A method and system for automatic monitoring of hydrological data

By conducting multi-level collection and catastrophic analysis of hydrological data, using geographical environment data and neural network algorithms to build a spatial and temporal continuous hydrological data map, the problem of insufficient monitoring points in traditional hydrological monitoring methods is solved, and accurate monitoring of hydrological environment and disasters are achieved. Early warning.

CN118708945BActive Publication Date: 2024-11-05HUBEI YIFANG TECH DEV +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411194481.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-05
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

Traditional hydrological monitoring methods rely on a single or small number of monitoring points, which is difficult to fully reflect the hydrological status of the entire region. There are shortcomings in hydrological data abnormal detection and disaster risk assessment, resulting in relatively accurate abnormal detection and disaster risk assessment capabilities for hydrological data. Low.

Method used

By acquiring geographical environment data for regional division, building a hydrological monitoring site network, performing multi-level data collection and spatiotemporal interpolation, generating a spatiotemporal continuous hydrological data map, using algorithms such as long and short-term memory neural networks to predict hydrological states and disaster risk assessment, and realizing intelligence Early warning.

Benefits of technology

It has improved the ability of accurate abnormal detection and disaster risk assessment of hydrological data, can timely identify and respond to abnormal situations in the hydrological environment, effectively warn of floods, river overflows and other disasters, and ensure the safety of water resources and the environment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118708945B_ABST
    Figure CN118708945B_ABST
Patent Text Reader

Abstract

The present application relates to an automatic monitoring method and system for hydrological data. The method comprises the following steps: obtaining geographic environment survey data; dividing the geographic environment survey data into regions to generate monitoring area division data, wherein the monitoring area division data includes key monitoring area data and conventional monitoring area data; constructing a monitoring site network through the monitoring area division data to generate a hydrological monitoring site network; performing multi-level monitoring data collection and adjustment on the hydrological monitoring site network using the key monitoring area data and the conventional monitoring area data, thereby generating monitoring site hydrological collection data; performing spatiotemporal interpolation on the monitoring site hydrological collection data and geographic environment survey data to generate site hydrological spatiotemporal interpolation data. The present invention improves the ability of accurate anomaly detection and disaster risk assessment of hydrological data by performing multi-level data collection and disaster analysis on hydrology.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of hydrological monitoring, and in particular to an automatic monitoring method and system for hydrological data. Background Art

[0002] The earliest hydrological monitoring methods mainly relied on manual measurements, which were not only time-consuming and labor-intensive, but also susceptible to human errors. With the advancement of science and technology, automated monitoring technologies have begun to be applied to hydrological data collection. The introduction of electronic sensors and data loggers has made it possible to automatically monitor basic hydrological parameters such as water level, flow rate and precipitation. These devices have greatly improved the accuracy and timeliness of data. Advances in communication technology have promoted the development of telemetry systems, which transmit real-time monitoring data remotely to central sites via radio, satellite or the Internet to facilitate real-time analysis and decision-making. During this period, global navigation satellite systems (GNSS) and radar technology were also introduced into hydrological monitoring, further improving the accuracy and coverage of monitoring. With the rise of the Internet of Things (IoT), big data and artificial intelligence (AI) technologies, the automatic monitoring system of hydrological data has been further improved. IoT devices can achieve a more dense and extensive monitoring network, while big data and AI technologies can process and analyze large amounts of hydrological data in real time, providing more accurate prediction and early warning services. However, the current traditional hydrological monitoring methods often rely on a single or a small number of monitoring points, which makes it difficult to fully reflect the hydrological conditions of the entire region. At the same time, there are deficiencies in hydrological data anomaly detection and comprehensive disaster risk assessment, resulting in low accurate anomaly detection and disaster risk assessment capabilities for hydrological data. Summary of the invention

[0003] Based on this, it is necessary to provide an automatic monitoring method and system for hydrological data to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for automatic monitoring of hydrological data is provided, the method comprising the following steps:

[0005] Step S1: Acquire geographic environment survey data; divide the geographic environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data;

[0006] Step S2: performing spatiotemporal interpolation on the hydrological collection data and geographic environment survey data of the monitoring site to generate spatiotemporal interpolation data of the site hydrological data; constructing a spatiotemporal map based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map;

[0007] Step S3: Perform hydrological time series change analysis on the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate site hydrological time series change data; perform model training on the site hydrological time series change data to generate a site hydrological monitoring state prediction model; import the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data;

[0008] Step S4: Based on the normal state prediction data of hydrological monitoring and the abnormal state prediction data of hydrological monitoring, the spatiotemporal continuous hydrological data map is monitored for hydrological data abnormality disasters, and normal hydrological abnormality monitoring data and abnormal hydrological comprehensive disaster risk index are generated; based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological abnormality monitoring data, the spatiotemporal continuous hydrological data map is subjected to hydrological intelligent early warning, and hydrological intelligent early warning data is obtained to execute the automatic monitoring operation of hydrological data.

[0009] The present invention obtains geographical environment data through modern technologies such as drones, ensuring high accuracy and wide coverage of the data, and providing reliable basic data for subsequent hydrological monitoring. The geographical environment data is divided into regions, key monitoring areas and conventional monitoring areas are clearly defined, and then a hydrological monitoring site network is constructed to ensure the reasonable distribution and coverage of the monitoring sites in space, effectively supporting the collection and analysis of hydrological data. Through spatiotemporal interpolation technology, the hydrological data of different monitoring sites are spatially and temporally continuous, and a spatiotemporal continuous hydrological data map is generated. This data processing method enables the dynamic changes of the hydrological environment to be accurately recorded and displayed, providing strong support for subsequent analysis and prediction. By analyzing the hydrological time series change data of the site and training algorithms such as long short-term memory neural networks (LSTM), the hydrological state changes of the monitoring site can be accurately predicted. This capability not only improves the real-time monitoring capability of hydrological environment changes, but also provides important data support for disaster warning. By using the model prediction data and abnormal monitoring data, the spatiotemporal continuous hydrological data map is comprehensively analyzed to generate normal hydrological anomaly monitoring data and abnormal hydrological comprehensive disaster risk index, which can timely identify and respond to abnormal conditions in the hydrological environment, effectively warn of disasters such as floods, river overflows, and soil landslides, thereby ensuring the safety of water resources and the sustainability of the environment. Therefore, the present invention improves the ability of accurate anomaly detection and disaster risk assessment of hydrological data by performing multi-level data collection and disaster analysis on hydrology.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: using a drone to obtain geographical environment survey data;

[0012] Step S12: Analyze the terrain and water system distribution of the geographical environment survey data, thereby generating terrain feature data and water system distribution feature data;

[0013] Step S13: dividing the geographical environment survey data into regions according to the terrain feature data and the water system distribution feature data to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data;

[0014] Step S14: confirming the deployment location of the hydrological monitoring station through the monitoring area division data to obtain the deployment location information data of the hydrological monitoring station; constructing the monitoring station network based on the deployment location information data of the hydrological monitoring station to generate the hydrological monitoring station network;

[0015] Step S15: Use the key monitoring area data and the conventional monitoring area data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data.

[0016] The present invention can achieve high-resolution geographic data acquisition by acquiring data through unmanned aerial vehicles, which is faster and more economical than traditional methods. These data can be used for accurate terrain modeling and environmental feature extraction, and provide detailed basic information for subsequent analysis and decision-making. The analysis of terrain characteristics and water system distribution data can reveal the geomorphic characteristics and water resource distribution of the region. These data provide a scientific basis for the subsequent regional division and deployment of hydrological monitoring stations, which is helpful for the rational planning of resource utilization and disaster prevention and mitigation. Regional division can focus monitoring work on key areas according to terrain and water system characteristics, optimize the configuration of monitoring resources, so as to ensure that important data is obtained in key areas in a timely manner, improve monitoring efficiency and data quality. After determining the deployment location of the monitoring station, a monitoring station network with wide and dense coverage area and such network structure can be established. Such network structure can effectively monitor hydrological changes under various geographical environmental conditions, and improve the response capability to water resource management and disaster warning. Multi-level monitoring data acquisition adjustment can ensure the quality and timeliness of data acquisition in different regions and different importance levels. These data are the basis for analyzing regional hydrological characteristics, predicting hydrological changes and formulating effective water resource management strategies.

