Flood risk assessment early warning method and system based on big data analysis

Abstract

The invention provides a flood risk assessment early warning method and system based on big data analysis, and relates to the field of natural disaster risk assessment. The big data analysis-based flood risk assessment and early warning method comprises the following steps of S1, collecting basic information of a regional river through big data; s2, real-time information of a regional river is collected through the Internet of Things; the invention further provides a flood risk assessment early warning system which comprises a big data acquisition module, an online monitoring module, a data processing module, a model construction module, a virtual operation module and a self-learning module. According to the scheme, the water level change condition in a prediction period is finally realized according to the predicted meteorological information, and the prediction, evaluation and early warning of flood risks in an analysis area range are realized; and reliable data support is provided for effective prediction, evaluation and early warning of natural disasters.

Description

Technical Field

[0002] The present invention relates to the field of natural disaster risk assessment, and in particular to a flood risk assessment and early warning method and system based on big data analysis. Background Art

[0003] Flood disaster warning is to analyze and process flood disaster data through machine learning technology to achieve risk assessment and accurate early warning of flood disasters, so as to take effective preventive measures in advance and reduce the losses caused by flood disasters.

[0004] In the relevant prior art, publication number: CN118643957A, discloses a flood disaster warning method and system based on machine learning, the method includes: data collection, data preprocessing, flood disaster data balancing, building a flood disaster risk assessment model and real-time flood disaster warning; introducing local density calculation weighted Euclidean distance, identifying and removing noise data, screening boundary data based on mean and standard deviation, extracting reference samples based on clustering, and synthesizing reference data; proposing a gated time convolution unit to extract the temporal characteristics of flood disasters, designing an adaptive graph convolution unit and an adaptive graph attention unit, and a gated fusion unit to extract the spatial characteristics of flood disasters, and obtaining the spatiotemporal characteristics of flood disasters based on regulatory factors.

[0005] The existing flood disaster warning methods are mainly limited by the frequency of data updates; the system functions are independent of each other; the data are relatively fixed, and the analysis and application are based on specific areas and specific types of data. The system cannot be flexibly applied to random areas with flood risks; it relies solely on hydrological and hydrodynamic analysis for calculations, and can only be verified based on historical flood data; and how to predict and evaluate flood disasters within the forecast period while monitoring precipitation information in real time requires further research.

[0006] Therefore, it is necessary to provide a flood risk assessment and early warning method based on big data analysis to solve the above technical problems. Summary of the invention

[0007] The present invention provides a flood risk assessment and early warning method based on big data analysis, which solves the problem in the related technology of how to predict and evaluate flood disasters within the prediction period while monitoring precipitation information in real time, which needs further research.

[0008] In order to solve the above technical problems, the present invention provides a flood risk assessment and early warning method based on big data analysis, comprising the following steps: Step S1, collecting basic information of regional rivers through big data; Step S2, collecting real-time information of regional rivers through the Internet of Things; Step S3, building a virtual operation model in combination with regional information and real-time information, for simulating a real regional river precipitation scenario according to actual meteorological information and rainfall information; Step S4, iteratively updating the virtual operation model according to the forecast data to generate flood conditions within the forecast period; Step S5: Feedback flood warning information based on the predicted flood situation within the period.

[0009] Preferably, the basic information in step S1 is mainly the required terrain, hydrology and gate distribution data obtained from the Internet and intranet sources.

[0010] Preferably, the real-time information in step S2 is mainly data on weather, precipitation and water level distribution monitored in real time by online equipment within the regional river range.

[0011] Preferably, in step S3, the virtual operation model constructs a self-learning model scope, builds the model based on historical data, and continuously updates the model environment with real-time data for continuous model iteration; the virtual operation model fits the real-time information, and continuously improves the analysis accuracy of the virtual operation model during the system operation cycle.

[0012] Preferably, the prediction data in step S4 includes meteorological forecast information, which is used to simulate precipitation operation within the prediction period according to the meteorological forecast information, generate precipitation and regional river water level changes within the prediction period, and be used for virtual prediction of flood disaster occurrence.

[0013] Preferably, the prediction data also includes water level change data, which is the change of water level in regional rivers within unit time; when there is a large error between meteorological information and actual rainfall information, the water level change is simulated within the prediction period based on the water level change data, and the regional river water level change within the prediction period is generated for virtual prediction of flood disasters.

[0014] Preferably, the flood warning information in step S5 is transmitted via remote communication by site maintenance personnel to provide early warning of flood disasters, so as to urge people in the area to evacuate in time.

