阅读说明：本技术 一种区域大气水文耦合预警决策系统及方法 (Regional atmosphere hydrologic coupling early warning decision system and method ) 是由 何杰颖 张升伟 于 2019-11-06 设计创作，主要内容包括：本发明涉及气象水文预报和气象信息处理技术领域,具体涉及一种区域大气水文耦合预警决策系统,包括：区域资料获取模块,用于实时获取区域资料；人工智能模块,用于得到区域内的可降水分析模型；降水预报分析模块,用于获得区域内的降水量和降水路径预报；校正模块,用于获得修正后的降水量和降水路径预报,并将其存储在数据库中；水文资料获取模块,用于获取地面径流资料；水文预报模块,用于提供该区域内的水位预报、预警和预泄决策；响应演示模块,用于模拟未来数小时内区域内的降水径流趋势和降水在地面向低处汇流,实现径流预警和预测；预警决策模块,用于根据预先设置的径流监测点阈值,确定预警决策方案。(The invention relates to the technical field of meteorological hydrological prediction and meteorological information processing, in particular to a regional atmosphere hydrological coupling early warning decision-making system, which comprises: the regional data acquisition module is used for acquiring regional data in real time; the artificial intelligence module is used for obtaining a degradable water analysis model in the region; the rainfall forecast analysis module is used for obtaining the rainfall amount and the rainfall path forecast in the area; the correction module is used for obtaining the corrected precipitation amount and precipitation path forecast and storing the corrected precipitation amount and precipitation path forecast in a database; the hydrological data acquisition module is used for acquiring surface runoff data; the hydrologic forecasting module is used for providing water level forecasting, early warning and pre-leakage decision in the area; the response demonstration module is used for simulating rainfall runoff trend in an area within hours in the future and convergence of rainfall at a low position of the ground surface, so that runoff early warning and prediction are realized; and the early warning decision-making module is used for determining an early warning decision-making scheme according to a preset runoff monitoring point threshold value.)

1. A regional atmospheric hydrologic coupling early warning decision-making system, the system comprising: the system comprises a regional data acquisition module, an artificial intelligence module, a precipitation forecast analysis module, a correction module, a hydrologic data acquisition module, a hydrologic forecast module, a response demonstration module and an early warning decision module;

the area data acquisition module is used for acquiring area data of a certain specific area in real time;

the artificial intelligence module is used for establishing a database by utilizing years of observation data of a wind cloud meteorological satellite and regional degradable water historical record data, and obtaining a degradable water analysis model in the region by adopting a mode of combining statistical analysis and a forecasting mode;

the rainfall forecast analysis module is used for carrying out rainfall path and intensity forecast analysis in real time through the area data acquired in real time and the rainfall analysis model in the area so as to acquire the rainfall and rainfall path forecast in the area;

the correction module is used for correcting the results of the conventional rainfall and the rainfall path in the area which are pre-stored in the database by using the rainfall and the rainfall path forecast in the area, obtaining the corrected rainfall and the corrected rainfall path forecast, and storing the corrected rainfall and the corrected rainfall path forecast in the database;

the hydrological data acquisition module is used for acquiring surface runoff data;

the hydrological forecasting module is used for simulating a hydrological process under a complex terrain condition by utilizing satellite meteorological products and hydrological data information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-leakage decision in the area;

the response demonstration module is used for simulating the rainfall runoff trend in the area in the hours in the future and the convergence of rainfall at the low position of the ground in combination with the ground runoff data according to the acquired corrected rainfall and rainfall path forecast, and finally entering the river course to realize runoff early warning and prediction;

and the early warning decision module is used for forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining early warning response level through multiple iterations, and making a final decision.

2. The system of claim 1, wherein the region data comprises: regional observation data, grid point data, satellite infrared and microwave data.

3. The system of claim 1, wherein the meteorological product comprises: atmospheric temperature and humidity profile, atmospheric water vapor content, liquid water in the cloud, air pressure, rainfall rate, strong convection and typhoon path.

4. The system of claim 1, wherein the hydrological processes in complex terrain conditions are simulated land, soil physics processes, and surface, subsurface water flow and waterlogging/water reservoir processes.

5. The system of claim 1, wherein the surface runoff data comprises: ground GIS, hydrological information and road surface node water conservancy facility.