[0017] Preferably, step S15 comprises the following steps:

[0018] Step S151: adjusting the site distribution of the hydrological monitoring site network in non-key areas according to the conventional monitoring area data, and generating conventional area site distribution adjustment data;

[0019] Step S152: Performing a hydrological monitoring demand assessment on the conventional area site distribution adjustment data and the key monitoring area data through the hydrological monitoring site network, thereby generating conventional area hydrological monitoring demand data and key area hydrological monitoring demand data;

[0020] Step S153: sorting the demand priorities of the conventional area hydrological monitoring demand data and the key area hydrological monitoring demand data to generate high-priority monitoring demand data and low-priority monitoring demand data;

[0021] Step S154: Based on the high-priority monitoring demand data and the hydrological monitoring demand data of the key areas, the data of the key monitoring areas are adjusted for high frequency of data collection, and the hydrological collection frequency adjustment data of the key area monitoring stations are generated; based on the low-priority monitoring demand data and the hydrological monitoring demand data of the conventional areas, the data of the conventional monitoring areas are adjusted for low frequency of data collection, and the hydrological collection frequency adjustment data of the conventional area monitoring stations are generated;

[0022] Step S155: collecting site hydrological data according to the hydrological collection frequency adjustment data of the key area monitoring site and the hydrological collection frequency adjustment data of the conventional area monitoring site to obtain the hydrological collection data of the monitoring site.

[0023] By adjusting the distribution of sites in non-key areas, the present invention can optimize the utilization efficiency of monitoring resources, ensure comprehensive monitoring coverage of conventional areas, and reduce resource waste. Assessing hydrological monitoring needs can ensure that sufficient monitoring coverage and data collection are provided in key areas and important time periods, which helps to accurately predict and respond to regional hydrological changes. Priority sorting can ensure that limited monitoring resources are used first for data collection in key areas and important time periods, improving the real-time and availability of data. High-frequency and low-frequency adjustments ensure the effective management of monitoring sites and the adjustment of data collection frequency under different demand priorities, so that monitoring data is more in line with actual needs. The final data collection ensures effective hydrological monitoring according to priorities and needs, provides real-time and comprehensive hydrological data, and supports water resources management, disaster warning, scientific research and other work.

[0024] Preferably, step S2 comprises the following steps:

[0025] Step S21: preprocessing the hydrological data collected at the monitoring site to generate standard hydrological data collected at the monitoring site, wherein the data preprocessing includes data cleaning, filling in missing values, processing of data outliers and data standardization;

[0026] Step S22: fusing the hydrological data collected at the standard monitoring site and the geographical environment survey data to generate site hydrological and geographical fusion data;

[0027] Step S23: performing spatiotemporal interpolation on the site hydrological and geographical fusion data to generate site hydrological spatiotemporal interpolation data;

[0028] Step S24: construct a spatiotemporal map based on the site hydrological spatiotemporal interpolation data, thereby generating a spatiotemporal continuous hydrological data map.

[0029] The present invention ensures the quality and consistency of monitoring data through data preprocessing. Cleaning data, filling missing values, processing outliers and standardizing data can improve the accuracy of subsequent data analysis and model building, and ensure the reliability of basic data. Data fusion combines hydrological collection data with geographic environment characteristic data to provide a more comprehensive data perspective. This fusion can reveal the correlation between hydrological data and the geographic environment, and help to more deeply understand the hydrological characteristics of the monitoring area. Spatiotemporal interpolation fills in data blank areas and generates continuous spatiotemporal hydrological data. This continuous data can be used for accurate hydrological model establishment and long-term trend analysis, and support disaster warning and resource management decisions. The spatiotemporal continuous hydrological data map shows the evolution and distribution of hydrological data in the spatiotemporal dimension. This map can help users intuitively understand the spatiotemporal variation laws of hydrological data and provide a visual data foundation for scientific research and decision support.

[0030] Preferably, step S24 includes the following steps:

[0031] Step S241: performing site geospatial analysis on the hydrological monitoring site network based on the site hydrological spatiotemporal interpolation data to generate site geospatial data;

[0032] Step S242: performing spatial matching on the site hydrological spatiotemporal interpolation data and the site geographic spatial data to generate site hydrological spatial matching data; performing temporal alignment on the site hydrological spatial matching data to generate site hydrological temporal alignment data;

[0033] Step S243: convert the site hydrological spatial matching data into a plane map through a heat map to generate a site hydrological spatial distribution map; use the site hydrological time alignment data to slide the site hydrological spatial distribution map on the time axis to generate a spatiotemporal continuous hydrological data map.

[0034] Through site geospatial analysis, the present invention can deeply understand the distribution characteristics and coverage of monitoring sites in geographic space, help optimize site layout, ensure that the monitoring network covers key areas, and improve the comprehensiveness and representativeness of data collection. Spatial matching and time alignment ensure the consistency and continuity of site hydrological data in space and time. This matching and alignment can reduce errors in data analysis and improve the credibility and utilization value of data. The site hydrological spatial distribution map shows the distribution density and trend of site hydrological data in geographic space through a heat map. This visualization method can intuitively convey the spatial distribution characteristics of the data and help users understand the hydrological conditions in the monitoring area. The spatiotemporal continuous hydrological data map combines the timeline sliding function to show the evolution and change trend of site hydrological data in the spatiotemporal dimension. This map helps to analyze the long-term trend and seasonal changes of hydrological changes, and provide a scientific basis for water resources management and environmental protection.

[0035] Preferably, step S3 comprises the following steps:

[0036] Step S31: performing hydrological time series change analysis on the hydrological data collected at the monitoring station according to the spatiotemporal continuous hydrological data map to generate station hydrological time series change data;

[0037] Step S32: dividing the data set of the site hydrological time series change data to generate a model training set and a model test set;

[0038] Step S33: using the long short-term memory neural network algorithm to perform model training on the model training set to generate a site hydrological monitoring state training model; performing model testing iterations on the site hydrological monitoring state training model through the model test set to generate a site hydrological monitoring state prediction model;

[0039] Step S34: importing the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data.

[0040] The present invention can identify and understand the periodic changes, trend changes and abnormal events of the site hydrological data through time series change analysis. These data provide a basis for subsequent model training and prediction, and help understand the dynamic changes of the hydrological system. Data set division ensures the independence and reliability of model training and testing. The training set is used to learn model parameters, and the test set is used to evaluate the generalization ability and prediction effect of the model, thereby improving the accuracy and reliability of the model. The long short-term memory neural network (LSTM) algorithm can capture the time series characteristics of hydrological data and is suitable for processing long-term dependencies and time series data. By training and testing the model, an accurate hydrological monitoring status prediction model can be generated for real-time prediction and monitoring of the hydrological state. The generated hydrological monitoring status prediction data can provide real-time hydrological monitoring status information, including predictions of normal and abnormal states. These prediction data can be used for timely warning and decision support to help managers effectively respond to changes and risks in the hydrological system.

[0041] Preferably, step S4 comprises the following steps:

[0042] Step S41: Based on the hydrological monitoring normal state prediction data, the spatiotemporal continuous hydrological data map is subjected to hydrological data abnormality monitoring to generate normal hydrological abnormality monitoring data;

[0043] Step S42: Performing hydrological disaster analysis on the spatiotemporal continuous hydrological data map according to the hydrological monitoring abnormal state prediction data, and generating an abnormal hydrological comprehensive disaster risk index;

[0044] Step S43: Perform hydrological intelligent early warning on the spatiotemporal continuous hydrological data map according to the hydrological abnormal disaster index and the normal hydrological anomaly monitoring data to obtain hydrological intelligent early warning data to perform automatic hydrological data monitoring operations.

[0045] The present invention can timely detect abnormal changes in hydrological data through abnormal monitoring data, including sudden events or changes in long-term trends. This monitoring helps to ensure the stable operation of the hydrological system and warn of potential problems in advance. Disaster analysis can comprehensively evaluate the risk level of abnormal hydrological data, including the type of hydrological disasters caused and the scope of impact. This analysis provides a scientific basis and early warning for responding to possible disasters. Intelligent early warning data can realize automatic monitoring and real-time response to the hydrological system. Through timely early warning and automated monitoring operations, the impact of hydrological disasters on the environment and society can be minimized, and water resource security and ecological environmental stability can be guaranteed.

