Natural disaster risk assessment method and device, electronic equipment and storage medium
By dividing the region into grids and training a graph convolutional neural network model, the problem of existing technologies being unable to consider the impact of adjacent counties and districts has been solved, thus achieving a more accurate natural disaster risk assessment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA PING AN PROPERTY INSURANCE CO LTD
- Filing Date
- 2021-11-29
- Publication Date
- 2026-07-31
AI Technical Summary
Existing natural disaster assessment methods cannot effectively consider the impact of neighboring counties, resulting in low accuracy of risk assessment.
By dividing the region into grids, constructing an undirected graph, and using a graph convolutional neural network model, combined with precipitation data and ground elevation data for iterative training, a disaster assessment model is established to conduct risk assessments for multiple districts and counties.
It improves the accuracy of natural disaster risk assessment and enables more comprehensive prediction of disaster impacts across multiple districts and counties.
Smart Images
Figure CN115936289B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to artificial intelligence technology, and more particularly to a method, apparatus, electronic device, and storage medium for natural disaster risk assessment. Background Technology
[0002] Against the backdrop of global climate change, natural disasters (such as rainstorms) are occurring more frequently, with a significant increasing trend in both intensity and scale, causing enormous direct and indirect losses to the economy and society. If disaster severity can be assessed in advance, disaster relief resources can be allocated proactively to reduce losses and mitigate risks.
[0003] However, current natural disaster assessment methods can only make single predictions based on data from a single district or county, ignoring the impact of neighboring districts or counties on the current district or county, resulting in low accuracy of risk assessment. Summary of the Invention
[0004] This invention provides a method, apparatus, electronic device, and storage medium for natural disaster risk assessment, the main purpose of which is to improve the accuracy of natural disaster risk assessment.
[0005] A preset area containing multiple districts and counties is divided into multiple grid areas using a grid of preset size; Obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; Construct an undirected graph by treating each district / county within the region as a node to obtain the first district / county node graph; Obtain general feature data for each district / county with district / county as the feature granularity, and construct the node features of the corresponding nodes in the first district / county node map for each historical precipitation event based on the grid area corresponding to each district / county using the general feature data, the precipitation value, and the ground elevation value, to obtain the corresponding second district / county node map. The historical precipitation risk value of each district and county at each historical precipitation event is obtained. The graph convolutional neural network model is iteratively trained using the historical precipitation risk value and the node map of the second district and county to obtain the disaster assessment model. When the region encounters a precipitation event, the expected precipitation value and current ground elevation value of each grid area in the region are obtained at the time of the precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, the node features of each node in the first district / county node map are constructed to obtain the current district / county node map. Based on the current county-level node map and the disaster assessment model, a disaster risk assessment is conducted for each county in the region to obtain the assessment results.
[0006] Optionally, the process of dividing a preset area containing multiple districts and counties into multiple grid regions using a grid of a preset size includes: A grid array is formed using the grids of a predetermined number; The center of the preset region is determined as the center of the central grid of the grid array; The region corresponding to each grid in the grid array is defined as the grid region.
[0007] Optionally, constructing an undirected graph by treating each district / county within the region as a node to obtain a first district / county node graph includes: The aforementioned districts and counties are designated as nodes; Connect the nodes corresponding to the adjacent districts and counties in the region to obtain the first district and county node map.
[0008] Optionally, the step of constructing the node features of the corresponding nodes in the first county-level node map for each historical precipitation event based on the general feature data, the precipitation values, and the ground elevation values, to obtain the corresponding second county-level node map, includes: Select the precipitation values of all grid areas corresponding to the district / county at each historical precipitation event, and determine the average or median of all selected precipitation values, or the score of a first preset proportion of all selected precipitation values, as the historical precipitation data of the corresponding district / county. The ground elevation values of all grid areas corresponding to the district / county at each historical precipitation event are selected, and the average or median of all selected ground elevation values, or the score of the second preset proportion among all selected ground elevation values, are determined as the historical district / county topographic data of the corresponding district / county. The historical precipitation data, historical topographic data, and general feature data of each district / county for each historical precipitation event are combined as the node features of the corresponding node in the first district / county node, thus obtaining the second district / county node map corresponding to that historical precipitation event.
