0056] Fig. 9 It is a schematic block diagram of the structure of a water event risk prediction device in one embodiment;

[0057] Fig.10 is a schematic diagram of the internal structure of a computer device in one embodiment;

[0058] Fig.11 Schematic diagram of the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure 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 disclosure and are not used to limit the present disclosure.

[0060] It should be noted that the terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0061] In this article, the term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship.

[0062] As described in the background technology, traditional water event analysis methods are usually aimed at a large range of precipitation objects, ignoring the subtle differences in local areas, resulting in insufficient accurate analysis of precipitation impacts. In addition, there are currently a large number of literature materials on water conservancy history flood control plans, rainstorm and flood records, historical use records of water conservancy projects, and scheduling plans, which provide important data and information for water resources management. However, these materials have the following deficiencies: lack of summary and analysis of key nodes in extreme hydrological events, unable to systematically reveal the occurrence, development and results of events; literature materials are mostly static records, lack of dynamic and real-time correlation analysis, unable to effectively support emergency management and real-time scheduling; it is difficult to use existing literature materials for comprehensive analysis to discover potential impact chains and correlations. Finally, at this stage, the analysis of the impact of extreme precipitation, water level, and flow in the upstream on the downstream mainly relies on the experience of water conservancy experts and professional water conservancy calculation models. Although these methods have certain scientificity and reliability, they have the following problems in causal inference: relying on expert experience, there is a risk of subjective judgment, and it is difficult to ensure the comprehensiveness and accuracy of the analysis. Although professional computational models can provide relatively accurate predictions, their complexity and high data requirements limit their widespread promotion in practical applications. They lack causal inference on the mutual influence between water-related objects, cannot systematically reveal the dynamic connection between upstream and downstream, and are difficult to provide global perspective support for comprehensive water event predictions.

[0063] In view of the above problems, the present disclosure provides a water event risk prediction method, which can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 obtains a pre-constructed water event knowledge graph; the water event knowledge graph includes: extracting data based on at least the historical data of water events in the target area to obtain triples that match the preset prompt words; converting the triples into structured data in the graph database; generating a water event knowledge graph based on at least the structured data in the graph database. The water event knowledge graph can be pre-established in the terminal 102 or in the server 104. The terminal 102 determines the spatial data and terrain data of the predicted area in the target area, and determines the risk type of the predicted area based on the spatial data, terrain data and risk factors related to the spatial data and terrain data of the predicted area. The terminal 102 predicts the risk level of the predicted area based on the risk type of the predicted area and the data of the predicted area in the water event knowledge graph. The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0064] In one embodiment, Figure 2 As shown in the figure, a water event risk prediction method is provided, which is applied to Figure 1 The terminal 102 in the example is used as an example to illustrate, and the following steps are included:

[0065] S202, obtaining a pre-constructed water event knowledge graph; the water event knowledge graph includes: extracting data based on at least historical water event data in the target area to obtain triples that match pre-set prompt words; converting the triples into structured data in a graph database; generating a water event knowledge graph based on at least the structured data in the graph database.

[0066] Among them, the water event knowledge graph is a structured data set involving knowledge in the field of water disaster prediction. It presents different types of water disaster-related knowledge in the form of a graph to facilitate better understanding and analysis of relevant information about water events. Water event historical data may include: literature materials such as water conservancy historical data and scheduling plans. Literature materials such as water conservancy historical data and scheduling plans refer to literature materials that record the development history of water conservancy projects, water resources utilization, water conservancy project design plans, water resources scheduling plans, etc. These documents include records and descriptions of various aspects of water conservancy project design, construction, operation and management, as well as investigation and research reports on water resources development and utilization in different time periods and regions. Taking the target area as an example, these water event historical data may include reservoir information, river information, etc. in the target area. Triples are ordered sets of three elements. They are usually used to represent a data structure of a fact, relationship or data. In fields such as knowledge graphs and semantic webs, triples are also often used to represent relationships between entities. The triples are usually presented in the form of subject-predicate-object, for example (subject, predicate, object), where the subject represents the main entity, the predicate represents the relationship or attribute, and the object represents the content associated with the main entity. In some embodiments of the present disclosure, the prompt words are usually words that extract key information from the historical data of water events.

[0067] Specifically, first, the prompt setting is fixed: a set of prompt words is set to extract information from the historical data of water events. For example, the prompt words may include "find the name of water conservancy facilities", "identify the dispatching rules", etc. The literature of the historical data of water events is input into LLM (Large Language Models). LLM is used to extract knowledge triples from the literature of the historical data of water events based on the prompt words. For example, a triple such as "reservoir-dispatching rules-flood season water release" is extracted from a description. The extracted triples are sorted in the form of entity-relationship-attribute. Among them, "prompt" usually refers to the initial text or question entered when using certain artificial intelligence models (such as GPT-3, BERT, etc.), which is used to trigger the model to generate corresponding text output. In this case, the initial text provided by the user is regarded as a prompt, and the model will generate a corresponding response or result based on this prompt. After obtaining the triples, the target graph database (such as Neo4j or NebulaGraph) can be connected. By establishing a connection with the target database, it can be ensured that subsequent operations are performed in the correct database environment. According to the entity information in the triples, the triples are converted into structured data in the graph data, so as to generate a water event knowledge graph based on the structured data.