[0046] Preferably, step S42 includes the following steps:

[0047] Step S421: extracting abnormal hydrological data from the spatiotemporal continuous hydrological data map according to the hydrological monitoring abnormal state prediction data to obtain abnormal hydrological monitoring data, wherein the abnormal hydrological data includes abnormal rainfall data, abnormal water level data, soil moisture data and abnormal area terrain data;

[0048] Step S422: performing runoff path and catchment area analysis on the abnormal rainfall data and the abnormal area terrain data to generate runoff path data and catchment area data; performing rainfall simulation on the runoff path data and catchment area data to generate flood disaster analysis data;

[0049] Step S423: analyzing the abnormal water level data to generate abnormal water level change data; predicting the abnormal water level change data to generate river overflow disaster analysis data;

[0050] Step S424: Calculate the slope of the abnormal area terrain data to obtain the abnormal area terrain slope data; perform slope risk assessment based on the soil moisture data and the abnormal area terrain slope data to generate landslide disaster analysis data; use the comprehensive disaster risk assessment calculation formula to perform multi-disaster comprehensive risk calculation on the flood disaster analysis data, river overflow disaster analysis data and landslide disaster analysis data to obtain the abnormal hydrological comprehensive disaster risk index.

[0051] The present invention can quickly identify and locate abnormal events and possible risk sources in the hydrological system by extracting abnormal hydrological data. Through these data, preventive measures and disaster management strategies can be taken to reduce potential disaster losses. Through runoff path and catchment area analysis and rainfall simulation, the impact of abnormal rainfall events on surface runoff and flood formation can be evaluated. This analysis helps to predict the probability of occurrence of flood disasters and the affected areas, and provides a scientific basis for emergency response and disaster warning. Water level change analysis and prediction can identify potential river overflow risks, especially when abnormal hydrological events occur. This analysis helps to formulate targeted flood control measures and river management strategies to protect coastal communities and infrastructure from water damage. Slope risk assessment and comprehensive disaster risk calculation can comprehensively consider the degree of potential impact of factors such as terrain, soil moisture and rainfall on landslide disasters. By calculating the comprehensive disaster risk index, the comprehensive risk level of different types of disasters can be quantified, providing a scientific basis for disaster prevention and emergency management.

[0052] Preferably, the calculation formula for comprehensive disaster risk assessment in step S424 is as follows:

[0053] ;

[0054] In the formula, Represented as in the area The results of the multi-hazard integrated risk assessment in Expressed as the weight coefficient of flood disaster, Expressed as the weight coefficient of river overflow disaster, Expressed as the weight coefficient of landslide hazard, Indicates flood hazard at location The level of disaster risk in the area Indicates the river overflow disaster at the location The level of disaster risk in the area Represents the landslide hazard at location The level of disaster risk in the area is expressed as the geographic spatial extent of the region being evaluated, Expressed as a position space coordinate vector product.

[0055] The present invention analyzes and integrates a comprehensive disaster risk assessment calculation formula. The basic principle of the formula is to comprehensively assess the risks of different disaster types within a spatial range and quantify them into a comprehensive risk index. The weight coefficient in the formula , , These coefficients determine the degree of influence of different disaster types on the comprehensive risk index. By adjusting these coefficients, the assessment results can be made more consistent with the relative impact of actual disasters. For example, in areas where floods are frequent, the weight of flood disasters may need to be increased to accurately reflect the overall risk of the area. Risk function , , , these functions describe the spatial position By properly defining and calculating these functions, the potential impact of disasters at each location can be accurately assessed based on factors such as topography, meteorology, and soil, thereby improving the accuracy and credibility of the assessment. The comprehensive disaster risk index obtained by summing up the risk scores of each disaster type reflects the risk of This comprehensive assessment helps to make scientific decisions in disaster management and risk planning. When using the conventional comprehensive disaster risk assessment calculation formula in this field, the following can be obtained: The comprehensive disaster risk assessment results in the region can be more accurately calculated by applying the comprehensive disaster risk assessment calculation formula provided by the present invention. The formula integrates the risks of multiple disaster types into one index, helping decision makers to fully understand the multi-disaster risks faced by the region, and helps to effectively allocate resources and formulate risk management strategies. By adjusting the weight coefficient, the importance of different disaster types can be more accurately reflected, thereby optimizing the priority of disaster response and prevention measures. Through the spatially distributed comprehensive disaster risk index, a disaster risk heat map or spatial distribution map can be produced to intuitively display the disaster risk situation in different locations in the region, which is conducive to the understanding and participation of the public and decision makers.

[0056] In this specification, an automatic monitoring system for hydrological data is provided, which is used to execute the above-mentioned automatic monitoring method for hydrological data. The automatic monitoring system for hydrological data includes:

[0057] The monitoring network deployment module is used to obtain geographical environment survey data; divide the geographical environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data;

[0058] The spatiotemporal map construction module is used to perform spatiotemporal interpolation on the hydrological collection data and geographical environment survey data of the monitoring site to generate the spatiotemporal interpolation data of the site hydrological data; the spatiotemporal map is constructed based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map;

[0059] The hydrological state prediction module is used to analyze the hydrological time series changes of the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate the site hydrological time series change data; to perform model training on the site hydrological time series change data to generate a site hydrological monitoring state prediction model; to import the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data;

[0060] The disaster anomaly analysis module is used to monitor the disasters caused by hydrological data anomalies on the spatiotemporal continuous hydrological data map based on the normal state prediction data of hydrological monitoring and the abnormal state prediction data of hydrological monitoring, and generate normal hydrological anomaly monitoring data and abnormal hydrological comprehensive disaster risk index; it performs hydrological intelligent early warning on the spatiotemporal continuous hydrological data map based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological anomaly monitoring data, and obtains hydrological intelligent early warning data to perform automatic monitoring of hydrological data.

[0061] The beneficial effect of the present invention is that by acquiring geographic environment survey data and performing regional division, the geographic characteristics and distribution of the monitoring area can be understood, providing basic data for the subsequent construction of a monitoring site network. According to the monitoring area division data, a hydrological monitoring site network is constructed, which helps to determine the location and layout of the monitoring site so as to effectively monitor the hydrological data. By collecting and adjusting the key monitoring area data and the conventional monitoring area data, the data collection of the hydrological monitoring site can be optimized, and the data quality and monitoring effect can be improved. By performing spatiotemporal interpolation on the hydrological collection data and the geographic environment survey data of the monitoring site, the site hydrological spatiotemporal interpolation data can be generated to provide more comprehensive and continuous hydrological information. Based on the site hydrological spatiotemporal interpolation data, a spatiotemporal continuous hydrological data map is constructed, and the hydrological data is displayed in the form of a map, which is convenient for comprehensive analysis and understanding of the hydrological data. By performing model training on the site hydrological time series change data, a site hydrological monitoring state prediction model can be generated to predict the site hydrological monitoring state. Based on the normal state prediction data of hydrological monitoring and the abnormal hydrological comprehensive disaster risk index, the spatiotemporal continuous hydrological data map is used to monitor the hydrological data abnormality disaster, and provide monitoring and evaluation of hydrological abnormalities. According to the abnormal hydrological comprehensive disaster risk index and the normal hydrological abnormality monitoring data, the spatiotemporal continuous hydrological data map is used for hydrological intelligent early warning, and hydrological intelligent early warning data is generated to provide early warning information for possible hydrological disasters. Therefore, the present invention improves the ability of accurate anomaly detection and disaster risk assessment of hydrological data by performing multi-level data collection and disaster analysis on hydrology. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of a process flow of an automatic monitoring method for hydrological data;

[0063] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0064] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0065] Figure 4 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0066] 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

[0067] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0068] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0069] It should be understood that although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, and the term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0070] To achieve this, please refer to Figures 1 to 4 , a method for automatic monitoring of hydrological data, the method comprising the following steps:

[0071] Step S1: Acquire geographic environment survey data; divide the geographic environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data;

[0072] Step S2: performing spatiotemporal interpolation on the hydrological collection data and geographic environment survey data of the monitoring site to generate spatiotemporal interpolation data of the site hydrological data; constructing a spatiotemporal map based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map;

[0073] Step S3: Perform hydrological time series change analysis on the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate site hydrological time series change data; perform model training on the site hydrological time series change data to generate a site hydrological monitoring state prediction model; import the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data;

[0074] Step S4: Based on the normal state prediction data of hydrological monitoring and the abnormal state prediction data of hydrological monitoring, the spatiotemporal continuous hydrological data map is monitored for hydrological data abnormality disasters, and normal hydrological abnormality monitoring data and abnormal hydrological comprehensive disaster risk index are generated; based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological abnormality monitoring data, the spatiotemporal continuous hydrological data map is subjected to hydrological intelligent early warning, and hydrological intelligent early warning data is obtained to execute the automatic monitoring operation of hydrological data.