[0009] Optionally, the step of iteratively training the graph convolutional neural network model using the historical precipitation risk values and the second county-level node map to obtain a disaster assessment model includes: The historical precipitation risk value of each district / county for each historical precipitation event is used as the label of the corresponding node in the second district / county node graph corresponding to that historical precipitation event, thus obtaining training samples; Use all the aforementioned training samples as the training set; The disaster assessment model is obtained by iteratively training the graph convolutional neural network model using the training set.
[0010] Optionally, the step of iteratively training the graph convolutional neural network model using the training set to obtain the disaster assessment model includes: Step A: Randomly select a training sample from the training set and input it into the graph convolutional neural network model to obtain the predicted label value corresponding to each node of the training sample; Step B: Based on the predicted label value and label corresponding to each node of the training sample, calculate the loss function value using a preset loss function; Step C: When the loss function value is greater than or equal to a preset threshold, change the model parameters of the graph convolutional neural network model and return to step A; Step D: When the loss function value is less than a preset threshold, stop training and output the disaster assessment model.
[0011] Optionally, before obtaining the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within the preset time interval, the method further includes: Use the current time as the right endpoint of the interval; The time interval is constructed using a preset time period as the interval length.
[0012] To address the above problems, the present invention also provides a natural disaster risk assessment device, the device comprising: The node graph construction module is used to divide a preset area containing multiple districts and counties into multiple grid areas using a grid of preset size; obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; construct an undirected graph with each district and county in the area as a node to obtain a first district and county node graph; obtain general feature data of each district and county with district and county as the feature granularity, and construct the node features of the corresponding nodes in the first district and county node graph for each historical precipitation event based on the grid area corresponding to each district and county using the general feature data, the precipitation value, and the ground elevation value to obtain a corresponding second district and county node graph; The model training module is used to obtain the historical precipitation risk value of each district and county at each historical precipitation event, and to iteratively train the graph convolutional neural network model using the historical precipitation risk value and the node map of the second district and county to obtain the disaster assessment model. The risk assessment module is used to obtain the expected precipitation value and current ground elevation value of each grid area in the region when the region encounters a precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, the module constructs the node features of each node in the first county node map to obtain the current county node map. Based on the current county node map and the disaster assessment model, the module performs a disaster risk assessment on each county in the region to obtain the assessment results.
[0013] To address the above problems, the present invention also provides an electronic device, the electronic device comprising: Memory, storing at least one computer program; and The processor executes the computer program stored in the memory to implement the natural disaster risk assessment method described above.
[0014] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the natural disaster risk assessment method described above.
[0015] In this embodiment of the invention, each district / county within the region is used as a node to construct an undirected graph, resulting in a first district / county node graph. General feature data for each district / county is obtained, using the district / county as the feature granularity. Based on the grid area corresponding to each district / county, the node features of the corresponding nodes in the first district / county node graph for each historical precipitation event are constructed using the general feature data, precipitation values, and ground elevation values, resulting in a corresponding second district / county node graph. In this embodiment, the second district / county node graph is constructed based on different district / county relationships and features. This second district / county node graph integrates the features of each district / county and adjacent districts / counties, resulting in richer feature dimensions and more accurate risk assessment. Therefore, the natural disaster risk assessment method, device, electronic equipment, and readable storage medium proposed in this embodiment improve the accuracy of natural disaster risk assessment. Attached Figure Description
[0016] Figure 1 A flowchart illustrating a natural disaster risk assessment method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the construction of a first county-level node map in a natural disaster risk assessment method provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of a module of a natural disaster risk assessment device provided in an embodiment of the present invention; Figure 4 A schematic diagram of the internal structure of an electronic device for implementing a natural disaster risk assessment method according to an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0017] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
[0018] This invention provides a method for natural disaster risk assessment. The executing entity of the natural disaster risk assessment method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application embodiment: a server, a terminal, etc. In other words, the natural disaster risk assessment method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster, etc. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.