[0068] In some exemplary embodiments, for example, if the triple involves entities such as "reservoir" and "river", it is necessary to create corresponding nodes in the graph database, and each node represents a specific entity object. After the node is created, edges are created between the nodes according to the relationship information in the triple. For example, if the triple indicates that there is a "dispatching rule" relationship between "reservoir" and "flood season water release", it is necessary to create an edge representing the "dispatching rule" between the "reservoir" node and the "flood season water release" node. These edges represent the relationship or connection between the nodes. To ensure that the nodes and edges can fully reflect the actual situation, it is necessary to set relevant attributes for them. For example, a node can have attributes such as name, location, and capacity, while an edge can have attributes such as relationship type and condition. By setting these attributes, the information in the graph database can be enriched to make it more operational and reference-worthy, thereby converting the triple into structured data in the graph database. Finally, data verification is performed to ensure that all triples have been correctly converted and recorded in the graph database. Checking whether the nodes and edges are created as expected and whether the attributes are set correctly is an important step to ensure data accuracy and integrity. Then, a water event knowledge graph is generated according to the data recorded in the graph database.

[0069] In other exemplary embodiments, after generating the water event knowledge graph, since some words in the water event historical data are expressed in different ways but are the same entity, it is necessary to perform reference resolution on the water event knowledge graph. For example, although "the river" and "this river" are expressed in different ways, they all refer to the same river entity. These synonyms and pronouns can be identified. The identified synonyms and pronouns are mapped to the corresponding nodes in the knowledge graph. This step ensures that entities with different expressions in the literature can correctly correspond to specific nodes in the knowledge graph. Context analysis is performed to determine the specific object of the pronoun by analyzing the context in the water event historical data. For example, in a description, the context of "the river" may refer to a specific river. This requires matching the context information with the nodes in the knowledge graph to ensure that the pronoun can be accurately mapped to the corresponding entity nodes. Finally, the application of reference resolution is performed. The reference resolution results obtained by identification and analysis are applied to the knowledge graph to ensure that the representation of each entity is consistent.

[0070] S204, determining the spatial data and topographic data of the prediction area in the target area, and determining the risk type of the prediction area based on the spatial data, topographic data of the prediction area and risk factors related to the spatial data and topographic data.

[0071] Among them, spatial data may include: topography, land use, urbanization level and other data of the predicted area. Topographic data may include: terrain height, terrain undulation, water system distribution and so on. Risk factors may be parameters of the degree of impact on risk. Risk types may include high risk, medium risk and low risk.

[0072] Specifically, the spatial data and terrain data of the predicted area in the target area can be determined through digital elevation models (DEM), topographic maps or satellite remote sensing data. Then the spatial data and terrain data are analyzed to determine the risk factors related to the spatial data and terrain data. The risk of the predicted area is scored according to the risk factors, spatial data and terrain data to determine the risk type of the predicted area.

[0073] S206: Predict the risk level of the predicted area based on the risk type of the predicted area and the data of the predicted area in the water event knowledge graph.

[0074] Specifically, the entity information and attribute information of the predicted area can be obtained from the water event knowledge graph. For example, the coordinate elevation, flow boundary conditions (m³ / s) and other attributes of the predicted area can be obtained from the water event knowledge graph, and the input data of the distributed hydrological model can be summarized according to the input instructions to calculate the water level, flow distribution and other information within the time range. Then, the historical flood records and dispatch rule attributes of the region in the water event knowledge graph are further read to obtain the flood risk impact level of the predicted area, such as once in ten years, once in fifty years, etc., so as to obtain the risk level of the predicted area. Among them, the distributed hydrological model is a mathematical model used to simulate and predict the hydrological cycle process in the basin. Unlike the traditional centralized hydrological model, the distributed hydrological model divides the basin into multiple sub-areas or grid units, and considers the surface characteristics such as land use, topography, and soil type in each unit to more accurately simulate the spatiotemporal changes of the hydrological process in the basin. Through the distributed hydrological model, hydrological processes such as precipitation, evaporation, runoff, and groundwater recharge can be simulated and predicted more finely, so as to better understand the complexity of the basin hydrological cycle. These models usually use mathematical equations and calculation methods to describe processes such as soil moisture balance, surface runoff, and groundwater runoff to support hydrological decision-making and applications such as water resources management, flood warning, and drought monitoring.