[0075] The present invention obtains geographical environment data through modern technologies such as drones, ensuring high accuracy and wide coverage of the data, and providing reliable basic data for subsequent hydrological monitoring. The geographical environment data is divided into regions, key monitoring areas and conventional monitoring areas are clearly defined, and then a hydrological monitoring site network is constructed to ensure the reasonable distribution and coverage of the monitoring sites in space, effectively supporting the collection and analysis of hydrological data. Through spatiotemporal interpolation technology, the hydrological data of different monitoring sites are spatially and temporally continuous, and a spatiotemporal continuous hydrological data map is generated. This data processing method enables the dynamic changes of the hydrological environment to be accurately recorded and displayed, providing strong support for subsequent analysis and prediction. By analyzing the hydrological time series change data of the site and training algorithms such as long short-term memory neural networks (LSTM), the hydrological state changes of the monitoring site can be accurately predicted. This capability not only improves the real-time monitoring capability of hydrological environment changes, but also provides important data support for disaster warning. By using the model prediction data and abnormal monitoring data, the spatiotemporal continuous hydrological data map is comprehensively analyzed to generate normal hydrological anomaly monitoring data and abnormal hydrological comprehensive disaster risk index, which can timely identify and respond to abnormal conditions in the hydrological environment, effectively warn of disasters such as floods, river overflows, and soil landslides, thereby ensuring the safety of water resources and the sustainability of the environment. Therefore, the present invention improves the ability of accurate anomaly detection and disaster risk assessment of hydrological data by performing multi-level data collection and disaster analysis on hydrology.

[0076] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of an automatic monitoring method for hydrological data of the present invention. In this example, the automatic monitoring method for hydrological data includes the following steps:

[0077] Step S1: Acquire geographic environment survey data; divide the geographic environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data;

[0078] In an embodiment of the present invention, the required geographic environment data, such as terrain, water system, land use and other data, are obtained by using an unmanned aerial vehicle or other geographic information acquisition tools. Use a geographic information system (GIS) or similar tools to analyze and process the survey data to divide different monitoring areas, which can be divided according to characteristics such as terrain and water system distribution. Generate monitoring area division data based on the analysis results, including key monitoring areas and conventional monitoring areas. Key monitoring areas are usually areas with significant hydrological characteristics or greater impact on water resources management, while conventional monitoring areas cover other areas. Based on the monitoring area division data, determine the location and layout of monitoring sites, which can be hydrological monitoring sites for collecting hydrological data. According to the characteristics of the key monitoring areas and conventional monitoring areas, adjust and optimize the monitoring site network to ensure that multi-level hydrological data can be effectively collected and hydrological monitoring data that can fully reflect the characteristics of the geographic environment can be generated;

[0079] Step S2: performing spatiotemporal interpolation on the hydrological collection data and geographic environment survey data of the monitoring site to generate spatiotemporal interpolation data of the site hydrological data; constructing a spatiotemporal map based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map;

[0080] In an embodiment of the present invention, the collected hydrological data and geographic environment survey data of the monitoring site are preprocessed, including data cleaning (removing outliers and erroneous data), filling missing data values ​​(such as interpolation filling) and data format unification, so as to ensure data quality and consistency. Select a suitable spatiotemporal interpolation method, which usually includes an interpolation method based on spatial distance (such as Kriging interpolation), an interpolation method based on time series (such as linear interpolation or interpolation function), an interpolation method based on statistical distribution (such as Gaussian process interpolation), etc. The selection of the interpolation method depends on the characteristics of the data and the required accuracy. Interpolation calculation is performed on the data of each monitoring site to generate spatially continuous and temporally continuous hydrological data. The interpolated hydrological data is integrated into a unified data set and prepared for the construction of a spatiotemporal map. Select a suitable method to construct a spatiotemporal map, usually using a geographic information system (GIS) or a professional spatiotemporal data analysis tool. The spatiotemporal map can be two-dimensional or three-dimensional, reflecting the hydrological characteristics at different time points and spatial positions. Visualization tools such as map software or professional data visualization software can be used to analyze and visualize the spatiotemporal map for further data understanding and decision support. The ultimate goal is to generate a complete, spatiotemporal and continuous hydrological data map that can be used to monitor hydrological trends, analyze the spatial distribution and temporal evolution of hydrological events, and provide a basis for water resources management and prediction.

[0081] Step S3: Perform hydrological time series change analysis on the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate site hydrological time series change data; perform model training on the site hydrological time series change data to generate a site hydrological monitoring state prediction model; import the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data;

[0082] In an embodiment of the present invention, a spatiotemporal continuous hydrological data map that has been constructed and hydrological data collected from monitoring sites in real time are prepared. Ensure data quality and consistency, and perform necessary preprocessing such as data cleaning and removal of outliers. Select an analysis method to reveal the temporal variation characteristics of hydrological data at the monitoring site, such as commonly used methods including time series analysis, trend analysis, periodic analysis, etc., as well as the application of data mining and machine learning techniques. Perform temporal variation analysis on the hydrological data at the monitoring site to identify the temporal variation patterns and trends of hydrological data, including seasonal changes, long-term trends, periodic fluctuations, etc. According to the characteristics of the temporal variation data and the prediction requirements, select a suitable prediction model. Commonly used models include statistical models (such as ARIMA), machine learning models (such as regression models, support vector machines), and deep learning models (such as recurrent neural networks RNN). Use historical temporal variation data as a training set to train and optimize the parameters of the selected model to improve the accuracy and stability of the prediction model. Use the reserved test data set to verify and evaluate the trained model to evaluate its prediction accuracy and generalization ability. The trained model is applied to actual data to generate site hydrological monitoring status prediction data, which includes predicted normal and abnormal hydrological monitoring states, and is helpful for real-time monitoring and decision support.

[0083] Step S4: Based on the normal state prediction data of hydrological monitoring and the abnormal state prediction data of hydrological monitoring, the spatiotemporal continuous hydrological data map is monitored for hydrological data abnormality disasters, and normal hydrological abnormality monitoring data and abnormal hydrological comprehensive disaster risk index are generated; based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological abnormality monitoring data, the spatiotemporal continuous hydrological data map is subjected to hydrological intelligent early warning, and hydrological intelligent early warning data is obtained to execute the automatic monitoring operation of hydrological data.

[0084] In the embodiment of the present invention, the predicted normal and abnormal state data of hydrological monitoring and the spatiotemporal continuous hydrological data map are prepared. These data are integrated into one for subsequent analysis and processing. Suitable methods and algorithms are selected to identify and monitor hydrological data anomalies in the spatiotemporal continuous hydrological data map. Common methods include statistical analysis, machine learning models, time series analysis, etc. Hydrological data anomalies in the spatiotemporal continuous hydrological data map are analyzed, such as sudden water level changes, abnormal precipitation conditions, unexpected flow fluctuations, etc. These anomalies indicate potential hydrological disaster risks. The monitored normal hydrological data anomalies are sorted and recorded to generate normal hydrological anomaly monitoring data, which reflect the normal change pattern of the hydrological system. Combined with the abnormal hydrological comprehensive disaster risk assessment model, the abnormal hydrological comprehensive disaster risk index is calculated and generated. This index comprehensively considers the disaster risks caused by different types of hydrological anomalies, such as floods, droughts, soil erosion, etc. Specific methods and technologies are selected to realize hydrological intelligent early warning, and timely remind and take necessary countermeasures, including rule-based early warning systems, data-driven prediction models, etc. Based on the calculated abnormal hydrological comprehensive disaster risk index and normal hydrological abnormality monitoring data, hydrological intelligent early warning data is generated. The early warning data includes information such as disaster risk level, affected areas and recommended response measures.

[0085] Preferably, step S1 comprises the following steps:

[0086] Step S11: using a drone to obtain geographical environment survey data;

[0087] Step S12: Analyze the terrain and water system distribution of the geographical environment survey data, thereby generating terrain feature data and water system distribution feature data;

[0088] Step S13: dividing the geographical environment survey data into regions according to the terrain feature data and the water system distribution feature data to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data;

[0089] Step S14: confirming the deployment location of the hydrological monitoring station through the monitoring area division data to obtain the deployment location information data of the hydrological monitoring station; constructing the monitoring station network based on the deployment location information data of the hydrological monitoring station to generate the hydrological monitoring station network;

[0090] Step S15: Use the key monitoring area data and the conventional monitoring area data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data.

