Disaster prediction method, device and equipment based on rainfall forecast data, medium and computer program product
By combining hydrodynamic model and machine learning model, a composite disaster sensitivity analysis chart is generated, and a single disaster warning in the existing technology ignores the linkage effect of multiple disasters is solved, and a comprehensive prediction and risk assessment of multiple disasters caused by rainfall is achieved. , improve the effectiveness of disaster prevention and mitigation measures and the scientific nature of early warning systems.
Patent Information
- Application Number
- CN202410926552.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-11
- Publication Date
- 2024-11-08
AI Technical Summary
The existing disaster prediction method based on rainfall forecasting mainly considers a single disaster, ignoring the linkage effect between multiple disasters that may be caused by rainfall, resulting in insufficient awareness of other potential risks, insufficient disaster prevention preparations, and uneven resource allocation. Impact the comprehensiveness and effectiveness of the early warning system.
By combining hydrodynamic model and machine learning-based geological disaster model, the AutoGluon model is used to construct a geological disaster machine learning model, collect rainfall forecast data, and generate flood sensitivity analysis maps, geological disaster sensitivity analysis maps and heavy rain warning maps, combining these maps Generate a composite disaster sensitivity analysis chart to comprehensively predict various potential disasters caused by rainfall.
A comprehensive risk assessment and prediction of various disasters caused by rainfall has been achieved, the effectiveness of disaster prevention and mitigation measures has been improved, the scientificity and accuracy of the early warning system has been enhanced, and a comprehensive disaster prevention and mitigation reference has been provided.
Smart Images

Figure CN118917457A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of natural disaster prediction, and in particular to a disaster prediction method, apparatus, device, medium and computer program product based on rainfall forecast data. Background Art
[0002] While high-resolution weather forecast models are developing rapidly, the technology of using weather forecasts to predict urban waterlogging or geological disasters has also developed to a relatively mature level. These technologies are based on rainfall forecast data, combined with terrain, land use, drainage system and other data, and generate urban waterlogging or geological disaster risk maps through physical mechanisms. These maps can show potential risk points and be used to plan response measures.
[0003] However, the existing methods of disaster forecasting based on rainfall basically only consider a single disaster, such as predicting urban waterlogging points or geological disaster areas through rainfall, which may affect the comprehensiveness and effectiveness of the early warning system. Specifically, rainfall may cause multiple disasters, such as waterlogging, floods, geological disasters, etc. Considering only a single disaster will ignore the linkage effect between these disasters. Early warning of a single disaster may lead to insufficient public awareness of other potential risks and failure to make comprehensive disaster prevention preparations. Early warning systems for single disasters may lead to uneven resource allocation, neglect prevention of other disasters, and lead to loopholes and deficiencies in disaster prevention and mitigation measures when formulating urban planning or emergency response plans.
[0004] Therefore, how to predict various potential disasters caused by rainfall based on rainfall forecast data is an urgent problem that needs to be solved. Summary of the invention
[0005] The present application provides a disaster prediction method based on rainfall forecast data, which can generate a composite disaster sensitivity analysis diagram by combining a hydrodynamic model, a geological disaster model constructed based on machine learning, and rainfall forecast data, and predict a variety of potential disasters caused by rainfall.
[0006] In order to achieve the above object, the present invention adopts the following technical scheme:
[0007] In a first aspect, the present invention provides a disaster prediction method based on rainfall forecast data, the method comprising:
[0008] Collect historical hydrological data and historical geological disaster data;
[0009] constructing a hydrodynamic model based on the historical hydrological data;
[0010] Use the AutoGluon model to train the historical geological disaster data and build a geological disaster machine learning model;
[0011] Collect rainfall forecast data;
[0012] Generating a rainstorm warning map based on the rainfall forecast data;
[0013] Inputting the rainfall forecast data into the hydrodynamic model and the geological disaster machine learning model respectively, to obtain a flood sensitivity analysis diagram and a geological disaster sensitivity analysis diagram respectively;
[0014] The flood sensitivity analysis map, the geological disaster sensitivity analysis map and the rainstorm warning map are combined to generate a composite disaster sensitivity analysis map.
[0015] In a preferred example of the present application, it can be further configured to include:
[0016] Before using the rainfall forecast data, the rainfall forecast data is downscaled using nearest neighbor interpolation.
[0017] In a preferred example of the present application, it can be further configured that the historical geological disaster data includes:
[0018] Historical rainfall, altitude, slope, aspect, curvature, normalized vegetation cover, distance from disaster site to road, distance from disaster site to water system, and land cover type.
[0019] In a preferred example of the present application, it can be further configured to include:
[0020] A rainfall-runoff machine learning model was built based on AutoGluon, and it was calibrated using historical rainfall and runoff data.