[0075] In the above-mentioned water event risk prediction method, data extraction is performed based on at least the historical data of water events in the target area to obtain a triple that matches the preset prompt word; the triple is converted into structured data in the graph database; and a water event knowledge graph is generated based on at least the structured data in the graph database, which includes multiple dimensions such as water conservancy facilities, river basin information, and historical events. The water event knowledge graph can be dynamically updated and maintained to reflect the latest hydrological information and scheduling schemes. In this way, we can establish the influence relationship between the objects in the water event, associate the upstream and downstream water conditions and work conditions and other perception information, and form a systematic impact chain analysis. When it is necessary to perform risk analysis on the predicted area, the spatial data and terrain data of the predicted area in the target area can be determined, and the risk type of the predicted area can be determined based on the spatial data, terrain data and risk factors related to the spatial data and terrain data of the predicted area. The risk of the predicted area can be analyzed spatially, thereby determining the risk type of the predicted area at the spatial level. Combining the risk type and the data of the predicted area in the water event knowledge graph, the water event knowledge graph can be fully utilized to accurately predict the risk level of the predicted area.

[0076] In one embodiment, Figure 3 As shown, the method also includes:

[0077] S302: Determine precipitation objects in the target area according to historical precipitation data of the target area, a preset precipitation threshold and a preset area threshold.

[0078] Among them, the precipitation object may be an area with more precipitation per unit time in some embodiments of the present disclosure. The preset area threshold may be determined based on historical precipitation data or may be set based on the experience of those skilled in the art. In this embodiment, there is no restriction on how the preset area threshold is determined.

[0079] Specifically, historical precipitation data of the target area can be obtained from a variety of data sources, such as meteorological observation stations, satellite remote sensing data or numerical weather forecast models. Those skilled in the art can select the corresponding data source according to actual needs. Each data source has its specific advantages and applicable scenarios. For example, meteorological observation stations provide real-time observation data, satellite remote sensing can cover a wide geographical area, and numerical weather forecast models provide forecast data for future time periods. Compare the historical precipitation data with the precipitation threshold, determine the area greater than the precipitation threshold, and then determine the area less than the area threshold. The final area can be the precipitation object of the target area.

[0080] S304: Generate a relationship graph based on the precipitation object, the geographic spatial information of the target area and the water event knowledge graph.

[0081] Among them, geographic spatial information may include: topographic maps, land use maps, and water system maps, etc.

[0082] Specifically, the geographic spatial information of precipitation objects and target areas can be superimposed, and associations can be established with the water event knowledge graph to generate a relationship graph.

[0083] S306, determining the current precipitation intensity of the target area, and determining the associated step number based on the current precipitation intensity.

[0084] S308: Determine the precipitation impact of the current precipitation intensity in the target area based on the number of associated steps and the relationship map.

[0085] Specifically, affected by floods and rainfall levels, it is necessary to determine the number of indirect associations, that is, to determine the length of the impact chain: Event intensity assessment: Evaluate the event intensity based on the current precipitation intensity of the precipitation object (such as once in 10 years, once in 50 years, etc.). According to the judgment of the intensity of the precipitation object in the knowledge graph, such as once in 10 years, once in 50 years, etc., dynamically adjust the association range, that is, determine the number of indirect association steps. The more steps, the more related objects involved, and the wider the perception domain. Therefore, the number of association steps can be determined based on the precipitation intensity, and the affected objects can be found from the relationship graph through the number of association steps, so as to determine the precipitation impact of the current precipitation intensity in the target area.

[0086] In this embodiment, by using historical precipitation data, a preset precipitation threshold and a preset area threshold, precipitation objects in the target area can be accurately determined, so that the affected entities can be accurately identified and evaluated according to the precipitation intensity, providing a scientific basis for disaster prevention and mitigation.

[0087] In one embodiment, Figure 4 As shown, the step of determining the precipitation object in the target area according to the historical precipitation data of the target area, the preset precipitation threshold and the preset area threshold includes:

[0088] S402: Acquire historical precipitation data of the target area, and determine a precipitation threshold value according to the historical precipitation data.

[0089] S404: Determine a first area in the target area where historical precipitation data exceeds the precipitation threshold.

[0090] S406: Based on the area of ​​the first area and a preset area threshold, determine a second area in the first area that is smaller than the area threshold.

[0091] S408: Determine a precipitation object based on the second area, and set attribute information of the precipitation object, where the attribute information includes: precipitation amount, area, and precipitation time.

[0092] Specifically, according to the selected data source, historical precipitation data of the target area is collected from it. This may involve accessing the data interface, downloading data files, or using special software to extract data. Next, data formatting is performed. The collected precipitation data is converted into a unified format for subsequent processing and analysis. For example, the data can be formatted as historical precipitation data within a certain time period to ensure that the data structure is clear and consistent. Finally, data storage is performed. The formatted historical precipitation data is stored in a database or data warehouse. Then the historical precipitation data is analyzed to determine the precipitation threshold. This can be done by statistically analyzing the average value, distribution, or other statistical indicators of the historical precipitation data. For example, the average precipitation in the target area in the same time period in the past few years can be calculated as a reference for the threshold. According to the results of the historical precipitation data analysis, an appropriate precipitation threshold is set. The setting of the threshold can be adjusted according to actual needs and application scenarios. For example, a value that is a certain multiple of the average precipitation can be selected as the threshold. Then the first area in the target area where the historical precipitation data is greater than the precipitation threshold is determined. The area exceeding the precipitation threshold is marked for visual display and further analysis. For example, color markings can be used to distinguish, blue indicates that the precipitation does not exceed the precipitation threshold, and red indicates that the precipitation exceeds the set precipitation threshold. Calculate the area of ​​each area that exceeds the threshold. Set a minimum area standard, that is, the area threshold. Only areas in the first area that meet or exceed this standard are considered valid precipitation events. Filter out areas in the first area that are smaller than the area threshold, and filter out areas that are larger than the area threshold to obtain the second area. Define the second area as a precipitation object. Set relevant attributes for each precipitation object, such as precipitation, area, duration, etc. Record precipitation objects in a database or knowledge graph.