6. A regional atmospheric hydrologic coupling early warning decision method, which is implemented based on the decision system of any one of claims 1-5, and comprises the following steps:

acquiring regional data of a specific region in real time;

performing rainfall path and intensity forecast analysis in real time through the area data acquired in real time to obtain the rainfall amount and rainfall path forecast in the area;

correcting the conventional precipitation and precipitation path results stored in a database in advance by using the obtained precipitation and precipitation path forecast in the area to obtain the corrected precipitation and precipitation path forecast, and storing the corrected precipitation and precipitation path forecast in the database;

simulating a hydrological process under a complex terrain condition by using a satellite meteorological product and hydrological information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-leakage decision in the area;

simulating the rainfall runoff trend in the area and the convergence of rainfall at the low position of the ground in the coming hours according to the obtained corrected rainfall amount and rainfall path forecast and by combining with the ground runoff data, and finally entering the process of a river channel to realize the early warning and prediction of runoff;

and forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining an early warning response level through multiple iterations, and making a final decision.

7. The regional atmospheric hydrologic coupling early warning decision-making method of claim 6, wherein the regional data comprises: regional observation data, grid point data, satellite infrared and microwave data.

8. The regional atmospheric hydrologic coupling early warning decision-making method of claim 6, wherein the surface runoff data comprises: ground GIS, hydrological information and road surface node water conservancy facility.

Technical Field

The invention belongs to the technical field of meteorological hydrological forecasting and meteorological information processing, and particularly relates to a regional atmospheric hydrological coupling early warning decision-making system method and system.

Background

Analysis of scale characteristics of a weather system and summary analysis of weather and climate characteristics of heavy rain show that conventional detection data cannot meet application requirements at present with the aim of monitoring and early warning of a disastrous small and medium-scale weather system. The conventional sounding stations are also generally spaced at 100 kilometers, and only two conventional observations are made each day. In addition, vertical distribution information of atmospheric parameters is particularly important for monitoring and early warning of disastrous weather.

The flood disasters of small and medium watersheds and urban floods in mountainous areas caused by strong rainfall generated by short-time strong thunderstorms seriously threaten the life and property safety of people, and whether the large storm can cause regional flood depends on factors such as flood peak, flood volume and the like. Therefore, a timely hydrological simulation of a predicted storm or the like is required to determine the flood situation of the area.

The prior early warning decision method has the following technical problems:

(1) a highly accurate rainfall prediction for _ qpf cannot be provided;

(2) the perfect atmospheric strong precipitation forecasting model, hydrological model, rainfall runoff model or runoff calculation model can not be established to complete the land-air coupling, and the runoff prediction of enough forecast period can not be provided.

Disclosure of Invention

The invention aims to solve the technical defects of the existing early warning decision method, and provides a regional atmosphere hydrologic coupling early warning decision system which integrates the collected data of the meteorological and hydrologic histories, modularization, parameter calibration and verification, has the WRF-Hydro coupling waterlogging forecast and the water level forecast, early warning and pre-drainage decision support functions of a river, lake and pond in a specific region, and finally perfects the air-ground coupling flood disaster combined early warning system of a specific region, can effectively reduce various disasters such as farmlands, water conservancy and traffic, and can realize reliable scientific disaster prevention decision support to make timely and accurate meteorological hydrologic forecast.

In order to achieve the purpose, the invention provides a regional atmospheric hydrologic coupling early warning decision-making system, which utilizes space-space (satellite, conventional observation, ground GIS and hydrologic information) data, combines a weather forecast research mode and a hydrologic model to carry out a land-gas coupling model of bidirectional coupling, fuses multi-source data, and combines global and regional forecast fields and reanalysis field data to realize meteorological hydrologic forecast for a certain specific region.

The system comprises: the system comprises a regional data acquisition module, an artificial intelligence module, a precipitation forecast analysis module, a correction module, a hydrologic data acquisition module, a hydrologic forecast module, a response demonstration module and an early warning decision module;

the area data acquisition module is used for acquiring area data of a certain specific area in real time;

the artificial intelligence module is used for establishing a database by utilizing years of observation data of a wind cloud meteorological satellite and regional degradable water historical record data, and obtaining a degradable water analysis model in the region by adopting a mode of combining statistical analysis and a forecasting mode;

the rainfall forecast analysis module is used for carrying out rainfall path and intensity forecast analysis in real time through the area data acquired in real time and the rainfall analysis model in the area so as to acquire the rainfall and rainfall path forecast in the area;