[0015] Preferably, corresponding alarm equipment is provided on the real-time monitoring equipment within the regional river range. When the alarm equipment receives flood disaster warning information, it automatically alarms within the regional river range, so that the alarm equipment can automatically alarm, facilitating simultaneous alarms at different monitoring points.

[0016] The present invention also provides a flood risk assessment and early warning system, which is used to provide support for the operation of the flood risk assessment and early warning method based on big data analysis. The flood risk assessment and early warning system includes a big data acquisition module, an online monitoring module, a data processing module, a model building module, a virtual operation module and a self-learning module; The big data acquisition module and the online monitoring module are both connected to the data processing module by signal, the big data acquisition module is used to collect the basic information, and the online monitoring module is used to collect the real-time information; The data processing module is used to clean and classify the collected data, identify the data quality, and assign weights; The data construction module and the virtual operation module are both connected to the data processing module by signal. The model construction module is used to construct the virtual operation model. The virtual operation module is used to add real weather and river distribution conditions and can adjust the depth of the river according to the actual rainfall. The self-learning module is signal-connected to the data processing module. The self-learning module automatically adjusts parameters based on a large number of model calculation results and corresponding disaster loss results, and continuously iterates and adjusts the corresponding results through machine learning generative adversarial networks and original data.

[0017] Preferably, it also includes a solution pushing module, which is used to push the flood warning information; the solution pushing module is also used to adaptively provide solution information during disaster warning and push the solution information.

[0018] Compared with related technologies, the flood risk assessment and early warning method based on big data analysis provided by the present invention has the following beneficial effects: By collecting historical data and building a virtual operation model, the meteorological information and water level information are merged into the virtual operation model to perform real-time simulation operations to obtain regional river water level changes and weather changes that are more in line with reality. During the simulation operation, the water level changes within the prediction period can be realized based on the predicted meteorological information, which can be used to analyze the prediction, assessment and early warning of flood risks in the region; providing reliable data support for the effective prediction, assessment and early warning of natural disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0020] Figure 1 A flow chart of a flood risk assessment and early warning method based on big data analysis provided by the present invention; Figure 2 A system block diagram of a flood risk assessment and early warning method based on big data analysis provided by the present invention; Figure 3 A warning flow chart of the flood risk assessment and warning method based on big data analysis provided by the present invention; Figure 4 A block diagram of the flood risk assessment and early warning system provided by the present invention; Figure 5 for Figure 4 A block diagram of the data processing module shown; Figure 6 for Figure 4 The block diagram of the big data acquisition module shown; Figure 7 for Figure 4 The block diagram of the online monitoring module shown; Figure 8 for Figure 4 A block diagram of the model building blocks shown; Fig. 9 for Figure 4 A block diagram of the virtual operation module shown; Fig.10 for Figure 4 The block diagram of the solution push module shown.

[0021] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] The present invention provides a flood risk assessment and early warning method based on big data analysis.

[0024] Please refer to Figures 1 to 3 In a first embodiment of the present invention, a flood risk assessment and early warning method based on big data analysis includes the following steps: Step S1, collecting basic information of regional rivers through big data; Step S2, collecting real-time information of regional rivers through the Internet of Things; Step S3, building a virtual operation model in combination with regional information and real-time information, for simulating a real regional river precipitation scenario according to actual meteorological information and rainfall information; Step S4, iteratively updating the virtual operation model according to the forecast data to generate flood conditions within the forecast period; Step S5: Feedback flood warning information based on the predicted flood situation within the period.

[0025] By collecting historical data and building a virtual operation model, the meteorological information and water level information are merged into the virtual operation model to perform real-time simulation operations to obtain regional river water level changes and weather changes that are more in line with reality. During the simulation operation, the water level changes within the prediction period can be realized based on the predicted meteorological information, which can be used to analyze the prediction, assessment and early warning of flood risks in the region; providing reliable data support for the effective prediction, assessment and early warning of natural disasters.

[0026] Due to the different carrying water depths at different points within the regional river range, when the predicted water level exceeds the carrying range, it is determined that a flood disaster will occur in the river range at that point; when the predicted water level does not exceed the carrying range, it is determined that a flood disaster will not occur in the river range at that point.

[0027] Specifically, the basic information in step S1 is mainly the required terrain, hydrology and gate distribution data obtained from the Internet and intranet sources, providing support for the simulation of the virtual operation model with geographical data, regional river distribution data and gate opening and closing distribution data.