[0093] In this embodiment, by determining the precipitation object, the area that is easily affected by precipitation can be accurately analyzed.

[0094] In one embodiment, Figure 5 As shown, the relationship graph is generated based on the precipitation object, the geographic space information of the target area and the water event knowledge graph, including:

[0095] S502, determining the geographic spatial information of the target area, and superimposing the precipitation object with the geographic spatial information to obtain superimposed comprehensive information.

[0096] S504: Perform entity recognition and data matching on the superimposed comprehensive information to determine entities in the superimposed comprehensive information.

[0097] S506: Generate a relationship graph based on the entities in the superimposed comprehensive information and the precipitation objects existing in the water event knowledge graph.

[0098] Among them, AND operation is a logical operation symbol, usually represented by the symbol "&" or "∧". In computers, AND operation means that if both operands are true (usually represented by 1), the result is true (1), otherwise the result is false (0).

[0099] Specifically, a GIS data layer (Geographic Information System data layer) can be prepared. It is a basic concept in GIS, referring to the organization of map data or spatial data in a geographic information system. GIS data layer is the most basic spatial data representation in GIS, used to store and represent geographic spatial information. GIS data layers can contain various types of spatial data, such as geometric elements such as points, lines, and surfaces, as well as attribute information associated with these elements. Each GIS data layer is usually based on a specific theme or thematic classification, and represents geographical phenomena through vector data or raster data. GIS data layers can be different types of data organization forms such as map layers, terrain layers, and land use layers. Overlay the precipitation object with the prepared GIS data layer. This can be achieved using GIS software or programming tools to integrate information from different data layers to form more comprehensive spatial information, that is, overlay comprehensive information. Perform entity recognition based on AND operations. Within the coverage of the precipitation object, identify water conservancy facilities or geographical elements such as rivers, reservoirs, and levees that may be involved. This can be achieved through spatial analysis tools and algorithms, such as identifying water bodies or artificial buildings in a specific area. Perform data matching. Match the entities in the identified overlay comprehensive information with the corresponding data in the GIS data layer to ensure the accuracy of the identified entity location and attributes. This includes comparing and verifying the spatial location and attribute information of the entity to ensure the consistency and reliability of the data. Use the entities identified by GIS as query conditions and use the query function of the graph database to retrieve relevant nodes from the knowledge graph. Extract directly related objects in the query results, such as upstream and downstream river sections, connected reservoirs, etc. Integrate the extracted related objects with the precipitation objects to form a complete relationship graph.

[0100] In one embodiment, Figure 6 As shown, the method also includes:

[0101] S602: Determine a precipitation object in the target area, and determine a downstream water system object that matches the precipitation object.

[0102] S604: Tracking the impact path according to the downstream water system object matching the precipitation object to determine the downstream area affected by the precipitation object.

[0103] S606: Update the downstream area affected by the precipitation object in the water event knowledge graph.

[0104] Specifically, the river topological relationship in the knowledge graph can be used to track the downstream water system of the precipitation object.

[0105] First, analyze the topological structure of the rivers in the water event knowledge graph and identify the upstream and downstream relationships of the rivers. This can be achieved through graph analysis methods, such as depth-first search (DFS) or breadth-first search (BFS), to determine the flow relationship and connection pattern between rivers. Perform impact path tracing. According to the topological structure of the river, trace the possible downstream impact path of the precipitation object. This includes identifying downstream paths and connections that may be affected by the precipitation object along the river network. Assemble the downstream areas into the impact chain. Based on the results of the impact path tracing, identify the downstream areas that may be affected. These areas may include geographic areas extending along the flow direction of the river, which may be affected by the precipitation object. Perform data annotation. Annotate the identified downstream areas in the water event knowledge graph.

[0106] S608, determining the flood peak propagation time, and marking the time windows affected by different entities in the knowledge graph based on the flood peak propagation time.

[0107] The flood peak propagation time refers to the time required for a flood to reach the highest water level (flood peak) in a river or stream. When rainfall causes a sharp rise in river water levels, the peak of the flood will arrive at a certain moment, and this flood peak will gradually propagate downstream over time. The flood peak propagation time depends on the topography of the river, water flow velocity, flow, and hydrological conditions along the way.