the correction module is used for correcting the results of the conventional rainfall and the rainfall path in the area which are pre-stored in the database by using the rainfall and the rainfall path forecast in the area, obtaining the corrected rainfall and the corrected rainfall path forecast, and storing the corrected rainfall and the corrected rainfall path forecast in the database;

the hydrological data acquisition module is used for acquiring surface runoff data;

the hydrological forecasting module is used for simulating a hydrological process under a complex terrain condition by utilizing satellite meteorological products and hydrological data information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-leakage decision in the area;

the response demonstration module is used for simulating the rainfall runoff trend in the area in the hours in the future and the convergence of rainfall at the low position of the ground in combination with the ground runoff data according to the acquired corrected rainfall and rainfall path forecast, and finally entering the river course to realize runoff early warning and prediction;

and the early warning decision module is used for forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining early warning response level through multiple iterations, and making a final decision.

As an improvement of the above technical solution, the area data includes: regional observation data, grid point data, satellite infrared and microwave data.

As an improvement of the above technical solution, the weather product includes: atmospheric temperature and humidity profile, atmospheric water vapor content, liquid water in the cloud, air pressure, rainfall rate, strong convection and typhoon path.

As an improvement of the technical scheme, the hydrological process under the complex terrain condition is a simulated land surface and soil physical process, and a surface and underground water flow and waterlogging/reservoir ponding process.

As an improvement of the above technical solution, the surface runoff data includes: ground GIS, hydrological information and road surface node water conservancy facility.

The invention also provides a regional atmosphere hydrologic coupling early warning decision method, which is realized based on the decision system and comprises the following steps:

acquiring regional data of a specific region in real time;

forecasting and analyzing precipitation paths and intensity in real time through the area data in the area acquired in real time to obtain precipitation amount and precipitation path forecast in the area;

correcting the conventional precipitation and precipitation path results stored in a database in advance by using the obtained precipitation and precipitation path forecast in the area to obtain the corrected precipitation and precipitation path forecast, and storing the corrected precipitation and precipitation path forecast in the database;

simulating a hydrological process under a complex terrain condition by using a satellite meteorological product and hydrological information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-leakage decision in the area;

according to the obtained corrected precipitation amount and precipitation path forecast, combining with surface runoff data, simulating the precipitation runoff trend in the Jingjin wing area in the coming hours and the convergence of precipitation at the low position of the surface, and finally entering the river course to realize runoff early warning and prediction;

forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining an early warning response level through multiple iterations, and making a final decision; and further determining a set of complete schemes for preventing platform leakage, preventing flood and draining waterlogging and preventing flood disasters.

As an improvement of the above technical solution, the area data includes: regional observation data, grid point data, satellite infrared and microwave data.

As an improvement of the above technical solution, the surface runoff data includes: ground GIS, hydrological information and road surface node water conservancy facility.

Compared with the prior art, the invention has the beneficial effects that:

in the system, the atmospheric hydrological information is used for iterative coupling, so that the calibration of the model parameters is more accurate, and accurate model output parameters such as precipitation, precipitation range, runoff prediction and the like are obtained, and decision judgment is better performed. The system has effective simulation and prediction capabilities on rainstorm and waterlogging, and has the functions of water level prediction, early warning and prejudice decision guidance of all lakes and reservoirs in the coverage area; the rainfall forecast analysis module and the rainfall forecast analysis module are used for quantitative rainfall forecast, and the hydrologic forecast module is used for forecasting the runoff, so that the rainfall forecast system has high accuracy, longer forecast period and certain advancement and use effect.

Drawings

FIG. 1 is a schematic structural diagram of a regional atmospheric hydrologic coupling early warning decision system according to the present invention;

fig. 2 is a flow chart of iterative computation for determining an early warning response level and making a final decision of the regional atmospheric hydrologic coupling early warning decision system of the present invention.

Detailed Description

The invention will now be further described with reference to the accompanying drawings.