[0019] Reference Figure 1 The flowchart shown is a schematic diagram of a natural disaster risk assessment method provided in an embodiment of the present invention. In this embodiment, the natural disaster risk assessment method includes: S1. Divide the preset area containing multiple districts and counties into grids using a grid of preset size to obtain multiple grid areas.
[0020] Optionally, the preset area mentioned in this embodiment of the invention is an area that needs to be assessed for natural disaster risks, which can be a country, province, city, or other region.
[0021] In detail, in this embodiment of the invention, since precipitation events often do not occur in a single district or county within a region, and the precipitation data is not identical across all districts and counties, this embodiment of the invention utilizes a grid of preset size to divide a preset region containing multiple districts and counties into multiple grid regions. These individual grid regions allow for better evaluation of the region's precipitation data. The precipitation event described in this embodiment of the invention can be one of several forms of precipitation, such as rain or snowfall.
[0022] Optionally, in embodiments of the present invention, the grids of a preset number are used to form a grid array; the center of the preset region is determined as the center of the grid array's central grid; and the region corresponding to each grid in the grid array is determined as the grid region.
[0023] Specifically, in this embodiment of the invention, the preset size grid is a fixed-size square grid, such as 1km. 1km, 3km The grid array is 3km in size and n. An array consisting of n identical grids.
[0024] For example: the grid array is 5 The grid array is 5 rows and 5 columns. Each row has five identical rectangular grids connected in sequence, and each column also has five identical rectangular grids connected in sequence. The center grid is the grid in the third row and third column of the grid array. The center coordinates of the region are determined as the center point of the center grid. The area covered by the grid is further determined according to the grid size to obtain the corresponding grid region. Each grid in the grid array corresponds to a grid region.
[0025] S2. Obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within the preset time interval.
[0026] In this embodiment of the invention, to ensure the reference value of historical precipitation events, the time factor needs to be considered; the historical precipitation events are each precipitation event that has occurred in the preset area within a preset time interval, and the precipitation value is the precipitation data corresponding to the current historical precipitation event in the grid area; the ground elevation value is the ground elevation of the grid center point of the current historical precipitation event in the grid area. In this embodiment of the invention, the precipitation data can be queried from the historical grid forecast database of a meteorological station, and the geographical elevation value can be queried from a land and resources information database.
[0027] Optionally, in this embodiment of the invention, before obtaining the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval, the method further includes: The time interval is obtained by constructing the interval using the current time as the right endpoint and a preset time period as the interval length.
[0028] S3. Construct an undirected graph by treating each district / county within the region as a node, and obtain the first district / county node graph.
[0029] In this embodiment of the invention, since the precipitation risk of each district / county is often affected by neighboring districts / counties, an undirected graph is constructed by treating all the districts / counties as nodes to obtain the first district / county node graph.
[0030] In detail, in this embodiment of the invention, an undirected graph is constructed using all the districts and counties as nodes to obtain a first district and county node graph, including: The districts and counties are used as nodes; the nodes of adjacent districts and counties in the region are connected to obtain the first district and county node map.
[0031] For example: There are three districts / counties: A, B, and C. District / county A is adjacent to district / county B, and district / county C is adjacent to district / county B. The corresponding first district / county node diagram would be as follows: Figure 2 As shown.
[0032] In another embodiment of the present invention, by utilizing the high throughput characteristics of blockchain nodes, the first district / county node map can be stored in the blockchain nodes, thereby improving the efficiency of data retrieval.
[0033] S4. Obtain general feature data for each district / county with district / county as the feature granularity, and construct the node features of the corresponding nodes in the first district / county node map for each historical precipitation event based on the grid area corresponding to each district / county, using the general feature data, the precipitation value, and the ground elevation value, to obtain the corresponding second district / county node map.
[0034] Optionally, the general feature data in this embodiment of the invention is feature data with districts and counties as the granularity, including but not limited to: the total number of insured policies, policy distribution density, total premium, total number of historical flood-damaged vehicle accidents, historical claim amount, etc.; population, GDP, etc., and the features in the general data are all numerical values of features with districts and counties as the feature granularity.