[0028] Specifically, the real-time information in step S2 is mainly data on weather, precipitation and water level distribution monitored in real time by online equipment within the regional river range.

[0029] Meteorological information is weather forecast information provided by real-time network data, which includes real-time rainfall information and predicted rainfall information; precipitation information is precipitation information measured in real time by rainfall measuring equipment, and water level information is water level information of regional rivers measured in real time by water level measuring equipment.

[0030] Specifically, in step S3, the virtual operation model constructs a self-learning model scope, builds the model based on historical data, and continuously updates the model environment with real-time data for continuous model iteration; the virtual operation model fits real-time information, and continuously improves the analysis accuracy of the virtual operation model during the system operation cycle.

[0031] According to the selected area, various data and information required for model construction are obtained from various data sources and knowledge graphs; the collected information is cleaned and classified, the data quality is identified, and the weight is given; the model is automatically selected based on the analysis area, and the initialization information of relevant parameters and data is given to automatically build a virtual operation model.

[0032] Specifically, the prediction data in step S4 includes meteorological forecast information, which is used to simulate precipitation operation within the prediction period according to the meteorological forecast information, generate precipitation and regional river water level changes within the prediction period, and is used for virtual prediction of flood disasters.

[0033] Specifically, the prediction data also includes water level change data, which is the change of water level in regional rivers within unit time; when there is a large error between meteorological information and actual rainfall information (the precipitation error ratio is much larger than a preset range), the water level change is simulated within the prediction period based on the water level change data, and the regional river water level change within the prediction period is generated for virtual prediction of flood disasters.

[0034] Specifically, in step S5, the flood warning information is transmitted via remote communication by the site maintenance personnel to provide early warning of flood disasters, so as to urge people in the area to evacuate in time.

[0035] Specifically, corresponding alarm equipment is installed on the real-time monitoring equipment within the regional river range. When the alarm equipment receives flood disaster warning information, it automatically alarms within the range of the regional river, so that the alarm equipment can automatically alarm, facilitating simultaneous alarms at different monitoring points.

[0036] In this embodiment, the analysis results with a higher degree of fit are ultimately output to the user in the form of graphs, tables and reports.

[0037] The flood risk assessment and early warning method based on big data analysis provided in this embodiment has the following beneficial effects: Through the analysis scope specified by the user, the system automatically expands the analysis scope according to the regional boundaries and the water system characteristics of the regional rivers, clarifies the extended area that affects the model calculation and results, and further obtains basic data, historical data, real-time monitoring data and forecast data in real time through the built-in database, data resources provided by the operating environment, and data publicly available on the Internet according to the expanded scope; further verifies, cleans and organizes the data to form a usable model database; The system automatically selects the appropriate model, starts calculation and evaluation according to the preset parameters, further adjusts the parameters according to the comparison with the measured data of the site or historical floods, and generates multiple sets of parameters according to the adversarial algorithm for simultaneous calculation and simulation, continuously improving the fit of the scheme, and ultimately achieving high-precision model parameter groups and analysis results; in the subsequent operation process, it continuously superimposes measured data, dynamically adjusts relevant weights, and further improves model accuracy; and outputs analysis results with a higher degree of fit to users.

[0038] The invention also provides a flood risk assessment and early warning system.

[0039] See also Figure 4 , a flood risk assessment and early warning system, used to provide support for the operation of the flood risk assessment and early warning method based on big data analysis, the flood risk assessment and early warning system includes a big data acquisition module, an online monitoring module, a data processing module, a model building module, a virtual operation module and a self-learning module; The big data acquisition module and the online monitoring module are both connected to the data processing module by signal, the big data acquisition module is used to collect the basic information, and the online monitoring module is used to collect the real-time information; The data processing module is used to clean and classify the collected data, identify the data quality, and assign weights; The data construction module and the virtual operation module are both connected to the data processing module by signal. The model construction module is used to construct the virtual operation model. The virtual operation module is used to add real meteorological and regional river distribution conditions, and can adjust the water level depth of regional rivers according to actual rainfall. The self-learning module is signal-connected to the data processing module. The self-learning module automatically adjusts parameters based on a large number of model calculation results and corresponding disaster loss results, and continuously iterates and adjusts the corresponding results through machine learning generative adversarial networks and original data.

[0040] It is convenient to automatically collect information related to geography, water areas, and opening and closing equipment within the area; and generate virtual operation data in combination with real-time meteorological forecast information and water level change information to simulate and predict rainfall and water level changes within the forecast period in advance. When a flood disaster is predicted, rapid feedback is provided through system self-detection.