[0091] In the embodiment of the present invention, a drone equipped with a high-precision camera and LIDAR (light detection and ranging) equipment is selected to ensure the accuracy of the data. The flight path of the drone is planned using GIS (geographic information system) software to cover the entire target survey area. The drone is flown under a preset route to collect geographical environment data, including information such as terrain, vegetation, and water system. The data collected by the drone is imported into the GIS software for data cleaning and processing. A digital elevation model (DEM) is used to analyze terrain features and generate terrain feature data such as slope, slope aspect, and elevation. The water distribution characteristics are analyzed by a water system extraction algorithm to generate water system distribution feature data, including information such as rivers, lakes, and wetlands. According to the characteristics of the terrain and water system, the division standards for key monitoring and conventional monitoring areas are formulated. The survey data is divided into regions using GIS software to generate monitoring area division data. Key monitoring areas include areas with frequent geological activities, areas with abundant water resources, etc., and conventional monitoring areas include relatively stable areas. Select appropriate hydrological monitoring site locations in key monitoring areas and conventional monitoring areas. Based on the terrain characteristics and water system distribution data, the specific deployment location of the hydrological monitoring station is confirmed, and the deployment location information data of the hydrological monitoring station is generated. According to the deployment location information, the hydrological monitoring station is built, and the necessary monitoring equipment, such as water level meter, flow meter, rain gauge, etc., is installed. According to the deployment location information, the hydrological monitoring station is built, and the necessary monitoring equipment, such as water level meter, flow meter, rain gauge, etc., is installed. The data collection frequency of each station is adjusted according to the importance of the area. The data collection frequency in the key monitoring area is higher, and the data collection frequency in the conventional monitoring area is relatively low. Each monitoring station collects hydrological data at the set frequency and transmits it to the data center in real time through the wireless sensor network. The collected hydrological data is stored and preliminarily processed in the data center to generate the hydrological collection data of the monitoring station for subsequent analysis and decision-making.

[0092] Preferably, step S15 comprises the following steps:

[0093] Step S151: adjusting the site distribution of the hydrological monitoring site network in non-key areas according to the conventional monitoring area data, and generating conventional area site distribution adjustment data;

[0094] Step S152: Performing a hydrological monitoring demand assessment on the conventional area site distribution adjustment data and the key monitoring area data through the hydrological monitoring site network, thereby generating conventional area hydrological monitoring demand data and key area hydrological monitoring demand data;

[0095] Step S153: sorting the demand priorities of the conventional area hydrological monitoring demand data and the key area hydrological monitoring demand data to generate high-priority monitoring demand data and low-priority monitoring demand data;

[0096] Step S154: Based on the high-priority monitoring demand data and the hydrological monitoring demand data of the key areas, the data of the key monitoring areas are adjusted for high frequency of data collection, and the hydrological collection frequency adjustment data of the key area monitoring stations are generated; based on the low-priority monitoring demand data and the hydrological monitoring demand data of the conventional areas, the data of the conventional monitoring areas are adjusted for low frequency of data collection, and the hydrological collection frequency adjustment data of the conventional area monitoring stations are generated;

[0097] Step S155: collecting site hydrological data according to the hydrological collection frequency adjustment data of the key area monitoring site and the hydrological collection frequency adjustment data of the conventional area monitoring site to obtain the hydrological collection data of the monitoring site.

[0098] In an embodiment of the present invention, the monitoring requirements of non-key areas are determined by analyzing the geographical features and hydrological conditions of conventional monitoring areas. The distribution of hydrological monitoring sites is adjusted to ensure that the site layout in the conventional monitoring area is reasonable and can cover all areas that need to be monitored. The adjusted site distribution is recorded to generate conventional area site distribution adjustment data. The conventional area site distribution adjustment data and key monitoring area data are analyzed in detail using the hydrological monitoring site network to evaluate the hydrological monitoring requirements of each area. Based on the evaluation results, conventional area hydrological monitoring demand data and key area hydrological monitoring demand data are generated. According to the importance and urgency of the requirements, the hydrological monitoring demand data of conventional areas and key areas are sorted. After sorting, high-priority monitoring demand data and low-priority monitoring demand data are generated to guide the subsequent acquisition frequency adjustment. For high-priority monitoring demand data and key area hydrological monitoring demand data, the data acquisition frequency of the key monitoring area is adjusted to improve the frequency and accuracy of data acquisition. The adjusted acquisition frequency is recorded to generate the hydrological acquisition frequency adjustment data of the key area monitoring site. According to the low-priority monitoring demand data and the hydrological monitoring demand data of the conventional area, adjust the data collection frequency of the conventional monitoring area and reduce the collection frequency to save resources. Record the adjusted collection frequency and generate the hydrological collection frequency adjustment data of the conventional area monitoring site. According to the generated high-frequency and low-frequency adjustment data, perform the corresponding hydrological data collection work to ensure that the monitoring data of the key areas and conventional areas are accurate, record the collected hydrological data, and generate the hydrological collection data of the monitoring site.

[0099] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0100] Step S21: preprocessing the hydrological data collected at the monitoring site to generate standard hydrological data collected at the monitoring site, wherein the data preprocessing includes data cleaning, filling in missing values, processing of data outliers and data standardization;

[0101] Step S22: fusing the hydrological data collected at the standard monitoring site and the geographical environment survey data to generate site hydrological and geographical fusion data;

[0102] Step S23: performing spatiotemporal interpolation on the site hydrological and geographical fusion data to generate site hydrological spatiotemporal interpolation data;

[0103] Step S24: construct a spatiotemporal map based on the site hydrological spatiotemporal interpolation data, thereby generating a spatiotemporal continuous hydrological data map.

[0104] In an embodiment of the present invention, by deleting obvious errors or invalid data points in the collected data, such as outliers caused by sensor failures, the format of all data items is ensured to be consistent, and the units and naming specifications are unified. For time series data, a linear interpolation method is used to fill short-term missing values, and a moving average method is used to fill missing data in a larger range. Statistical methods or machine learning algorithms (such as IQR or Z-score) are used to detect outliers in the data. The detected outliers are replaced or corrected to ensure the rationality and continuity of the data. The data is scaled to the same range (such as 0-1) at a certain ratio for subsequent analysis and processing. The data is normalized by standard deviation so that the data has the same scale. Ensure that the hydrological collection data of the standard monitoring site is aligned with the geographical environment survey data in time and space, and extract relevant terrain features and water system features from the geographical environment survey data. The hydrological collection data of the standard monitoring site is integrated with the geographical environment survey data to generate site hydrological and geographical fusion data containing multi-dimensional information such as time, space, terrain, and hydrology. Select an appropriate spatiotemporal interpolation model, such as Kriging interpolation (Kriging), inverse distance weighted interpolation (IDW), etc. Use the selected model to interpolate the site hydrological and geographical fusion data to generate spatiotemporal interpolation data covering the entire monitoring area. Verify the interpolation results through cross-validation and other methods to ensure the accuracy of the interpolation data. Visualize the spatiotemporal interpolation data to generate hydrological data maps of different time and space dimensions. Continuously connect the data maps of different time steps to generate a spatiotemporal continuous hydrological data map. Analyze the spatiotemporal continuous hydrological data map to identify the changing trends and abnormalities of hydrological characteristics.

[0105] Preferably, step S24 includes the following steps:

[0106] Step S241: performing site geospatial analysis on the hydrological monitoring site network based on the site hydrological spatiotemporal interpolation data to generate site geospatial data;

[0107] Step S242: performing spatial matching on the site hydrological spatiotemporal interpolation data and the site geographic spatial data to generate site hydrological spatial matching data; performing temporal alignment on the site hydrological spatial matching data to generate site hydrological temporal alignment data;

[0108] Step S243: convert the site hydrological spatial matching data into a plane map through a heat map to generate a site hydrological spatial distribution map; use the site hydrological time alignment data to slide the site hydrological spatial distribution map on the time axis to generate a spatiotemporal continuous hydrological data map.