[0035] Furthermore, in this embodiment of the invention, the node features of the corresponding nodes in the first county-level node map for each historical precipitation event are constructed based on the general feature data, the precipitation values, and the ground elevation values, to obtain the corresponding second county-level node map, including: Precipitation features are extracted by selecting precipitation values from all grid areas corresponding to the district / county at each historical precipitation event to obtain the historical precipitation data of the corresponding district / county. Optionally, in this embodiment of the invention, all grid areas corresponding to each region are all grid areas within the administrative region of that district or county.
[0036] In this embodiment of the invention, precipitation values of all grid areas corresponding to the district / county at each historical precipitation event are selected, and 70% of the selected precipitation values are determined as the historical precipitation data of the corresponding district / county.
[0037] Optionally, in another embodiment of the present invention, the average, median, 50% score, etc. of all selected precipitation values can be determined as the historical precipitation data of the corresponding district / county.
[0038] Furthermore, in this embodiment of the invention, the ground elevation values of all grid areas corresponding to the district / county at each historical precipitation event are selected for terrain feature extraction to obtain the historical terrain data of the corresponding district / county.
[0039] Specifically, in this embodiment of the invention, the ground elevation values of all grid areas corresponding to the district / county at each historical precipitation event are selected, and 70% of the selected ground elevation values are determined as the historical district / county topographic data.
[0040] Alternatively, in another embodiment of the present invention, the average, median, 50% score, etc. of all selected precipitation values can be determined as the historical topographic data of the corresponding district / county.
[0041] Furthermore, in this embodiment of the invention, the historical precipitation data, historical topographic data, and general feature data of each district and county in each historical precipitation event are combined as the node features of the corresponding node in the first district and county node to obtain the second district and county node map corresponding to the historical precipitation event.
[0042] S5. Obtain the historical precipitation risk value for each district / county at each historical precipitation event, and use the historical precipitation risk value and the node graph of the second district / county to iteratively train the graph convolutional neural network model to obtain the disaster assessment model.
[0043] Optionally, in this embodiment of the invention, the historical precipitation risk value is the level or value of the severity of the disaster in the district / county assessed at each historical precipitation time. For example, the historical precipitation risk value can be the accident rate of vehicle insurance in the district / county; the graph convolutional neural network model is the type of the initial artificial intelligence model trained to obtain the disaster assessment model.
[0044] In detail, in this embodiment of the invention, the historical precipitation risk value of each district / county for each historical precipitation event is used as the label of the corresponding node in the second district / county node graph corresponding to the historical precipitation event, so as to obtain training samples.
[0045] In this embodiment of the invention, each historical precipitation event corresponds to a training sample. All the training samples are used as a training set, and the pre-constructed graph convolutional neural network model is iteratively trained using the training set to obtain the disaster assessment model.
[0046] Optionally, in this embodiment of the invention, all the training samples are used as a training set to iteratively train the pre-constructed graph convolutional neural network model to obtain the disaster assessment model, including: Step A: Randomly select a training sample from the training set and input it into the graph convolutional neural network model to obtain the predicted label value corresponding to each node of the training sample; Step B: Based on the predicted label value and label corresponding to each node of the training sample, calculate the loss function value using a preset loss function; Step C: When the loss function value is greater than or equal to a preset threshold, change the model parameters of the graph convolutional neural network model and return to step A; Step D: When the loss function value is less than a preset threshold, stop training and output the disaster assessment model.
[0047] Optionally, in this embodiment of the invention, the focal loss function is selected as the loss function during training.
[0048] Optionally, in another embodiment of the present invention, a preset number of training samples may be selected as the training set, wherein the preset number of samples is the integer value of 80% of the total number of samples.
[0049] S6. When the region encounters a precipitation event, obtain the expected precipitation value and current ground elevation value of each grid area in the region at the time of the precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, construct the node features of each node in the first district / county node map to obtain the current district / county node map.
[0050] In detail, when the area described in this embodiment of the invention is experiencing a precipitation event, since precipitation is a continuous process, it is necessary to obtain the expected precipitation value, that is, the expected total precipitation of the grid area of this precipitation event. The expected precipitation value can be obtained from the grid forecast database of a meteorological station.