[0021] Inputting the rainfall forecast data into the rainfall runoff machine learning model to obtain predicted runoff;
[0022] The predicted runoff is input into the hydrodynamic model.
[0023] In a preferred example of the present application, it can be further configured that a hydrodynamic model is constructed based on the historical hydrological data, including:
[0024] Combined with historical hydrological data, a hydrodynamic model was constructed based on the digital elevation model, and the water levels of historical flood hydrological stations were used to calibrate the parameters of this model.
[0025] In a preferred example of the present application, it can be further configured that the historical hydrological data includes:
[0026] Water body mask, channel width, and channel depth information.
[0027] In a second aspect, the present application provides a disaster prediction device based on rainfall forecast data, the device comprising:
[0028] Data collection module, used to collect historical hydrological data and historical geological disaster data; collect rainfall forecast data;
[0029] A model building module is used to build a hydrodynamic model based on the historical hydrological data; use the AutoGluon model to train the historical geological disaster data to build a geological disaster machine learning model;
[0030] A map generation module is used to generate a rainstorm warning map based on the rainfall forecast data; input the rainfall forecast data into the hydrodynamic model and the geological disaster machine learning model respectively to obtain a flood sensitivity analysis map and a geological disaster sensitivity analysis map respectively;
[0031] The composite module is used to combine the flood sensitivity analysis map, the geological disaster sensitivity analysis map and the rainstorm warning map to generate a composite disaster sensitivity analysis map.
[0032] In a third aspect, the present application provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the disaster prediction method based on rainfall forecast data as described in any one of the above items are implemented.
[0033] In a fourth aspect, the present application provides a computer-readable storage medium having a program stored thereon, wherein when the program is executed by a processor, a disaster prediction method based on rainfall forecast data as described in any one of the above items is implemented.
[0034] In a fifth aspect, the present application provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the disaster prediction method based on rainfall forecast data as described in any one of the above items.
[0035] In summary, compared with the prior art, the technical solution provided in the embodiment of the present application has at least the following beneficial effects:
[0036] The present application provides a disaster prediction method based on rainfall forecast data, which integrates multiple disaster risks, including heavy rain, floods and geological disasters, to obtain a composite disaster sensitivity analysis diagram, which comprehensively displays the various disaster risks of the region under specific rainfall conditions, providing a comprehensive reference for disaster prevention and mitigation. The present application provides a comprehensive composite disaster risk assessment. This comprehensive analysis can more accurately predict the various disasters caused by rainfall and enhance the effectiveness of disaster prevention and mitigation. It uses the machine learning algorithm AutoGluon model to construct a geological disaster machine learning model. The model analyzes a large amount of historical disaster data to identify the complex relationship between rainfall and geological disasters, thereby improving the accuracy of disaster prediction. In addition, by integrating meteorological data, geographic information data, geological data and historical disaster data, a comprehensive temporal and spatial overlay analysis is performed to improve the scientific nature of the early warning system. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 A flowchart of disaster prediction based on rainfall forecast data is provided for one embodiment of the present application.
[0038] Figure 2 A conceptual diagram of a hydrodynamic model for disaster prediction based on rainfall forecast data provided in one embodiment of the present application.
[0039] Figure 3 An ROC curve diagram of a geological disaster machine learning model for disaster prediction based on rainfall forecast data provided in one embodiment of the present application.
[0040] Figure 4 A graph of model testing results of a geological disaster machine learning model for disaster prediction based on rainfall forecast data provided in one embodiment of the present application.
[0041] Figure 5 A module diagram of a disaster prediction device based on rainfall forecast data provided in one embodiment of the present application. DETAILED DESCRIPTION
[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0043] In one embodiment of the present application, a disaster prediction method based on rainfall forecast data is provided. Figure 1 As shown, the method includes:
[0044] S100: Collect historical hydrological data and historical geological disaster data;
[0045] Specifically, the historical hydrological data include the flow and water level of historical flood hydrological stations, water body masks, river width and river depth information, etc. The historical geological disaster data include rainfall, altitude, slope, aspect, curvature, normalized vegetation coverage, normalized vegetation index, the distance from the disaster point to the road, the distance from the disaster point to the water system, disaster point data, water system, road and land cover type, etc. Among them, altitude, slope, aspect, curvature, etc. belong to DEM (Digital Elevation Model) data, collected from MERIT DEM, soil cover type is collected from Pengcheng Laboratory, roads and water systems are collected from OpenStreetMap, rainfall is collected from ERA5, normalized vegetation index is collected from MOD13Q1, and disaster point data comes from internal statistics of Sun Yat-sen University.
[0046] S200: constructing a hydrodynamic model based on the historical hydrological data;
[0047] Specifically, based on the digital elevation model, the hydrodynamic model is constructed in combination with historical hydrological data, namely water body mask, river width, and river depth information, and the water levels of historical flood hydrological stations are used for parameter calibration. Figure 2 shown.