[0108] Specifically, the peak propagation time can be estimated based on the calculation results of the historical hydrological model and the literature records. The historical hydrological model refers to the research that uses historical hydrological data and different model methods to simulate and predict hydrological processes. These models can be based on statistical methods, physical processes or machine learning techniques. The purpose of the historical hydrological model is usually to analyze past hydrological data, understand the regularity of hydrological processes, and establish a prediction model to estimate and predict future hydrological conditions. Construct a time delay model. According to the peak propagation time, a time delay model for peak propagation is established. The time of future peak propagation is estimated by the peak arrival time and propagation path in the historical data. Call the peak propagation event attributes of the river in the knowledge graph. Use the attribute information of the river peak propagation event recorded in the knowledge graph, such as the records of historical flood events and the results of model calculations, to further estimate the impact time. Finally, perform time window annotation. Annotate the time windows that may be affected by different downstream objects in the knowledge graph. These time windows reflect the flood propagation time range obtained based on model calculations and historical data analysis. The flood peak propagation event includes, for example, water system, river name, river section name, flood peak propagation time AB point, occurrence time (year, month, day), reliability, flood peak flow (m³ / s), occurrence time (year, month, day), affected area and other information.

[0109] S610, performing a superposition analysis on the precipitation object to determine a cross-impact area, and updating the water event knowledge graph based on the cross-impact area.

[0110] Specifically, additional assessments can be conducted on areas affected by multiple hydrological events such as extreme precipitation in a short period of time. The possible superimposed impacts of multiple precipitation objects on the same downstream area are usually considered, and the entities involved are merged into the same knowledge graph view for association analysis. For example, the data of multiple precipitation objects can be integrated to analyze their possible cross-impacts, and the data of multiple precipitation objects can be superimposed and analyzed to identify possible cross-impact areas, and the water event knowledge graph can be updated based on the cross-impact areas. The updated water event knowledge graph can be used to assess the risks of cross-impact areas and determine the severity of the superimposed impacts.

[0111] In one embodiment, Figure 7 As shown, the method also includes:

[0112] S702, determining the time series data of historical water events in the water event knowledge graph, and determining the periodic pattern of the time series data.

[0113] S704, identifying change points in the time series, determining time nodes based on the change points and the periodic pattern; and determining the category of the periodic pattern based on the time nodes.

[0114] Among them, historical water events can usually be historical extreme hydrological events. Historical extreme hydrological events refer to extreme hydrological events that occurred in a specific area in the past, such as extreme precipitation, floods, droughts, etc. These events usually have clear data or documentary records in local records and have a significant impact on the local hydrological cycle and climate conditions. The time series data of historical extreme hydrological events refers to the data series that records extreme hydrological events that occurred in the past period of time, such as precipitation, river water level, flood flow, etc. These data are usually arranged in chronological order to analyze and study the laws of hydrological changes and extreme events in a specific area.

[0115] Specifically, the time series data of historical extreme hydrological events can be extracted from the knowledge graph. Time series analysis techniques (such as autoregressive models and moving average models) are used to analyze the time series data to identify periodic patterns, and then the change point detection algorithm (PELT algorithm) is applied to identify significant change points in the time series. Based on significant change points and periodic patterns, key time nodes (such as flood peak arrival time and water level critical point) are determined. The identified time series patterns are classified according to key time nodes, such as rapid rise type, continuous high water level type, etc.

[0116] S706, configuring a hydrological event perception chain, and building a water event impact chain based on the hydrological event perception chain and the temporal relationship in the water event knowledge spectrum.

[0117] Among them, the perception chain can be an agent, which is based on machine learning and artificial intelligence technology and is specifically used to analyze hydrological event data, predict hydrological processes, assess risks, and formulate corresponding response measures. This agent can provide accurate hydrological event analysis and prediction results by learning and pattern recognition of a large amount of hydrological data. The water event impact chain may include: upstream rainfall, midstream water level changes, downstream floods, etc.

[0118] Specifically, the configured hydrological event analysis agent can understand and process hydrological domain knowledge and user needs, and can also proxy database query and other operations. Design query templates to extract relevant entities and relationships from the knowledge graph. Use graph algorithms (such as shortest path and full path analysis) to build event impact paths. Through the Agent call chain, gradually build a complete water event impact chain based on the temporal relationship of the knowledge graph.

[0119] S708, setting a plurality of precipitation scenarios for the forecast area, performing simulation forecasting based on the water event knowledge graph and the plurality of precipitation scenarios for the forecast area, and determining the time when each entity in the water event knowledge graph is affected by the precipitation scenario.

[0120] Specifically, the knowledge graph can also be used to estimate the time when the entity is affected, such as the flood peak crossing time and the subsequent highest water level period. For example, different precipitation scenarios can be set, such as precipitation events of different intensities and durations. The water event knowledge graph and multiple precipitation scenarios in the predicted area are used for simulation and prediction, and the hydraulic model operation records and historical scenario matching are used to infer the impact path and impact time under different scenarios.

[0121] In this embodiment, based on the water event knowledge graph, we can infer the impact of extreme hydrological events at different time points. It can help analyze the occurrence, development and results of extreme hydrological events, not only provide scientific disaster prevention and mitigation measures, but also support real-time monitoring and dynamic updates to ensure the timeliness and accuracy of the inference results.