As shown in fig. 1, the present invention provides a regional atmospheric hydrologic coupling early warning decision system, which utilizes a weather forecast research model and a hydrologic model, namely a WRF model and a WRF-Hydro model, based on a global forecast field, reanalysis lattice data and regional high-resolution observation data, combining satellite infrared and microwave data, and basic GIS information of the road surface, road surface node water conservancy facilities, such as fish ponds, reservoirs, lakes, river networks, sluice gates and pump stations, develop an extreme weather flood disaster early warning decision support system in a specific area, realize typhoon path and intensity forecast, rainfall process and rainfall forecast, and then typhoon early warning, rainfall process and rainfall amount early warning, river, lake, pond and hydrological forecast, regional production convergence process forecast, urban river water level, surface water forecast, regional flood forecast, regional platform prevention and drainage decision, regional flood prevention and drainage decision and flood disaster forecast are realized.

The system comprises: the system comprises a regional data acquisition module, an artificial intelligence module, a precipitation forecast analysis module, a correction module, a hydrologic data acquisition module, a hydrologic forecast module, a response demonstration module and an early warning decision module;

the regional data acquisition module is used for acquiring regional data in real time;

wherein the area data comprises: regional observation data, lattice point data, satellite infrared and microwave data;

the artificial intelligence module is used for establishing a database by utilizing years of observation data of a wind cloud meteorological satellite and regional degradable water historical record data, and obtaining a degradable water analysis model with high adaptability and accuracy in a region by adopting a mode of combining statistical analysis and a forecasting mode;

the rainfall forecast analysis module is used for carrying out rainfall path and intensity forecast analysis in real time through the area data acquired in real time and the rainfall analysis model in the area so as to acquire the rainfall and rainfall path forecast in the area; specifically, by acquiring regional data in the region in real time and a degradable water analysis model in the region, combining global reanalysis data, setting the region by using an atmospheric assimilation mode WRFDA (WRFDA), parameterizing the atmospheric assimilation mode, setting forecast duration and resolution, and carrying out rainfall path and intensity forecast analysis to obtain rainfall and rainfall path forecast in the region;

the method comprises the steps of utilizing a satellite to read regional data on line, obtaining longitude and latitude and time, carrying out format conversion on the regional data after quality control and space-time matching, namely converting an HDF format into a binary format, carrying out regional data assimilation based on an improved background error covariance matrix atmosphere assimilation mode, forecasting an intensive rainfall path and intensity, and providing optimized intensive rainfall path and intensity forecast by combining conventional rainfall forecast.

Specifically, by utilizing an improved atmospheric mode parameter forecasting and assimilating mode, a high-resolution time-varying background error covariance matrix is adopted, the time-varying interval is 15 minutes, and then a satellite microwave data assimilating mode is adopted to obtain 24-48 hours of degradable water data and precipitation data according to an artificial intelligence method and a big data analysis method; the improved atmospheric mode parameter prediction and assimilation mode in the implementation is not a static background error covariance matrix in a conventional assimilation mode, but a high-resolution time-varying background error covariance matrix is adopted, and the prediction accuracy is effectively improved.

The correction module is used for correcting the results of the conventional rainfall and the rainfall path in the area which are stored in the database in advance by using the rainfall and the rainfall path forecast in the area, obtaining the corrected rainfall and the corrected rainfall path forecast in the area and storing the corrected rainfall and the corrected rainfall path forecast in the database;

the hydrological data acquisition module is used for acquiring surface runoff data;

wherein the surface runoff data comprises: ground GIS, hydrological information and surface node water infrastructure, e.g. ponds, reservoirs, lakes, river networks, sluices and pumping stations.

The hydrological forecasting module is used for simulating a hydrological process under a complex terrain condition by utilizing satellite meteorological products and hydrological data information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-leakage decision in the area;

wherein the weather product comprises: atmospheric temperature and humidity profile, atmospheric water vapor content, liquid water in the cloud, air pressure, rainfall rate, strong convection and typhoon path.

The hydrological process under the complex terrain condition is a simulated land surface and soil physical process, and a surface and underground water flow and waterlogging/reservoir ponding process.

Specifically, the atmospheric hydrologic coupling mode refers to the coupling of an atmospheric assimilation model and a hydrologic model to study water circulation, so as to achieve the aim of accurate runoff prediction. The high-resolution output results of the atmospheric assimilation model, such as atmospheric temperature and humidity profiles, atmospheric water vapor content, liquid water in cloud, air pressure, rainfall rate, strong convection, typhoon path and other conventional meteorological observation data and satellite data, are used as meteorological forcing factors to drive hydrological model input of regional scales and drainage basin scales, and a parameter calibration scheme is combined with land hydrological information, atmospheric information and sea, road and land boundary information. According to regional landform and hydrological information, system setting is needed, and static parameter setting and preprocessing of a forecasting system are needed; the coupling mode of the atmospheric assimilation model and the hydrological model is bidirectional coupling, the result of the hydrological model is input into the atmospheric model in a rasterization mode, rainfall, historical runoff and historical evapotranspiration are used as links of the interaction relation of the two models, and the interaction mechanism of the hydrological model and the atmospheric model is considered.