[0048] S300: Using the AutoGluon model to train the historical geological disaster data to build a geological disaster machine learning model;
[0049] Specifically, the effect of the geological disaster machine learning model is as follows: Figure 3 As shown, Figure 3 is the ROC curve of the model. The test results of the geological disaster machine learning model are as follows Figure 4 shown.
[0050] S400: collecting rainfall forecast data;
[0051] Specifically, the rainfall forecast data is the rainfall prediction data in the weather forecast.
[0052] S500: generating a rainstorm warning map based on the rainfall forecast data;
[0053] Specifically, a rainstorm warning map for a certain area is drawn based on the rainfall forecast data, and the rainstorm warning map indicates the rainstorm risk levels in different areas.
[0054] S600: Inputting the rainfall forecast data into the hydrodynamic model and the geological disaster machine learning model respectively to obtain a flood sensitivity analysis diagram and a geological disaster sensitivity analysis diagram respectively;
[0055] Specifically, the rainfall forecast data is input into the hydrodynamic model, and the model generates a flood sensitivity analysis diagram by simulating river flooding, which shows the flood risk status in the study area under the forecast rainfall conditions. The geological disaster sensitivity analysis diagram shows the risk level of geological disasters in the study area under the forecast rainfall conditions.
[0056] S700: Generate a composite disaster sensitivity analysis map by combining the flood sensitivity analysis map, the geological disaster sensitivity analysis map and the rainstorm warning map.
[0057] Specifically, the composite disaster sensitivity analysis map comprehensively displays the various disaster risks of a region under specific rainfall conditions, providing a comprehensive reference for disaster prevention and mitigation.
[0058] In this embodiment, by integrating multiple disaster risks, including rainstorms, floods, and geological disasters, a composite disaster sensitivity analysis map is obtained. The map comprehensively shows the multiple disaster risks of the region under specific rainfall conditions, providing a comprehensive reference for disaster prevention and mitigation. Moreover, this comprehensive analysis can more accurately predict the various disasters caused by rainfall and enhance the effectiveness of disaster prevention and mitigation. By using the machine learning algorithm AutoGluon model to build a geological disaster machine learning model, the model analyzes a large amount of historical disaster data, identifies the complex relationship between rainfall and geological disasters, and improves the accuracy of disaster prediction. In addition, by integrating meteorological data, geographic information data, geological data, and historical disaster data, a comprehensive temporal and spatial superposition analysis is performed to improve the scientific nature of the early warning system.
[0059] In some embodiments, it also includes:
[0060] Before using the rainfall forecast data, the rainfall forecast data is downscaled using nearest neighbor interpolation.
[0061] In this embodiment, by performing spatial downscaling processing on the forecast rainfall data, a high-precision disaster sensitivity analysis grid is obtained, thereby improving the accuracy of generating a composite disaster sensitivity analysis map.
[0062] In some embodiments, the historical geological disaster data includes:
[0063] Historical rainfall, altitude, slope, aspect, curvature, normalized vegetation coverage, distance from disaster site to road, distance from disaster site to water system, and land cover type.
[0064] In this embodiment, the accuracy of generating a composite disaster sensitivity analysis diagram is improved by comprehensively considering multi-source data.
[0065] In some embodiments, it also includes:
[0066] A rainfall-runoff machine learning model was built based on AutoGluon, and it was calibrated using historical rainfall and runoff data.
[0067] Inputting the rainfall forecast data into the rainfall runoff machine learning model to obtain predicted runoff;
[0068] The predicted runoff is input into the hydrodynamic model.
[0069] In specific implementation, the rainfall-runoff model is used to simulate the runoff generated by the forecast rainfall in different river basins, and the runoff is used as the input of the hydrodynamic model.
[0070] In this embodiment, the prediction results of the machine learning model are used as the input of the physical mechanism model to improve the accuracy and reliability of the overall prediction.
[0071] In some embodiments, constructing a hydrodynamic model based on the historical hydrological data includes:
[0072] Combined with historical hydrological data, a hydrodynamic model was constructed based on the digital elevation model, and the water levels of historical flood hydrological stations were used to calibrate the parameters of this model.
[0073] In this embodiment, the accuracy and reliability of the overall prediction are improved
[0074] In some embodiments, the historical hydrological data includes:
[0075] Water body mask, channel width, and channel depth information.
[0076] In this embodiment, the accuracy of generating a composite disaster sensitivity analysis diagram is improved by comprehensively considering multi-source data.