[0122] In one embodiment, Figure 8 As shown, the risk type of the predicted area is determined based on the spatial data, topographic data of the predicted area and the risk factors related to the spatial data and topographic data, including:

[0123] S802: superimpose the spatial data to obtain multi-layer spatial data.

[0124] S804: Perform a first analysis on the multi-layer spatial data according to the levels to obtain a first analysis result.

[0125] S806: Perform a second analysis on the terrain data to obtain a second analysis result.

[0126] S808: Determine risk factors that affect the risk based on the first analysis result and the second analysis result.

[0127] S810, calculating a risk index of the predicted area based on the risk factor, and determining a risk type of the predicted area based on the risk index and a preset risk threshold interval.

[0128] Specifically, multiple layers of spatial data are integrated for comprehensive analysis. Spatial data at different levels are superimposed to create a multi-layer spatial data set. This is achieved through GIS software or spatial databases to ensure spatial consistency, scale consistency and precision alignment of the data set. The spatial data is first analyzed layer by layer, considering the impact of different levels on flood propagation. The integrated multi-layer spatial data is analyzed according to the level to obtain the first analysis result. Starting from the terrain data, the terrain data is secondly analyzed to analyze the impact of terrain on flood diffusion and water accumulation; then combined with land use data, the impact of different land types on flood water retention capacity and permeability is analyzed; finally, considering the urbanization degree data, the impact of urban expansion on flood runoff and drainage system is analyzed to obtain the second analysis result. Then, based on the first analysis results and the second analysis results, the risk factors affecting the risk are defined, including terrain characteristics (DEM), distribution and performance of flood control facilities (such as embankments, reservoirs), population density, land use type (such as urbanization degree, farmland, forest cover), historical breach, historical dangerous sections, etc. A comprehensive risk score is calculated for the predicted area. The comprehensive risk score can be calculated based on the defined risk factors using a weighted or non-weighted method to calculate the risk index of each spatial unit. This can be achieved through mathematical models or expert judgment to ensure the objectivity and accuracy of the score. The threshold interval for risk level division is set, and the risk threshold interval is used to determine the risk type corresponding to the risk index, such as low risk, medium risk, and high risk.

[0129] In this embodiment, based on the constructed knowledge graph, the impact of extreme hydrological events in different spatial ranges can be inferred, supporting real-time monitoring and dynamic updating, and improving the timeliness and accuracy of the inference results.

[0130] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0131] Based on the same inventive concept, the disclosed embodiment also provides a water event risk prediction device for implementing the water event risk prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more water event risk prediction device embodiments provided below can refer to the limitations of the water event risk prediction method above, and will not be repeated here.

[0132] In one embodiment, Fig. 9 As shown, a water event risk prediction device 900 is provided, including: a knowledge graph acquisition module 902, a risk type determination module 904 and a risk level prediction module 906, wherein:

[0133] The knowledge graph acquisition module 902 is used to acquire a pre-built water event knowledge graph; the water event knowledge graph includes: extracting data based on at least historical water event data in the target area to obtain triples matching pre-set prompt words; converting the triples into structured data in a graph database; generating a water event knowledge graph based on at least the structured data in the graph database;

[0134] The risk type determination module 904 is used to determine the spatial data and terrain data of the prediction area in the target area, and determine the risk type of the prediction area based on the spatial data, terrain data of the prediction area and risk factors related to the spatial data and terrain data;

[0135] The risk level prediction module 906 is used to predict the risk level of the prediction area based on the risk type of the prediction area and the data of the prediction area in the water event knowledge graph.

[0136] In one embodiment of the device, the device further comprises:

[0137] The impact determination module is used to determine the precipitation objects in the target area according to the historical precipitation data of the target area, the preset precipitation threshold and the preset area threshold; generate a relationship map based on the precipitation objects, the geographic spatial information of the target area and the water event knowledge map; determine the current precipitation intensity of the target area, and determine the number of associated steps based on the current precipitation intensity; determine the precipitation impact of the current precipitation intensity of the target area based on the number of associated steps and the relationship map.

[0138] In one embodiment of the device, the impact determination module includes: a precipitation object determination module, which is used to obtain historical precipitation data of a target area and determine a precipitation threshold based on the historical precipitation data; determine a first area in the target area where the historical precipitation data exceeds the precipitation threshold; based on the area area of ​​the first area and a preset area threshold, determine a second area in the first area that is smaller than the area threshold; determine a precipitation object based on the second area, and set attribute information of the precipitation object, wherein the attribute information includes: precipitation amount, area, and precipitation time.

[0139] In one embodiment of the device, the impact determination module includes: a relationship graph generation module, which is used to determine the geographic spatial information of the target area, superimpose the precipitation object with the geographic spatial information to obtain superimposed comprehensive information; perform entity recognition and data matching on the superimposed comprehensive information to determine the entities in the superimposed comprehensive information; and generate a relationship graph based on the entities in the superimposed comprehensive information and the precipitation objects existing in the water event knowledge graph.