The response demonstration module is used for simulating the rainfall runoff trend in the area in the hours in the future and the convergence of the rainfall at the low part of the ground surface according to the obtained corrected rainfall and rainfall path forecast and by combining with the ground runoff data, and finally entering the river course to realize runoff early warning and prediction.

And the early warning decision module is used for forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining early warning response level through multiple iterations, and making a final decision. Specifically, a runoff monitoring point threshold value, such as a water level, a flow rate and the like, is calculated according to a forecast result output by the atmospheric mode. The calculation is actually a black box empirical algorithm, and is obtained by empirical statistics of historical data.

According to the strong rainfall path and the strength forecasting result, by combining basic GIS information of the road surface and road surface nodes such as fish ponds, reservoirs, lakes, river networks, water gates, pump stations and other water conservancy facilities, searching for an optimized hydrological network parameter threshold value by utilizing a Monte Carlo algorithm, a neural network algorithm and an artificial intelligent optimization algorithm, establishing a set of complete water level forecasting, early warning and pre-leakage decision guidance interfaces based on the water level measured values of the ponds of the rivers and the lakes, the characteristics of respective geographic positions and hydrological states, and intelligently displaying real-time changes and relevant response decisions of the ponds of the rivers and the lakes in the future extreme disaster weather process.

Based on the actual water level measurement values of the rivers, lakes, ponds and the characteristics of respective geographic positions and hydrological states, a set of complete water level forecasting, early warning and pre-drainage decision guidance interfaces is established, and real-time changes and relevant response decisions of the rivers, lakes, ponds and the ponds in the area in the future in extreme disaster weather process are intelligently displayed. And verifying the flood disaster forecasting and early warning decision system in the area in practical application, and comparing forecasting and early warning results used by services to perfect the forecasting, early warning and pre-discharge decision of the water level of the related river and lake reservoir.

The system utilizes an atmosphere assimilation model and a hydrological model, based on a global forecast field, reanalysis grid data and regional high-resolution observation data, combines satellite infrared and microwave data and basic GIS information of a road surface, develops a regional extreme weather flood disaster early warning decision system by using road surface nodes such as fishponds, reservoirs, lakes, river networks, sluice gates, pump stations and other water conservancy facilities, realizes typhoon path and intensity forecast, precipitation process and precipitation quantity forecast, and further realizes typhoon early warning, precipitation process and rainfall quantity early warning, river, lake and pond hydrological forecast, regional production convergence process forecast, urban river water level, road surface ponding forecast, regional flood forecast, regional platform prevention pre-discharge decision, regional flood prevention and drainage decision and flood disaster forecast.

The system has effective simulating and forecasting capabilities on rainstorm and waterlogging, covers all lake and reservoir pond water level forecasting, early warning and pre-drainage decision guidance in the specific area range, and can provide visual, accurate and quantitative rainstorm and waterlogging forecasting for forecasters.

Based on the prediction system, the invention also provides a regional atmosphere hydrologic coupling early warning decision method, which comprises the following steps:

acquiring regional data in a specific region in real time;

wherein the area data comprises: regional observation data, lattice point data, satellite infrared and microwave data;

performing rainfall path and intensity forecast analysis in real time through the area data acquired in real time to obtain the rainfall amount and rainfall path forecast in the area;

correcting the conventional precipitation and precipitation path results stored in a database in advance by using the obtained precipitation and precipitation path forecast in the area to obtain the corrected precipitation and precipitation path forecast, and storing the corrected precipitation and precipitation path forecast in the database;

simulating a hydrological process under a complex terrain condition by using a satellite meteorological product and hydrological information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-leakage decision in the area;