[0041] Please refer to Figure 4 and Figure 5Specifically, the data processing module includes a central processing unit, a disaster analysis unit and a data storage unit. The central processing unit is used for processing collected data, predicted data and feedback data; the disaster analysis unit is used for evaluating the predicted flood disaster and determining whether the water level has entered a flood disaster state; the data storage unit is used for storing comparison data of the flood disaster state, and is also used for storing system operation data and virtual operation model data.

[0042] Facilitates the processing, analysis and storage of system data.

[0043] Please refer to Figure 4 and Figure 6 Specifically, the big data collection module includes a regional river collection unit, a historical data collection unit, a weather forecast collection unit and a regional facility collection unit; the regional river collection unit is used to collect geographical information, regional river distribution information and regional river depth information within the regional range; the historical data collection unit is used to summarize the historical data of flood disasters that have occurred in the regional range in the past; the weather forecast collection unit is used to collect large-scale weather information and weather forecast information; the regional facility collection unit is used to summarize the number and distribution area of ​​opening and closing gates within the regional river range.

[0044] It is convenient to collect terrain data, historical flood disaster data, weather forecast data and gate opening and closing distribution data within the regional scope, and provide data support for the construction of the basic model.

[0045] Please refer to Figure 4 and Figure 7 Specifically, the online monitoring module includes a water level collection unit, a rainfall collection unit and an opening and closing gate collection unit; the water level collection unit is used to collect water level information within the regional river range in real time; the rainfall collection unit is used to collect rainfall information within the regional river range in real time; the opening and closing gate collection unit is used to collect status information corresponding to the opening and closing gates within the regional river range in real time.

[0046] It is convenient to conduct online monitoring of on-site water level, rainfall and gate opening and closing status within the regional river range, providing data support for simulation operation.

[0047] Please refer to Figure 4 and Figure 8Specifically, the model construction module includes a terrain generation unit, a water area generation unit and a rainfall generation unit. The terrain generation unit constructs in proportion according to the terrain information within the regional range to generate a virtual geographical environment; the water area generation unit is used to add a changeable water area environment to the constructed geographical model to generate a virtual water area environment; the rainfall generation unit is used to simulate the rainfall environment to generate a virtual rainfall environment; the virtual geographical environment, virtual water area environment and virtual rainfall environment are combined to generate a real virtual scene model.

[0048] It is convenient to obtain real virtual operation data based on the combination of terrain environment, regional river conditions and meteorological conditions, so as to simulate real rainfall and water level changes, so as to accurately obtain the water level changes in the river basin in different regions.

[0049] Please refer to Figure 4 and Fig. 9 Specifically, the virtual operation module includes a rainfall analysis unit, a water level analysis unit, a virtual conversion unit and a virtual synthesis unit. The rainfall analysis unit is used to generate predicted rainfall information according to the rainfall within the prediction period; the water level analysis unit is used to generate predicted water level information according to the current water level change; the virtual conversion unit is used to convert the predicted rainfall information and the predicted water level information into predicted environmental information, and the virtual synthesis unit is used to merge the predicted environmental information and the virtual scene model to obtain a predicted virtual scene model.

[0050] In this embodiment, the prediction virtual scene model is updated in real time following the meteorological changes and the water level changes collected in real time, so as to facilitate the real-time analysis of the flood disaster situation within the prediction period.

[0051] The predictive virtual scene model allows for easy and intuitive viewing of the scope and area of ​​flood disasters, allowing managers to review, analyze and record flood disasters within the forecast period.

[0052] In this embodiment, the water level analysis unit is also used to record the water level change per unit time. When there is a difference between the rainfall predicted by the meteorological forecast and the actual measured rainfall, the water level change within the period is predicted by the water level change.

[0053] Please refer to Figure 4 and Fig.10 Furthermore, the flood risk assessment and early warning system also includes a solution push module, which is used to push the flood warning information; the solution push module is also used to adaptively provide solution information during disaster warning and push the solution information.

[0054] Specifically, the solution push module includes a disaster assessment and early warning unit, a solution analysis unit and a solution push unit. The disaster assessment and early warning unit is used to analyze whether there is a flood disaster within the prediction period, and to promptly and synchronously warn management personnel and maintenance personnel in the area according to the flood disaster situation, so that the warning information is sent synchronously to improve the efficiency of information dissemination; the solution analysis unit is used to automatically simulate and analyze flood discharge plans to realize the analysis of flood discharge plans under the condition of safe diversion; the solution push unit is used to automatically push feasible drainage and control plans to the management user. After receiving the solution, the management user sends the opening and closing signal to the corresponding opening and closing site to realize flood discharge and diversion before the flood disaster occurs, so that excess water can be timely diverted to the safe river basin.