[0051] Furthermore, in this embodiment of the invention, the expected precipitation values of all grid areas corresponding to the district / county are selected, and feature extraction is performed on all selected expected precipitation values to obtain the corresponding expected precipitation data of the district / county.
[0052] In this embodiment of the invention, the current ground elevation values of all grid areas corresponding to the district / county are obtained, and the selected current ground elevation values are used for feature extraction to obtain the corresponding current terrain data of the district / county.
[0053] The feature extraction steps described in this embodiment are the same as those described above, and will not be repeated here.
[0054] Furthermore, in this embodiment of the invention, the expected precipitation data, current topographic data, and general feature data corresponding to each district / county are combined as the node features of the corresponding nodes in the first district / county node map to obtain the current district / county node map.
[0055] S7. Based on the current county node map and the disaster assessment model, conduct a disaster risk assessment for each county in the region to obtain the assessment results.
[0056] In detail, in this embodiment of the invention, the current county / district node map is input into the disaster assessment model to obtain the disaster assessment value of the county / district corresponding to each node in the current county / district node map.
[0057] Furthermore, disaster risk assessment is conducted using the disaster assessment value. If the disaster assessment value is greater than a preset disaster threshold, the disaster assessment result for the corresponding district / county is high risk; if the disaster assessment threshold is less than the preset disaster threshold, the disaster assessment result for the corresponding district / county is low risk.
[0058] like Figure 3 The diagram shown is a functional block diagram of the natural disaster risk assessment device of the present invention.
[0059] The natural disaster risk assessment device 100 of the present invention can be installed in an electronic device. Depending on the functions implemented, the natural disaster risk assessment device may include a node graph construction module 101, a model training module 102, and a risk assessment module 103. The module described in this invention may also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.
[0060] In this embodiment, the functions of each module / unit are as follows: The node graph construction module 101 is used to divide a preset area containing multiple districts and counties into multiple grid areas using a grid of a preset size; obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; construct an undirected graph with each district and county in the area as a node to obtain a first district and county node graph; obtain general feature data of each district and county with district and county as the feature granularity, and construct the node features of the corresponding nodes in the first district and county node graph for each historical precipitation event based on the grid area corresponding to each district and county using the general feature data, the precipitation value, and the ground elevation value to obtain a corresponding second district and county node graph; The model training module 102 is used to obtain the historical precipitation risk value of each district and county at each historical precipitation event, and to iteratively train the graph convolutional neural network model using the historical precipitation risk value and the node graph of the second district and county to obtain the disaster assessment model. The risk assessment module 103 is used to obtain the expected precipitation value and current ground elevation value of each grid area in the region when the region encounters a precipitation event, construct the node features of each node in the first county node map based on the expected precipitation value, the current ground elevation value, and the general feature data, and obtain the current county node map; and perform disaster risk assessment on each county in the region based on the current county node map and the disaster assessment model to obtain the assessment result.
[0061] In detail, each module in the natural disaster risk assessment device 100 described in this embodiment of the invention adopts the same characteristics as described above during use. Figure 1 The method used is the same as the natural disaster risk assessment method described in the article and can produce the same technical effect, so it will not be elaborated here.
[0062] like Figure 4 The diagram shown is a schematic representation of the electronic device used in this invention to implement the natural disaster risk assessment method.
[0063] The electronic device may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program, such as a natural disaster risk assessment program, stored in the memory 11 and capable of running on the processor 10.
[0064] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code of a natural disaster risk assessment program, but also to temporarily store data that has been output or will be output.
[0065] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules (such as natural disaster risk assessment programs) stored in the memory 11, and calls data stored in the memory 11 to perform various functions of the electronic device and process data.
[0066] The communication bus 12 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The communication bus 12 is configured to enable communication between the memory 11 and at least one processor 10, etc. For ease of illustration, only one thick line is used in the figure, but this does not indicate that there is only one bus or one type of bus.
[0067] Figure 4 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 4 The structure shown does not constitute a limitation on the electronic device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.
[0068] For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault classification circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.
[0069] Optionally, the communication interface 13 may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device and other electronic devices.
[0070] Optionally, the communication interface 13 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.
[0071] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.