specifically, the atmospheric hydrologic coupling mode refers to the coupling of an atmospheric assimilation model and a hydrologic model to study water circulation, so as to achieve the aim of accurate runoff prediction. The high-resolution output results of the atmospheric assimilation model, such as atmospheric temperature and humidity profiles, atmospheric water vapor content, liquid water in cloud, air pressure, rainfall rate, strong convection, typhoon path and other conventional meteorological observation data and satellite data, are used as meteorological forcing factors to drive hydrological model input of regional scales and drainage basin scales, and a parameter calibration scheme is combined with land hydrological information, atmospheric information and sea, road and land boundary information. According to regional landform and hydrological information, system setting is needed, and static parameter setting and preprocessing of a forecasting system are needed; the coupling mode is bidirectional coupling, the result of the hydrological model is input into the atmospheric model in a rasterization mode, rainfall, historical runoff and historical evapotranspiration are used as links of the interaction relationship of the two models, and the interaction mechanism of the hydrological model and the atmospheric model is considered;

and simulating the rainfall runoff trend in the area and the convergence of rainfall at the low position of the ground in the coming hours according to the obtained corrected rainfall amount and rainfall path forecast and by combining with the ground runoff data, and finally entering the river course to realize runoff early warning and prediction.

Forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining an early warning response level through multiple iterations, and making a final decision; and further determining a set of complete schemes for preventing platform leakage, preventing flood and draining waterlogging and preventing flood disasters.

Specifically, principal component analysis is carried out on ground information (such as a fishpond, a reservoir, a lake, a river network, a sluice, a pump station and other water conservancy facilities), main parameters input by a neural network model are determined according to a specified region, weight function training is carried out on historical monitoring data of runoff by utilizing a backward transmission algorithm to obtain a new runoff forecasting result, the new runoff forecasting result is recorded and fed back to the neural network model for input, the forecast rainfall and the runoff forecasting are intelligently judged through repeated iteration, and a current early warning grade is determined by utilizing a lookup table method according to the match of a historical rainfall forecasting value, a historical runoff recording value and a historical response grade to obtain accurate decision judgment; see in particular fig. 2.

Firstly, extracting useful information aiming at high correlation of ground data, performing data dimensionality reduction by using knowledge of linear algebra, converting a plurality of variables into a few irrelevant comprehensive variables to represent GIS information in a region, and preparing for shortening the training time of a neural network and improving the training precision. The neural network method adopts a three-layer backward transmission neural network structure, AN input layer is AN element after principal component analysis, hidden layers A1, A2 and … AN represent N hidden layer neurons, the number of the hidden layer neurons is determined by the number M of the input neurons, AN output layer is runoff prediction, N is equal to the sum of the number M of the input layers plus the number 1 of the output layers plus the planned iteration times X to obtain the square, matched historical ground information and runoff information are trained to obtain a neural network output value, namely runoff prediction, the predicted value is fed back to the predicted rainfall amount, the neural network training process is continuously carried out according to the variable determined by principal component analysis, the iterative runoff predicted value and the rainfall amount are matched by utilizing a lookup table method according to the historical rainfall amount predicted value, the historical runoff recording value and the historical response level, the current early warning level is intelligently determined, the rainfall amount and the runoff predicted value recorded by each iteration are obtained, and obtaining accurate decision judgment.

Example 1.

In the present embodiment, a specific area is set as the kyford zone.

The embodiment provides a regional atmosphere hydrologic coupling early warning decision-making system, and the system includes: the system comprises an area data acquisition module, an artificial intelligence module, a precipitation forecast analysis module, a correction module, a hydrological data acquisition module, a hydrological forecast module and a response demonstration module;

the area data acquisition module is used for acquiring the area data of the Jingjin Ji area in real time; the Beijing jin Ji regional data comprises: regional observation data, lattice point data, satellite infrared and microwave data;

the artificial intelligence module is used for establishing a database by utilizing years of observation data of a wind cloud meteorological satellite and degradable water history record data of the Jingjin Ji area, and obtaining a degradable water analysis model in the Jingjin Ji area by adopting a mode of combining statistical analysis and a forecasting mode;

the rainfall forecast analysis module is used for carrying out rainfall path and intensity forecast analysis in real time through the area data acquired in real time and the rainfall analysis model in the Jingjin Ji area to obtain the rainfall and rainfall path forecast in the Jingjin Ji area;

the correcting module is used for correcting the conventional precipitation amount and precipitation path result in the Jingjin Ji area which is pre-stored in the database by using the precipitation amount and precipitation path forecast in the Jingjin Ji area to obtain the corrected precipitation amount and precipitation path forecast, and storing the corrected precipitation amount and precipitation path forecast in the database;

the hydrological data acquisition module is used for acquiring surface runoff data; the surface runoff data comprises: ground GIS, hydrological information and surface node water infrastructure, e.g. ponds, reservoirs, lakes, river networks, sluices and pumping stations.