[0055] When a flood disaster occurs within the forecast period, the system automatically analyzes the flood distribution area, and simulates and analyzes the opening and closing plans of the gates at each point based on the situation in the distribution area to obtain a solution that can discharge the flood in a timely manner. This solution will not cause flooding in other river basins while meeting the flood discharge requirements. The system can automatically analyze and push the opening and closing plans of the gates, reduce manual intervention, and realize intelligent analysis and push of solutions.

[0056] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A flood risk assessment and early warning method based on big data analysis, characterized in that: The following steps are involved: Step S1, collecting basic information of regional rivers through big data; Step S2, collecting real-time information of regional rivers through the Internet of Things; Step S3, building a virtual operation model in combination with regional information and real-time information, for simulating a real regional river precipitation scenario according to actual meteorological information and rainfall information; Step S4, iteratively updating the virtual operation model according to the forecast data to generate flood conditions within the forecast period; Step S5: Feedback flood warning information based on the predicted flood situation within the period.

2. The flood risk assessment and early warning method based on big data analysis according to claim 1 is characterized in that: The basic information in step S1 is mainly the required terrain, hydrology and gate distribution data obtained from the Internet and intranet sources.

3. The flood risk assessment and early warning method based on big data analysis according to claim 2 is characterized in that: The real-time information in step S2 is mainly data on weather, precipitation and water level distribution monitored in real time by online equipment within the regional river range.

4. The flood risk assessment and early warning method based on big data analysis according to claim 3 is characterized in that: In step S3, the virtual operation model constructs a self-learning model scope, builds the model based on historical data, and continuously updates the model environment with real-time data for continuous model iteration; the virtual operation model fits real-time information and continuously improves the analysis accuracy of the virtual operation model during the system operation cycle.

5. The flood risk assessment and early warning method based on big data analysis according to claim 4 is characterized in that: The prediction data in step S4 includes meteorological prediction information, which is used to simulate precipitation operation within the prediction period according to the meteorological prediction information, generate precipitation and regional river water level changes within the prediction period, and is used for virtual prediction of flood disasters.

6. The flood risk assessment and early warning method based on big data analysis according to claim 5 is characterized in that: The prediction data also includes water level change data, which is the change of water level in regional rivers within unit time. When there is a large error between meteorological information and actual rainfall information, the water level change is simulated within the prediction period based on the water level change data to generate the regional river water level change within the prediction period for virtual prediction of flood disasters.

7. The flood risk assessment and early warning method based on big data analysis according to claim 6 is characterized in that: In step S5, the flood warning information is communicated to the site maintenance personnel via remote communication to provide early warning of flood disasters, so as to urge people in the area to evacuate in time.

8. The flood risk assessment and early warning method based on big data analysis according to claim 7 is characterized in that: Corresponding alarm equipment is equipped on the real-time monitoring equipment within the regional river range. When the alarm equipment receives flood disaster warning information, it automatically alarms within the range of the regional river, so that the alarm equipment can automatically alarm, facilitating simultaneous alarms at different monitoring points.

9. A flood risk assessment and early warning system, used to support the operation of the flood risk assessment and early warning method based on big data analysis as described in any one of claims 1 to 8, characterized in that: The flood risk assessment and early warning system includes a big data acquisition module, an online monitoring module, a data processing module, a model building module, a virtual operation module and a self-learning module; The big data acquisition module and the online monitoring module are both connected to the data processing module by signal, the big data acquisition module is used to collect the basic information, and the online monitoring module is used to collect the real-time information; The data processing module is used to clean and classify the collected data, identify the data quality, and assign weights; The data construction module and the virtual operation module are both connected to the data processing module by signal. The model construction module is used to construct the virtual operation model. The virtual operation module is used to add real weather and river distribution conditions and can adjust the depth of the river according to the actual rainfall. The self-learning module is signal-connected to the data processing module. The self-learning module automatically adjusts parameters based on a large number of model calculation results and corresponding disaster loss results, and continuously iterates and adjusts the corresponding results through machine learning generative adversarial networks and original data.

10. The flood risk assessment and early warning system according to claim 9, characterized in that: It also includes a solution push module, which is used to push the flood warning information; the solution push module is also used to adaptively provide solution information during disaster warning and push the solution information.

Images