[0109] In an embodiment of the present invention, the specific geographical location of the hydrological monitoring site is determined by using GPS or other positioning technologies. The geographic spatial characteristics of each monitoring site are analyzed, including altitude, slope, vegetation coverage, etc. The geographical location and geographic spatial characteristics of each monitoring site are integrated to generate site geospatial data. The site hydrological spatiotemporal interpolation data is matched with the site geospatial data using a spatial matching algorithm, and the hydrological data of each site is associated with its geographical location. Site hydrological spatial matching data containing geographical location and corresponding hydrological data is generated. The site hydrological spatial matching data is time-aligned to ensure the consistency of the data in time, and the site hydrological time-aligned data is generated. A heat map is generated using the site hydrological spatial matching data to intuitively display the spatial distribution of the hydrological data. The heat map is converted into a plane map to display the spatial distribution of the hydrological data of each monitoring site, and a site hydrological spatial distribution map is generated. Using the site hydrological time-aligned data, the site hydrological spatial distribution map is slid on the time axis to dynamically display the changes in the hydrological data at different time points. Through the timeline sliding function, a spatiotemporal continuous hydrological data map covering the entire monitoring area is generated. Specifically, the geographical coordinates of each hydrological monitoring site are collected using a GPS device. For example, site A is located at 35.6895 degrees north latitude and 139.6917 degrees east longitude, with an altitude of 50 meters, a slope of 5 degrees, and a vegetation coverage of 30%. Match the hydrological data (such as water level and flow) of site A with its geographical location and characteristics. Ensure that all data are aligned at the same time point, for example, all site data at 10:00 on July 1, 2024 are aligned. Generate a heat map from site A to site Z to show the distribution of water levels. Convert the heat map into a plane map to show the water level of each site. By sliding the timeline, view the changes in hydrological data from July 1, 2024 to July 7, 2024, and generate a spatiotemporal continuous hydrological data map.

[0110] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0111] Step S31: performing hydrological time series change analysis on the hydrological data collected at the monitoring station according to the spatiotemporal continuous hydrological data map to generate station hydrological time series change data;

[0112] Step S32: dividing the data set of the site hydrological time series change data to generate a model training set and a model test set;

[0113] Step S33: using the long short-term memory neural network algorithm to perform model training on the model training set to generate a site hydrological monitoring state training model; performing model testing iterations on the site hydrological monitoring state training model through the model test set to generate a site hydrological monitoring state prediction model;

[0114] Step S34: importing the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data.

[0115] In an embodiment of the present invention, the hydrological data collected at each monitoring site, including water level, flow, rainfall, etc., are extracted from the spatiotemporal continuous hydrological data map. The extracted data is subjected to time series change analysis, the change trend of the hydrological data at each monitoring site in different time periods is observed, and the site hydrological time series change data is generated. The periodicity and trend changes of the hydrological data are further analyzed by statistical methods such as time series decomposition, moving average, differential analysis, etc. According to the time span and data integrity, the site hydrological time series change data is divided into a model training set and a model test set. The data of most time periods are selected for model training to ensure that the data set contains sufficient samples and change patterns. The data of the remaining time periods are selected for model testing to ensure that the time periods of the test data and the training data do not overlap, and a model test set is generated. A long short-term memory (LSTM) neural network model is designed and constructed, the input layer includes various indicators of hydrological data, the hidden layer is composed of several LSTM units, and the output layer generates prediction results. The model training set is imported into the LSTM model, the model is trained using the back propagation algorithm and the gradient descent method, the model parameters are adjusted, and a site hydrological monitoring status training model is generated. Import the model test set into the trained LSTM model for model testing. Evaluate the model performance through loss functions and evaluation indicators (such as RMSE and MAE), and iteratively optimize the model parameters according to the test results to finally generate the site hydrological monitoring status prediction model. Import the latest site hydrological time series change data into the site hydrological monitoring status prediction model. Use the LSTM model to predict the imported data to generate hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data. Analyze the prediction results, issue warnings for hydrological monitoring abnormal states, and provide corresponding response measures. For example, extract water level, flow, rainfall and other data from site A to site Z from January 2024 to June 2024. Through time series analysis, it is found that the water level of site A rises significantly from January to March each year, and gradually decreases from May to June. Select the data from January to April 2024 as the training set. Select the data from May to June 2024 as the test set. Design an LSTM neural network model with water level, flow, and rainfall as the input layer, two layers of LSTM units as the hidden layer, and hydrological status prediction as the output layer. Import the training set into the LSTM model and train the model parameters. Import the test set into the model for testing, evaluate the model performance through evaluation indicators such as RMSE, iteratively optimize the model parameters, and finally generate a prediction model. Import the latest hydrological data from July 2024 into the model. The model prediction results show that the water level at site A rises abnormally in the second week of July, and generate hydrological monitoring abnormal status prediction data. According to the prediction results, issue an early warning and formulate response measures, such as increasing the monitoring frequency of the site and making necessary flood prevention preparations.

[0116] As an example of the present invention, refer to Figure 4As shown, in this example, step S4 includes:

[0117] Step S41: Based on the hydrological monitoring normal state prediction data, the spatiotemporal continuous hydrological data map is subjected to hydrological data abnormality monitoring to generate normal hydrological abnormality monitoring data;

[0118] Step S42: Performing hydrological disaster analysis on the spatiotemporal continuous hydrological data map according to the hydrological monitoring abnormal state prediction data, and generating an abnormal hydrological comprehensive disaster risk index;

[0119] Step S43: Perform hydrological intelligent early warning on the spatiotemporal continuous hydrological data map according to the abnormal hydrological comprehensive disaster risk index and the normal hydrological anomaly monitoring data to obtain hydrological intelligent early warning data to perform automatic hydrological data monitoring operations.

[0120] In an embodiment of the present invention, the normal state prediction data of hydrological monitoring is extracted from the hydrological monitoring state prediction model. The extracted data is used to compare and analyze the spatiotemporal continuous hydrological data map, monitor the changes in hydrological data, and identify possible abnormalities. The identified abnormal change data is recorded and normal hydrological abnormal change monitoring data is generated. The abnormal state prediction data of hydrological monitoring is extracted from the hydrological monitoring state prediction model. A hydrological disaster analysis model is constructed, taking into account various disaster factors such as a sharp rise in water levels, floods, and droughts. The extracted abnormal state data is analyzed using the disaster analysis model, and the abnormal hydrological comprehensive disaster risk index is calculated and generated. Based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological abnormal change monitoring data, a hydrological intelligent early warning model is constructed. The spatiotemporal continuous hydrological data map is analyzed using the early warning model to generate hydrological intelligent early warning data. According to the generated hydrological intelligent early warning data, the corresponding hydrological data monitoring and response operations are automatically performed, and more specifically, the normal state prediction data of site A in July 2024 is extracted. By comparing the actual hydrological data in July, it was found that the water level was significantly different from the predicted value, and possible anomalies were identified, and the anomaly data was recorded as normal hydrological anomaly monitoring data. The abnormal state prediction data of site A in July 2024 was extracted, and a flood risk model considering the sharp rise in water level was constructed. The model was used to calculate the flood risk of site A in July, and an abnormal hydrological comprehensive disaster risk index was generated. An intelligent early warning model was constructed based on the abnormal hydrological comprehensive disaster risk index and normal hydrological anomaly monitoring data. The early warning model was used to analyze the hydrological data of site A, generate early warning data, and indicate possible flood risks. According to the early warning data, the monitoring frequency of site A was automatically increased, and emergency measures were deployed downstream.

[0121] Preferably, step S42 includes the following steps:

[0122] Step S421: extracting abnormal hydrological data from the spatiotemporal continuous hydrological data map according to the hydrological monitoring abnormal state prediction data to obtain abnormal hydrological monitoring data, wherein the abnormal hydrological data includes abnormal rainfall data, abnormal water level data, soil moisture data and abnormal area terrain data;

[0123] Step S422: performing runoff path and catchment area analysis on the abnormal rainfall data and the abnormal area terrain data to generate runoff path data and catchment area data; performing rainfall simulation on the runoff path data and catchment area data to generate flood disaster analysis data;

[0124] Step S423: analyzing the abnormal water level data to generate abnormal water level change data; predicting the abnormal water level change data to generate river overflow disaster analysis data;

[0125] Step S424: Calculate the slope of the abnormal area terrain data to obtain the abnormal area terrain slope data; perform slope risk assessment based on the soil moisture data and the abnormal area terrain slope data to generate landslide disaster analysis data; use the comprehensive disaster risk assessment calculation formula to perform multi-disaster comprehensive risk calculation on the flood disaster analysis data, river overflow disaster analysis data and landslide disaster analysis data to obtain the abnormal hydrological comprehensive disaster risk index.