[0072] The natural disaster risk assessment program stored in the memory 11 of the electronic device is a combination of multiple computer programs, which, when run in the processor 10, can achieve the following: A preset area containing multiple districts and counties is divided into multiple grid areas using a grid of preset size; Obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; Construct an undirected graph by treating each district / county within the region as a node to obtain the first district / county node graph; Obtain general feature data for each district / county with district / county as the feature granularity, and construct the node features of the corresponding nodes in the first district / county node map for each historical precipitation event based on the grid area corresponding to each district / county using the general feature data, the precipitation value, and the ground elevation value, to obtain the corresponding second district / county node map. The historical precipitation risk value of each district and county at each historical precipitation event is obtained. The graph convolutional neural network model is iteratively trained using the historical precipitation risk value and the node map of the second district and county to obtain the disaster assessment model. When the region encounters a precipitation event, the expected precipitation value and current ground elevation value of each grid area in the region are obtained at the time of the precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, the node features of each node in the first district / county node map are constructed to obtain the current district / county node map. Based on the current county-level node map and the disaster assessment model, a disaster risk assessment is conducted for each county in the region to obtain the assessment results.
[0073] Specifically, the processor 10's implementation method of the above-mentioned computer program can be found in [reference needed]. Figure 1The descriptions of the relevant steps in the corresponding embodiments are not repeated here.
[0074] Furthermore, if the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable medium can be non-volatile or volatile. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).
[0075] Embodiments of the present invention may also provide a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following: A preset area containing multiple districts and counties is divided into multiple grid areas using a grid of preset size; Obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; Construct an undirected graph by treating each district / county within the region as a node to obtain the first district / county node graph; Obtain general feature data for each district / county with district / county as the feature granularity, and construct the node features of the corresponding nodes in the first district / county node map for each historical precipitation event based on the grid area corresponding to each district / county using the general feature data, the precipitation value, and the ground elevation value, to obtain the corresponding second district / county node map. The historical precipitation risk value of each district and county at each historical precipitation event is obtained. The graph convolutional neural network model is iteratively trained using the historical precipitation risk value and the node map of the second district and county to obtain the disaster assessment model. When the region encounters a precipitation event, the expected precipitation value and current ground elevation value of each grid area in the region are obtained at the time of the precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, the node features of each node in the first district / county node map are constructed to obtain the current district / county node map. Based on the current county-level node map and the disaster assessment model, a disaster risk assessment is conducted for each county in the region to obtain the assessment results.
[0076] Furthermore, the computer's usable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store the operating system, applications required for at least one function, etc.; and the data storage area may store data created based on the use of blockchain nodes, etc.
[0077] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.
[0078] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0079] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.
[0080] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.
[0081] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.
[0082] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.
[0083] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.
[0084] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The term "second class" is used to indicate names and does not indicate any specific order.
[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims
1. A method for natural disaster risk assessment, characterized by, The method includes: A preset area containing multiple districts and counties is divided into multiple grid areas using a grid of preset size; Obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; Construct an undirected graph by treating each district / county within the region as a node to obtain the first district / county node graph; Obtain general feature data for each district / county with district / county as the feature granularity, and construct the node features of the corresponding nodes in the first district / county node map for each historical precipitation event based on the grid area corresponding to each district / county using the general feature data, the precipitation value, and the ground elevation value, to obtain the corresponding second district / county node map. The historical precipitation risk value of each district and county at each historical precipitation event is obtained. The graph convolutional neural network model is iteratively trained using the historical precipitation risk value and the node map of the second district and county to obtain the disaster assessment model. When the region encounters a precipitation event, the expected precipitation value and current ground elevation value of each grid area in the region are obtained at the time of the precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, the node features of each node in the first district / county node map are constructed to obtain the current district / county node map. Based on the current county-level node map and the disaster assessment model, a disaster risk assessment is conducted for each county in the region to obtain the assessment results; The process of constructing node features for corresponding nodes in the first county-level node map for each historical precipitation event using the general feature data, precipitation values, and ground elevation values, to obtain the corresponding second county-level node map, includes: Select the precipitation values of all grid areas corresponding to the district / county at each historical precipitation event, and determine the average or median of all selected precipitation values, or the score of a first preset proportion of all selected precipitation values, as the historical precipitation data of the corresponding district / county. The ground elevation values of all grid areas corresponding to the district / county at each historical precipitation event are selected, and the average or median of all selected ground elevation values, or the score of the second preset proportion among all selected ground elevation values, are determined as the historical district / county topographic data of the corresponding district / county. The historical precipitation data, historical topographic data, and general feature data of each district / county for each historical precipitation event are combined as the node features of the corresponding node in the first district / county node map to obtain the second district / county node map corresponding to that historical precipitation event.