[0140] In one embodiment of the device, the device also includes: a first update module, used to determine the precipitation object in the target area, and determine the downstream water system object matching the precipitation object; perform influence path tracking based on the downstream water system object matching the precipitation object, and determine the downstream area affected by the precipitation object; and update the downstream area affected by the precipitation object in the water event knowledge graph.

[0141] The labeling module is used to determine the flood peak propagation time, and label the time windows affected by different entities in the knowledge graph based on the flood peak propagation time.

[0142] The second updating module is used to perform superposition analysis on the precipitation object, determine the cross-impact area, and update the water event knowledge graph based on the cross-impact area.

[0143] In one embodiment of the device, the device further comprises:

[0144] The category determination module is used to determine the time series data of historical water events in the water event knowledge graph and determine the periodic pattern of the time series data. The change points in the time series are identified, and the time nodes are determined based on the change points and the periodic pattern; and the category of the periodic pattern is determined based on the time nodes.

[0145] The influence chain construction module is used to configure the hydrological event perception chain and to construct the water event influence chain based on the temporal relationship between the hydrological event perception chain and the water event knowledge spectrum.

[0146] A simulation module is used to set multiple precipitation scenarios in the forecast area, perform simulation prediction based on the water event knowledge graph and the multiple precipitation scenarios in the forecast area, and determine the time when each entity in the water event knowledge graph is affected by the precipitation scenario.

[0147] In one embodiment of the device, the risk type determination module 904 is also used to superimpose the spatial data to obtain multi-layer spatial data; perform a first analysis on the multi-layer spatial data according to the hierarchy to obtain a first analysis result; perform a second analysis on the terrain data to obtain a second analysis result; determine the risk factors affecting the risk based on the first analysis result and the second analysis result; calculate the risk index of the predicted area based on the risk factors, and determine the risk type of the predicted area based on the risk index and a pre-set risk threshold interval.

[0148] Each module in the above-mentioned water event risk prediction device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0149] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig.10 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store water event knowledge graph data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a water event risk prediction method is implemented.

[0150] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Fig.11As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a water event risk prediction method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0151] Those skilled in the art will understand that Fig.10 The structure shown in or 11 is merely a block diagram of a partial structure related to the scheme of the present disclosure, and does not constitute a limitation on the computer device to which the scheme of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0152] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0153] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0154] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0155] It should be noted that the historical data on water events, historical precipitation data, etc. involved in this disclosure are all information and data authorized by users or fully authorized by all parties.

[0156] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present disclosure may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided by the present disclosure may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but are not limited to this.

[0157] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] The above-described embodiments only express several implementation methods of the present disclosure, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present disclosure, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the attached claims.

Claims

1. A water event risk prediction method, characterized in that: The method comprises: Obtaining a pre-constructed water event knowledge graph; the water event knowledge graph comprises: extracting data based on at least historical water event data in a target area to obtain triples matching pre-set prompt words; converting the triples into structured data in a graph database; generating a water event knowledge graph based at least on the structured data in the graph database; Determining spatial data and topographic data of a prediction area in the target area, and determining a risk type of the prediction area based on the spatial data, topographic data of the prediction area and risk factors associated with the spatial data and topographic data, including: Integrate multi-layer spatial data, in which spatial data at different levels are superimposed to create a multi-layer spatial data set; The first analysis of spatial data is carried out layer by layer, considering the impact of different layers on flood propagation, and the first analysis results are obtained; Starting from topographic data, we analyze the impact of topography on flood diffusion and water accumulation; combining land use data, we analyze the impact of different land types on flood water retention and permeability; considering urbanization data, we analyze the impact of urban expansion on flood runoff and drainage systems, and obtain the second analysis result; Based on the results of the first and second analyses, define the risk factors that affect the risk, including topographic features, the distribution and performance of flood control facilities, population density, land use type, historical breaches, and historical dangerous sections; Calculate a risk index for each spatial unit based on the defined risk factors using either a weighted or non-weighted approach; and Set the threshold interval for risk level classification, and use the risk threshold interval to determine the risk type corresponding to the risk index; Predicting the risk level of the predicted area based on the risk type of the predicted area and the data of the predicted area in the water event knowledge graph includes: Obtain entity information and attribute information of the predicted area from the water event knowledge graph, including coordinate elevation and flow boundary conditions, and summarize them as input data of the distributed hydrological model according to the input instructions, and calculate the water level and flow distribution information within the time range; Read the historical flood records and dispatch rule attributes of the region in the water event knowledge graph to obtain the flood risk impact level of the predicted region, thereby obtaining the risk level of the predicted region; Among them, the distributed hydrological model is a mathematical model used to simulate and predict the hydrological cycle process in a watershed.

2. The method according to claim 1, characterized in that The method further comprises: Determine precipitation objects in the target area based on historical precipitation data of the target area, a preset precipitation threshold, and a preset area threshold; Generate a relationship graph based on the precipitation object, the geospatial information of the target area and the water event knowledge graph; Determining a current precipitation intensity in the target area, and determining an associated step number based on the current precipitation intensity; The precipitation impact of the current precipitation intensity in the target area is determined based on the associated step number and the relationship map.