[0126] In an embodiment of the present invention, abnormal state prediction data is extracted from a hydrological monitoring state prediction model. The extracted data is classified to obtain abnormal rainfall data, abnormal water level data, soil moisture data, and abnormal area terrain data. The classified data is stored to form abnormal hydrological monitoring data. Runoff path analysis is performed using abnormal rainfall data and abnormal area terrain data to generate runoff path data. Water catchment area analysis is performed based on runoff path data to generate water catchment area data. Rainfall simulation is performed using runoff path data and water catchment area data to generate flood disaster analysis data. River water level change analysis is performed on abnormal water level data to generate river abnormal water level change data. Water level change prediction is performed using river abnormal water level change data to generate river overflow disaster analysis data. Slope calculation is performed on abnormal area terrain data to obtain abnormal area terrain slope data. Slope risk assessment is performed based on soil moisture data and abnormal area terrain slope data to generate landslide disaster analysis data. The comprehensive disaster risk assessment calculation formula is used to calculate the multi-hazard comprehensive risk of flood disaster analysis data, river overflow disaster analysis data and landslide disaster analysis data to obtain the abnormal hydrological comprehensive disaster risk index. For example, the abnormal state prediction data of site A in July 2024 is extracted. The abnormal rainfall data (abnormal rainfall in July 2024), abnormal water level data (abnormal water level in July 2024), soil moisture data (soil moisture in July 2024), and abnormal regional terrain data (terrain data around site A) are classified and stored as abnormal hydrological monitoring data. The runoff path analysis is performed using the abnormal rainfall data and the terrain data of site A to generate runoff path data. The catchment area analysis is performed based on the runoff path data to generate catchment area data. The rainfall simulation is performed using the runoff path and catchment area data to generate flood disaster analysis data. The river water level change analysis is performed on the abnormal water level data of site A to generate river abnormal water level change data. The water level change prediction is performed using the river abnormal water level change data to generate river overflow disaster analysis data. The slope of the abnormal area terrain data around site A is calculated to obtain the terrain slope data. Risk assessment is performed based on soil moisture data and slope data to generate landslide disaster analysis data. The flood disaster analysis data, river overflow disaster analysis data and landslide disaster analysis data are comprehensively calculated using the comprehensive disaster risk assessment calculation formula to obtain the abnormal hydrological comprehensive disaster risk index.

[0127] Preferably, the calculation formula for comprehensive disaster risk assessment in step S424 is as follows:

[0128] ;

[0129] In the formula, Represented as in the area The results of the multi-hazard integrated risk assessment in Expressed as the weight coefficient of flood disaster, Expressed as the weight coefficient of river overflow disaster, Expressed as the weight coefficient of landslide hazard, Indicates flood hazard at location The level of disaster risk in the area Indicates the river overflow disaster at the location The level of disaster risk in the area Represents the landslide hazard at location The level of disaster risk in the area is expressed as the geographic spatial extent of the region being evaluated, Expressed as a position space coordinate vector product.

[0130] The present invention analyzes and integrates a comprehensive disaster risk assessment calculation formula. The basic principle of the formula is to comprehensively assess the risks of different disaster types within a spatial range and quantify them into a comprehensive risk index. The weight coefficient in the formula , , These coefficients determine the degree of influence of different disaster types on the comprehensive risk index. By adjusting these coefficients, the assessment results can be made more consistent with the relative impact of actual disasters. For example, in areas where floods are frequent, the weight of flood disasters may need to be increased to accurately reflect the overall risk of the area. Risk function , , , these functions describe the spatial position By properly defining and calculating these functions, the potential impact of disasters at each location can be accurately assessed based on factors such as topography, meteorology, and soil, thereby improving the accuracy and credibility of the assessment. The comprehensive disaster risk index obtained by summing up the risk scores of each disaster type reflects the risk of This comprehensive assessment helps to make scientific decisions in disaster management and risk planning. When using the conventional comprehensive disaster risk assessment calculation formula in this field, the following can be obtained: The comprehensive disaster risk assessment results in the region can be more accurately calculated by applying the comprehensive disaster risk assessment calculation formula provided by the present invention. The formula integrates the risks of multiple disaster types into one index, helping decision makers to fully understand the multi-disaster risks faced by the region, and helps to effectively allocate resources and formulate risk management strategies. By adjusting the weight coefficient, the importance of different disaster types can be more accurately reflected, thereby optimizing the priority of disaster response and prevention measures. Through the spatially distributed comprehensive disaster risk index, a disaster risk heat map or spatial distribution map can be produced to intuitively display the disaster risk situation in different locations in the region, which is conducive to the understanding and participation of the public and decision makers.

[0131] In this specification, an automatic monitoring system for hydrological data is provided, which is used to execute the above-mentioned automatic monitoring method for hydrological data. The automatic monitoring system for hydrological data includes:

[0132] The monitoring network deployment module is used to obtain geographical environment survey data; divide the geographical environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data;

[0133] The spatiotemporal map construction module is used to perform spatiotemporal interpolation on the hydrological collection data and geographical environment survey data of the monitoring site to generate the spatiotemporal interpolation data of the site hydrological data; the spatiotemporal map is constructed based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map;

[0134] The hydrological state prediction module is used to analyze the hydrological time series changes of the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate the site hydrological time series change data; to perform model training on the site hydrological time series change data to generate a site hydrological monitoring state prediction model; to import the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data;

[0135] The disaster anomaly analysis module is used to monitor the disasters caused by hydrological data anomalies on the spatiotemporal continuous hydrological data map based on the normal state prediction data of hydrological monitoring and the abnormal state prediction data of hydrological monitoring, and generate normal hydrological anomaly monitoring data and abnormal hydrological comprehensive disaster risk index; it performs hydrological intelligent early warning on the spatiotemporal continuous hydrological data map based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological anomaly monitoring data, and obtains hydrological intelligent early warning data to perform automatic monitoring of hydrological data.

[0136] The beneficial effect of the present invention is that by acquiring geographic environment survey data and performing regional division, the geographic characteristics and distribution of the monitoring area can be understood, providing basic data for the subsequent construction of a monitoring site network. According to the monitoring area division data, a hydrological monitoring site network is constructed, which helps to determine the location and layout of the monitoring site so as to effectively monitor the hydrological data. By collecting and adjusting the key monitoring area data and the conventional monitoring area data, the data collection of the hydrological monitoring site can be optimized, and the data quality and monitoring effect can be improved. By performing spatiotemporal interpolation on the hydrological collection data and the geographic environment survey data of the monitoring site, the site hydrological spatiotemporal interpolation data can be generated to provide more comprehensive and continuous hydrological information. Based on the site hydrological spatiotemporal interpolation data, a spatiotemporal continuous hydrological data map is constructed, and the hydrological data is displayed in the form of a map, which is convenient for comprehensive analysis and understanding of the hydrological data. By performing model training on the site hydrological time series change data, a site hydrological monitoring state prediction model can be generated to predict the site hydrological monitoring state. Based on the normal state prediction data of hydrological monitoring and the abnormal hydrological comprehensive disaster risk index, the spatiotemporal continuous hydrological data map is used to monitor the hydrological data abnormality disaster, and provide monitoring and evaluation of hydrological abnormalities. According to the abnormal hydrological comprehensive disaster risk index and the normal hydrological abnormality monitoring data, the spatiotemporal continuous hydrological data map is used for hydrological intelligent early warning, and hydrological intelligent early warning data is generated to provide early warning information for possible hydrological disasters. Therefore, the present invention improves the ability of accurate anomaly detection and disaster risk assessment of hydrological data by performing multi-level data collection and disaster analysis on hydrology.

[0137] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0138] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for automatic monitoring of hydrological data, characterized in that: The following steps are involved: Step S1: Obtaining geographical environment survey data; Divide the geographical environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data; Step S2: performing spatiotemporal interpolation on the hydrological collection data and geographic environment survey data of the monitoring site to generate spatiotemporal interpolation data of the site hydrological data; constructing a spatiotemporal map based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map; Step S3: analyzing the hydrological time series changes of the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate the hydrological time series change data of the site; Model training is performed on the site hydrological time series change data to generate a site hydrological monitoring status prediction model; the site hydrological time series change data is imported into the site hydrological monitoring status prediction model to obtain site hydrological monitoring status prediction data, wherein the site hydrological monitoring status prediction data includes hydrological monitoring normal status prediction data and hydrological monitoring abnormal status prediction data; Step S4: Based on the hydrological monitoring normal state prediction data and the hydrological monitoring abnormal state prediction data, the spatiotemporal continuous hydrological data map is subjected to hydrological data abnormal change disaster monitoring, and normal hydrological abnormal change monitoring data and abnormal hydrological comprehensive disaster risk index are generated; Based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological anomaly monitoring data, the spatiotemporal continuous hydrological data map is used for hydrological intelligent early warning, and the hydrological intelligent early warning data is obtained to perform automatic monitoring of hydrological data.