2. The natural disaster risk assessment method of claim 1, wherein, The process involves dividing a preset area containing multiple districts and counties into multiple grid regions using a grid of a preset size, including: A grid array is formed using the grids of a predetermined number; The center of the preset region is determined as the center of the central grid of the grid array; The region corresponding to each grid in the grid array is defined as the grid region.
3. The natural disaster risk assessment method of claim 1, wherein, The step of constructing an undirected graph by treating each district / county within the region as a node to obtain the first district / county node graph includes: The aforementioned districts and counties are designated as nodes; Connect the nodes corresponding to the adjacent districts and counties in the region to obtain the first district and county node map.
4. The natural disaster risk assessment method as described in claim 1, characterized in that, The process of iteratively training a graph convolutional neural network model using the historical precipitation risk values and the second county-level node map to obtain a disaster assessment model includes: The historical precipitation risk value of each district / county for each historical precipitation event is used as the label of the corresponding node in the second district / county node graph corresponding to that historical precipitation event, thus obtaining training samples; Use all the aforementioned training samples as the training set; The disaster assessment model is obtained by iteratively training the graph convolutional neural network model using the training set.
5. The natural disaster risk assessment method as described in claim 4, characterized in that, The step of iteratively training the graph convolutional neural network model using the training set to obtain the disaster assessment model includes: Step A: Randomly select a training sample from the training set and input it into the graph convolutional neural network model to obtain the predicted label value corresponding to each node of the training sample; Step B: Based on the predicted label value and label corresponding to each node of the training sample, calculate the loss function value using a preset loss function; Step C: When the loss function value is greater than or equal to a preset threshold, change the model parameters of the graph convolutional neural network model and return to step A; Step D: When the loss function value is less than a preset threshold, stop training and output the disaster assessment model.
6. The natural disaster risk assessment method as described in any one of claims 1 to 5, characterized in that, Before obtaining the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within the preset time interval, the method further includes: Use the current time as the right endpoint of the interval; The preset time interval is constructed using a preset time period as the interval length.
7. A natural disaster risk assessment device, used to implement the natural disaster risk assessment method as described in any one of claims 1 to 6, characterized in that, include: The node graph construction module is used to divide a preset area containing multiple districts and counties into multiple grid areas using a grid of preset size; Obtain the precipitation value and ground elevation value of each grid area for each historical precipitation event in the preset area within a preset time interval; construct an undirected graph with each district / county in the area as a node to obtain a first district / county node graph; obtain general feature data of each district / county with district / county as the feature granularity, and construct the node features of the corresponding nodes in the first district / county node graph for each historical precipitation event based on the grid area corresponding to each district / county using the general feature data, the precipitation value, and the ground elevation value to obtain a corresponding second district / county node graph; The model training module is used to obtain the historical precipitation risk value of each district and county at each historical precipitation event, and to iteratively train the graph convolutional neural network model using the historical precipitation risk value and the node map of the second district and county to obtain the disaster assessment model. The risk assessment module is used to obtain the expected precipitation value and current ground elevation value of each grid area in the region when the region encounters a precipitation event. Based on the expected precipitation value, the current ground elevation value, and the general feature data, the module constructs the node features of each node in the first county node map to obtain the current county node map. Based on the current county node map and the disaster assessment model, the module performs a disaster risk assessment on each county in the region to obtain the assessment results.
8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory that is communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the natural disaster risk assessment method as described in any one of claims 1 to 6.
9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the natural disaster risk assessment method as described in any one of claims 1 to 6.