3. The method according to claim 2, characterized in that: The step of determining the precipitation object in the target area according to the historical precipitation data of the target area, the preset precipitation threshold and the preset area threshold includes: Acquire historical precipitation data of the target area, and determine a precipitation threshold value according to the historical precipitation data; Determine a first area in the target region where historical precipitation data exceeds the precipitation threshold; Based on the area of ​​the first area and a preset area threshold, determining a second area in the first area that is smaller than the area threshold; A precipitation object is determined based on the second area, and attribute information of the precipitation object is set, where the attribute information includes: precipitation amount, area, and precipitation time.

4. The method according to claim 2, characterized in that: The generating of a relationship graph based on the precipitation object, the geospatial information of the target area and the water event knowledge graph includes: Determine the geographic spatial information of the target area, and superimpose the precipitation object with the geographic spatial information to obtain superimposed comprehensive information; Performing entity recognition and data matching on the superimposed comprehensive information to determine entities in the superimposed comprehensive information; A relationship graph is generated based on entities in the superimposed comprehensive information and the precipitation objects present in the water event knowledge graph.

5. The method according to claim 1, characterized in that The method further comprises: Determine the precipitation object in the target area, and determine the downstream water system object matching the precipitation object; Tracking the impact path according to the downstream water system object matching the precipitation object to determine the downstream area affected by the precipitation object; Updating the downstream area affected by the precipitation object in the water event knowledge graph; Determine the flood peak propagation time, and mark the time windows affected by different entities in the knowledge graph based on the flood peak propagation time; Overlay analysis is performed on the precipitation object to determine the cross-impact area, and the water event knowledge graph is updated based on the cross-impact area.

6. The method according to claim 1, characterized in that The method further comprises: Determining time series data of historical water events in the water event knowledge graph, and determining a periodic pattern of the time series data; Identifying a change point in the time series, determining a time node based on the change point and the periodic pattern; and determining a category of the periodic pattern based on the time node; Configure a hydrological event perception chain, and build a water event impact chain based on the hydrological event perception chain and the temporal relationship in the water event knowledge graph; A plurality of precipitation scenarios are set for the forecast area, simulation prediction is performed based on the water event knowledge graph and the plurality of precipitation scenarios for the forecast area, and the time when each entity in the water event knowledge graph is affected by the precipitation scenario is determined.

7. The method according to claim 1, characterized in that The step of determining the risk type of the predicted area based on the spatial data, topographic data of the predicted area and risk factors related to the spatial data and topographic data comprises: Superimposing the spatial data to obtain multi-layer spatial data; Performing a first analysis on the multi-layer spatial data according to the levels to obtain a first analysis result; Performing a second analysis on the terrain data to obtain a second analysis result; Determining risk factors affecting risk based on the first analysis result and the second analysis result; A risk index of the predicted area is calculated based on the risk factor, and a risk type of the predicted area is determined based on the risk index and a preset risk threshold interval.

8. A water event risk prediction device, characterized in that: The device comprises: A knowledge graph acquisition module is used to acquire a pre-built water event knowledge graph; the water event knowledge graph includes: extracting data based on at least historical water event data in a target area to obtain triples matching pre-set prompt words; converting the triples into structured data in a graph database; generating a water event knowledge graph based on at least the structured data in the graph database; The risk type determination module is used to determine the spatial data and terrain data of the predicted area in the target area, and determine the risk type of the predicted area based on the spatial data, terrain data of the predicted area and risk factors related to the spatial data and terrain data, including: Integrate multi-layer spatial data, in which spatial data at different levels are superimposed to create a multi-layer spatial data set; The first analysis of spatial data is carried out layer by layer, considering the impact of different layers on flood propagation, and the first analysis results are obtained; Starting from topographic data, we analyze the impact of topography on flood diffusion and water accumulation; combining land use data, we analyze the impact of different land types on flood water retention and permeability; considering urbanization data, we analyze the impact of urban expansion on flood runoff and drainage systems, and obtain the second analysis result; Based on the results of the first and second analyses, define the risk factors that affect the risk, including topographic features, the distribution and performance of flood control facilities, population density, land use type, historical breaches, and historical dangerous sections; Calculate the risk index for each spatial unit based on the defined risk factors using either a weighted or non-weighted approach; and Set the threshold interval for risk level classification, and use the risk threshold interval to determine the risk type corresponding to the risk index; A risk level prediction module is used to predict the risk level of the predicted area based on the risk type of the predicted area and the data of the predicted area in the water event knowledge graph, including: Obtain entity information and attribute information of the predicted area from the water event knowledge graph, including coordinate elevation and flow boundary conditions, and summarize them as input data of the distributed hydrological model according to the input instructions, and calculate the water level and flow distribution information within the time range; Read the historical flood records and dispatch rule attributes of the region in the water event knowledge graph to obtain the flood risk impact level of the predicted region, thereby obtaining the risk level of the predicted region; Among them, the distributed hydrological model is a mathematical model used to simulate and predict the hydrological cycle process in a watershed.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.