2. The automatic monitoring method for hydrological data according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: using a drone to obtain geographical environment survey data; Step S12: Analyze the terrain and water system distribution of the geographical environment survey data, thereby generating terrain feature data and water system distribution feature data; Step S13: dividing the geographical environment survey data into regions according to the terrain feature data and the water system distribution feature data to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; Step S14: confirming the deployment location of the hydrological monitoring station through the monitoring area division data to obtain the deployment location information data of the hydrological monitoring station; constructing the monitoring station network based on the deployment location information data of the hydrological monitoring station to generate the hydrological monitoring station network; Step S15: Use the key monitoring area data and the conventional monitoring area data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data.

3. The automatic monitoring method for hydrological data according to claim 2, characterized in that: Step S15 includes the following steps: Step S151: adjusting the site distribution of the hydrological monitoring site network in non-key areas according to the conventional monitoring area data, and generating conventional area site distribution adjustment data; Step S152: Performing a hydrological monitoring demand assessment on the conventional area site distribution adjustment data and the key monitoring area data through the hydrological monitoring site network, thereby generating conventional area hydrological monitoring demand data and key area hydrological monitoring demand data; Step S153: sorting the demand priorities of the conventional area hydrological monitoring demand data and the key area hydrological monitoring demand data to generate high-priority monitoring demand data and low-priority monitoring demand data; Step S154: Based on the high-priority monitoring demand data and the hydrological monitoring demand data of the key areas, the data of the key monitoring areas are adjusted for high frequency of data collection, and the hydrological collection frequency adjustment data of the key area monitoring stations are generated; based on the low-priority monitoring demand data and the hydrological monitoring demand data of the conventional areas, the data of the conventional monitoring areas are adjusted for low frequency of data collection, and the hydrological collection frequency adjustment data of the conventional area monitoring stations are generated; Step S155: collecting site hydrological data according to the hydrological collection frequency adjustment data of the key area monitoring site and the hydrological collection frequency adjustment data of the conventional area monitoring site to obtain the hydrological collection data of the monitoring site.

4. The automatic monitoring method for hydrological data according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: preprocessing the hydrological data collected at the monitoring site to generate standard hydrological data collected at the monitoring site, wherein the data preprocessing includes data cleaning, filling in missing values, processing of data outliers and data standardization; Step S22: fusing the hydrological data collected at the standard monitoring site and the geographical environment survey data to generate site hydrological and geographical fusion data; Step S23: performing spatiotemporal interpolation on the site hydrological and geographical fusion data to generate site hydrological spatiotemporal interpolation data; Step S24: construct a spatiotemporal map based on the site hydrological spatiotemporal interpolation data, thereby generating a spatiotemporal continuous hydrological data map.

5. The automatic monitoring method for hydrological data according to claim 4, characterized in that: Step S24 includes the following steps: Step S241: performing site geospatial analysis on the hydrological monitoring site network based on the site hydrological spatiotemporal interpolation data to generate site geospatial data; Step S242: performing spatial matching on the site hydrological spatiotemporal interpolation data and the site geographic spatial data to generate site hydrological spatial matching data; performing temporal alignment on the site hydrological spatial matching data to generate site hydrological temporal alignment data; Step S243: convert the site hydrological spatial matching data into a plane map through a heat map to generate a site hydrological spatial distribution map; use the site hydrological time alignment data to slide the site hydrological spatial distribution map on the time axis to generate a spatiotemporal continuous hydrological data map.

6. The automatic monitoring method for hydrological data according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing hydrological time series change analysis on the hydrological data collected at the monitoring station according to the spatiotemporal continuous hydrological data map to generate station hydrological time series change data; Step S32: dividing the data set of the site hydrological time series change data to generate a model training set and a model test set; Step S33: using the long short-term memory neural network algorithm to perform model training on the model training set to generate a site hydrological monitoring state training model; performing model testing iterations on the site hydrological monitoring state training model through the model test set to generate a site hydrological monitoring state prediction model; Step S34: importing the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data.

7. The automatic monitoring method for hydrological data according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: Based on the hydrological monitoring normal state prediction data, the spatiotemporal continuous hydrological data map is subjected to hydrological data abnormality monitoring to generate normal hydrological abnormality monitoring data; Step S42: Performing hydrological disaster analysis on the spatiotemporal continuous hydrological data map according to the hydrological monitoring abnormal state prediction data, and generating an abnormal hydrological comprehensive disaster risk index; Step S43: Perform hydrological intelligent early warning on the spatiotemporal continuous hydrological data map according to the abnormal hydrological comprehensive disaster risk index and the normal hydrological anomaly monitoring data to obtain hydrological intelligent early warning data to perform automatic hydrological data monitoring operations.

8. The automatic monitoring method for hydrological data according to claim 7, characterized in that: Step S42 includes the following steps: Step S421: extracting abnormal hydrological data from the spatiotemporal continuous hydrological data map according to the hydrological monitoring abnormal state prediction data to obtain abnormal hydrological monitoring data, wherein the abnormal hydrological data includes abnormal rainfall data, abnormal water level data, soil moisture data and abnormal area terrain data; Step S422: performing runoff path and catchment area analysis on the abnormal rainfall data and the abnormal area terrain data to generate runoff path data and catchment area data; performing rainfall simulation on the runoff path data and catchment area data to generate flood disaster analysis data; Step S423: analyzing the abnormal water level data to generate abnormal water level change data; predicting the abnormal water level change data to generate river overflow disaster analysis data; Step S424: Calculate the slope of the abnormal area terrain data to obtain the abnormal area terrain slope data; perform slope risk assessment based on the soil moisture data and the abnormal area terrain slope data to generate landslide disaster analysis data; use the comprehensive disaster risk assessment calculation formula to perform multi-disaster comprehensive risk calculation on the flood disaster analysis data, river overflow disaster analysis data and landslide disaster analysis data to obtain the abnormal hydrological comprehensive disaster risk index.

9. The automatic monitoring method for hydrological data according to claim 8, characterized in that: The calculation formula for comprehensive disaster risk assessment in step S424 is as follows: ; In the formula, Represented as in the area The results of the multi-hazard integrated risk assessment in Expressed as the weight coefficient of flood disaster, Expressed as the weight coefficient of river overflow disaster, Expressed as the weight coefficient of landslide hazard, Indicates flood hazard at location The level of disaster risk in the area Indicates the river overflow disaster at the location The level of disaster risk in the area Represents the landslide hazard at location The level of disaster risk in the area is expressed as the geographic spatial extent of the region being evaluated, Expressed as a position space coordinate vector product.

10. An automatic monitoring system for hydrological data, characterized in that: Used to execute the automatic monitoring method for hydrological data according to claim 1, the automatic monitoring system for hydrological data comprises: The monitoring network deployment module is used to obtain geographical environment survey data; divide the geographical environment survey data into regions to generate monitoring region division data, wherein the monitoring region division data includes key monitoring region data and conventional monitoring region data; construct a monitoring site network through the monitoring region division data to generate a hydrological monitoring site network; use the key monitoring region data and conventional monitoring region data to perform multi-level monitoring data collection and adjustment on the hydrological monitoring site network, thereby generating monitoring site hydrological collection data; The spatiotemporal map construction module is used to perform spatiotemporal interpolation on the hydrological collection data and geographical environment survey data of the monitoring site to generate the spatiotemporal interpolation data of the site hydrological data; the spatiotemporal map is constructed based on the spatiotemporal interpolation data of the site hydrological data to generate a spatiotemporal continuous hydrological data map; The hydrological state prediction module is used to analyze the hydrological time series changes of the hydrological data collected at the monitoring site according to the spatiotemporal continuous hydrological data map to generate the site hydrological time series change data; to perform model training on the site hydrological time series change data to generate a site hydrological monitoring state prediction model; to import the site hydrological time series change data into the site hydrological monitoring state prediction model to perform site hydrological monitoring state prediction data, wherein the site hydrological monitoring state prediction data includes hydrological monitoring normal state prediction data and hydrological monitoring abnormal state prediction data; The disaster anomaly analysis module is used to monitor the disasters caused by hydrological data anomalies on the spatiotemporal continuous hydrological data map based on the normal state prediction data of hydrological monitoring and the abnormal state prediction data of hydrological monitoring, and generate normal hydrological anomaly monitoring data and abnormal hydrological comprehensive disaster risk index; it performs hydrological intelligent early warning on the spatiotemporal continuous hydrological data map based on the abnormal hydrological comprehensive disaster risk index and the normal hydrological anomaly monitoring data, and obtains hydrological intelligent early warning data to perform automatic monitoring of hydrological data.

Citation Information

Cited By

  • Hydrological environment risk assessment method and system based on multi-source data fusion

    CN122198650A