The hydrological forecasting module is used for simulating a hydrological process under a complex terrain condition by utilizing satellite meteorological products and hydrological data information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-discharge decision in the Jingjin Ji area;

the response demonstration module is used for simulating the rainfall runoff trend in the Jingjin Ji area and the convergence of rainfall at the lower part of the ground in the coming hours according to the acquired corrected rainfall amount and rainfall path forecast and combining with the ground runoff data, and finally entering the river course to realize runoff early warning and prediction.

The early warning decision module is used for carrying out weight analysis and decision judgment on early warning response levels according to preset runoff monitoring point thresholds, providing fault tolerance analysis reports, determining decision schemes and further determining a set of complete prevention schemes for preventing platform leakage, flood prevention and drainage and flood disasters in Jingjin Ji areas.

Based on the prediction system, the embodiment also provides a regional atmospheric hydrologic coupling early warning decision method, which comprises the following steps:

acquiring regional data of the Jingjin Ji region in real time;

wherein the area data comprises: regional observation data, lattice point data, satellite infrared and microwave data;

carrying out rainfall path and intensity forecast analysis in real time through the area data acquired in real time to obtain the rainfall amount and rainfall path forecast in the Jingjin Ji area;

correcting the conventional precipitation amount and precipitation path results stored in a database in advance by using the obtained precipitation amount and precipitation path forecast in the Jingjin Ji area to obtain the corrected precipitation amount and precipitation path forecast, and storing the corrected precipitation amount and precipitation path forecast in the database;

simulating a hydrological process under a complex terrain condition by using a satellite meteorological product and hydrological information based on an atmospheric hydrological coupling mode, and providing water level forecasting, early warning and pre-discharge decision in the Jingjin Ji area;

specifically, the atmospheric hydrologic coupling mode refers to the coupling of an atmospheric assimilation model and a hydrologic model to study water circulation, so as to achieve the aim of accurate runoff prediction. The high-resolution output results of the atmospheric assimilation model, such as atmospheric temperature and humidity profiles, atmospheric water vapor content, liquid water in cloud, air pressure, rainfall rate, strong convection, typhoon path and other conventional meteorological observation data and satellite data, are used as meteorological forcing factors to drive hydrological model input of regional scales and drainage basin scales, and a parameter calibration scheme is combined with land hydrological information, atmospheric information and sea, road and land boundary information. According to regional landform and hydrological information, system setting is needed, and static parameter setting and preprocessing of a forecasting system are needed; the coupling mode is bidirectional coupling, the result of the hydrological model is input into the atmospheric model in a rasterization mode, rainfall, historical runoff and historical evapotranspiration are used as links of the interaction relationship of the two models, and the interaction mechanism of the hydrological model and the atmospheric model is considered;

and simulating the rainfall runoff trend in the Jingjin wing area and the convergence of rainfall at the lower part of the ground in the coming hours according to the obtained corrected rainfall amount and rainfall path forecast and by combining with the ground runoff data, and finally entering a river course to realize runoff early warning and prediction.

Forecasting and updating the runoff monitoring points according to a preset runoff monitoring point threshold value and in combination with the forecasted rainfall, feeding back the forecasted rainfall, determining an early warning response level through multiple iterations, and making a final decision; and further determining a set of complete schemes for preventing platform leakage, preventing flood and draining waterlogging and preventing flood disasters.

Specifically, principal component analysis is carried out on ground information (such as a fishpond, a reservoir, a lake, a river network, a sluice, a pump station and other water conservancy facilities), main parameters input by a neural network model are determined according to a specified Jingjin wing area, weight function training is carried out on historical monitoring data of runoff by using a backward transmission algorithm to obtain a new runoff forecasting result, the new runoff forecasting result is recorded and fed back to the neural network model for input, the rainfall forecast and the runoff forecast are intelligently judged through repeated iteration, and a current early warning grade is determined by using a lookup table method according to the matching of a historical rainfall forecasting value, a historical runoff recording value and a historical response grade, so that accurate decision judgment is obtained.

Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and are not limited. Although the present invention has been described in detail with reference to the embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention as defined in the appended claims.