[0077] This application also provides a disaster prediction device based on rainfall forecast data, see Figure 5 As shown, the device comprises:
[0078] The data collection module 100 is used to collect historical hydrological data and historical geological disaster data; collect rainfall forecast data;
[0079] A model building module 200 is used to build a hydrodynamic model based on the historical hydrological data; use the AutoGluon model to train the historical geological disaster data to build a geological disaster machine learning model;
[0080] A map generation module 300 is used to generate a rainstorm warning map based on the rainfall forecast data; input the rainfall forecast data into the hydrodynamic model and the geological disaster machine learning model respectively to obtain a flood sensitivity analysis map and a geological disaster sensitivity analysis map respectively;
[0081] The composite module 400 is used to combine the flood sensitivity analysis map, the geological disaster sensitivity analysis map and the rainstorm warning map to generate a composite disaster sensitivity analysis map.
[0082] The functional implementation of each module in the above-mentioned disaster prediction device based on rainfall forecast data corresponds to the steps in the above-mentioned disaster prediction method embodiment based on rainfall forecast data, and its functions and implementation processes are no longer described one by one here.
[0083] The present application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the disaster prediction method based on rainfall forecast data as described in any of the above embodiments are implemented.
[0084] The present application also provides a computer-readable storage medium, on which a program is stored, wherein the computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive and / or a memory stick, etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The working process, working details and technical effects of the computer-readable storage medium provided in this embodiment can be found in the above embodiment of a disaster prediction method based on rainfall forecast data, which will not be repeated here.
[0085] The application also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the disaster prediction method based on rainfall forecast data as described in any of the above embodiments.
[0086] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods 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 memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0087] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not 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. The above-mentioned embodiments only express several implementation methods of the present application, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present application, several deformations and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent of this application shall be based on the attached claims.
Claims
1. A disaster prediction method based on rainfall forecast data, characterized in that: include: Collect historical hydrological data and historical geological disaster data; constructing a hydrodynamic model based on the historical hydrological data; Use the AutoGluon model to train the historical geological disaster data and build a geological disaster machine learning model; Collect rainfall forecast data; Generating a rainstorm warning map based on the rainfall forecast data; Inputting the rainfall forecast data into the hydrodynamic model and the geological disaster machine learning model respectively, to obtain a flood sensitivity analysis diagram and a geological disaster sensitivity analysis diagram respectively; The flood sensitivity analysis map, the geological disaster sensitivity analysis map and the rainstorm warning map are combined to generate a composite disaster sensitivity analysis map.
2. The disaster prediction method based on rainfall forecast data according to claim 1, characterized in that: Also includes: Before using the rainfall forecast data, the rainfall forecast data is downscaled using nearest neighbor interpolation.
3. The disaster prediction method based on rainfall forecast data according to claim 1, characterized in that: The historical geological disaster data include: Historical rainfall, altitude, slope, aspect, curvature, normalized vegetation cover, distance from disaster site to road, distance from disaster site to water system, and land cover type.
4. The disaster prediction method based on rainfall forecast data according to claim 1, characterized in that: Also includes: A rainfall-runoff machine learning model was built based on AutoGluon, and it was calibrated using historical rainfall and runoff data. Inputting the rainfall forecast data into the rainfall runoff machine learning model to obtain predicted runoff; The predicted runoff is input into the hydrodynamic model.
5. The disaster prediction method based on rainfall forecast data according to claim 1, characterized in that: Constructing a hydrodynamic model based on the historical hydrological data includes: Combined with historical hydrological data, a hydrodynamic model was constructed based on the digital elevation model, and the water levels of historical flood hydrological stations were used to calibrate the parameters of this model.
6. The disaster prediction method based on rainfall forecast data according to claim 5, characterized in that: The historical hydrological data include: Water body mask, channel width, and channel depth information.
7. A disaster prediction device based on rainfall forecast data, characterized in that: include: Data collection module, used to collect historical hydrological data and historical geological disaster data; Collect rainfall forecast data; A model building module is used to build a hydrodynamic model based on the historical hydrological data; use the AutoGluon model to train the historical geological disaster data to build a geological disaster machine learning model; A map generation module is used to generate a rainstorm warning map based on the rainfall forecast data; input the rainfall forecast data into the hydrodynamic model and the geological disaster machine learning model respectively to obtain a flood sensitivity analysis map and a geological disaster sensitivity analysis map respectively; The composite module is used to combine the flood sensitivity analysis map, the geological disaster sensitivity analysis map and the rainstorm warning map to generate a composite disaster sensitivity analysis map.
8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the disaster prediction method based on rainfall forecast data according to any one of claims 1 to 6 are implemented.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, wherein when the program is executed by a processor, the disaster prediction method based on rainfall forecast data according to any one of claims 1 to 6 is implemented.
10. A computer program product comprising computer instructions, characterized in that: The computer instructions implement the steps of the method according to claims 1 to 6 when executed by a processor.
